Language Learning & Technology 18

Volume 18 Number 1
February 2014
Articles
Columns
Fostering Foreign Language Learning

Through Technology-Enhanced Intercultural
Projects
Abstract | Article PDF
Jen Jun Chen, National Sun Yat-sen University
Shu Ching Yang, National Sun Yat-sen University
pp. 5775
Emerging Technologies
Edited by Robert Godwin-Jones
Using Peer Computer-Mediated Corrective

Feedback to Support EFL Learners' Writing
Ali AbuSeileek, Al al-Bayt University
Awatif Abualsha'r, Al al-Bayt University
pp. 7695
Direct and Indirect Access to Corpora: An
Exploratory Case Study Comparing Students
Error Correction and Learning Strategy Use in
L2 Writing
Hyunsook Yoon, Hankuk University of Foreign
Studies
JungWon Jo, Hankuk University of Foreign
Studies
pp. 96117
Effects of Captioning on Video
Comprehension and Incidental Vocabulary
Learning
Maribel Montero Perez, iMinds - ITEC KU Leuven
Kulak
Elke Peters, KU Leuven Kulak
Geraldine Clarebout, KU Leuven Kulak
Piet Desmet, iMinds - ITEC KU Leuven Kulak
pp. 118141
Towards Transparent Computing:

Content Authoring Using Open Standards
Article PDF
by Robert Godwin-Jones
pp. 110
Action Research
Edited by Greg Kessler
Can a Web-Based Course Improve
Communicative Competence of Foreign-born
Nurses?
Article PDF
by Eileen Van Schaik, Emily Lynch, Susan Stoner,
& Lorna Sikorski
pp. 1122
Exploring How Collaborative Dialogues
Facilitate Synchronous Collaborative Writing
Article PDF
by Hui-Chin Yeh
pp. 2337
Announcements
News From Sponsoring Organizations
Article PDF
pp. 3841
Reviews
Edited by Paige Ware
Researching Online Foreign Language
Interaction and Exchange: Theories, Methods,
and Challenges
Dooly, M., & O'Dowd, R.
Article PDF
Reviewed by Linda Bradley
pp. 4245
Contact: Editors or Managing Editor

Copyright 2014 Language Learning & Technology, ISSN 1094-3501.
Articles are copyrighted by their respective authors.
Ecological Affordance and Anxiety in an Oral

Asynchronous Computer-Mediated
Environment
Levi McNeil, Sookmyung Women's University
pp. 142159
Use of Wikis to Promote Collaborative EFL
Writing
Zelilha Aydin, Ozyegin University
Senem Yildiz, Bogazici University
pp. 160180
2013 Reviewer Acknowledgements

Article PDF
pp. 181183
Language Learning with Technology: Ideas for

Integrating Technology in the Classroom
Stanley, G.
Article PDF
Reviewed by Nancy Montgomery
pp. 4648
Encounters: Chinese Language and Culture
Ning, C. Y., & Montanaro, J. S.
Article PDF
Reviewed by Yaqiong Cui
pp. 4952
Technology in Interlanguage Pragmatics
Research and Teaching
Naoko, T., & Sykes, J.
Article PDF
Reviewed by Feng Xiao
pp. 5356
Contact: Editors or Managing Editor

About Language Learning & Technology

Language Learning & Technology is a refereed journal that began publication in July 1997. LLT
disseminates research to foreign and second language educators worldwide on issues related to
technology and language education.
Language Learning & Technology is sponsored and funded by the University of Hawaii National
Foreign Language Resource Center (NFLRC) and the Michigan State University Center for
Language Education and Research (CLEAR).
Language Learning & Technology is a fully refereed journal with an editorial board of scholars in
the fields of second language acquisition and computer-assisted language learning. The focus of
the publication is not technology per se, but rather issues related to language learning and
language teaching, and how they are affected or enhanced by the use of technologies.
Back and current issues of Language Learning & Technology are indexed in the Current
Contents/Social and Behavioral Sciences, ISI Alerting Services, Institute for Scientific
Information's (ISI) Social Sciences Citation Index (SSCI), Linguistics Abstracts, PsycINFO, and
Social SciSearch databases.
Since 2007, Language Learning & Technology has ranked in the top 20 Linguistics journals and
in the top 20 Education journals in the Thomson Reuters Journal Citation Reports. The European
Reference Index for the Humanities (ERIH) classifies Language Learning & Technology as INT2
in the field of Pedagogical and Educational research, an international publication with
significant visibility and influence in the various research domains in different countries.
Language Learning & Technology is published three times per year (February, June, and
October).

Sponsors, Board, and Editorial Staff

Volume 18, Number 1
SPONSORS
University of Hawaii National Foreign Language Resource Center (NFLRC)
Michigan State University Center for Language Education and Research (CLEAR)
ADVISORY AND EDITORIAL BOARDS

Advisory Board
Susan Gass
Michigan State University
Julio Rodrguez
University of Hawaii at Mnoa
Editorial Board
Nike Arnold
Robert Blake
Klaus Brandl
Thierry Chanier
Scott Crossley
Tracey Derwing
Lara Ducate
Robert Godwin-Jones
Regine Hampel
Debra Hardison
Claire Kennedy
Markus Ktter
Eva Lam
Jenifer Larson-Hall
Joshua Lawrence
Lina Lee
Meei-Ling Liaw
Paul Kei Matsuda
Jill Pellettieri
Mark Pegrum
Hayo Reinders
Bryan Smith
Patrick Snellings
Susana Sotillo
Phillip A. Towndrow
Pavel Trofimovich
Portland State University

University of California, Davis
University of Washington
Universite Blaise Pascal
Georgia State University
University of Alberta
University of South Carolina
Virginia Commonwealth University
The Open University
Griffith University, Brisbane
University of Siegen
Northwestern University
Kyushu University
University of California, Irvine
University of New Hampshire
National Taichung University
Arizona State University
Santa Clara University
University of Western Australia
Middlesex University, London
Arizona State University
University of Amsterdam
Montclair State University
National Institute of Education, Singapore
Concordia University

The contents of this publication were developed under a grant from the Department of Education (CFDA 84.229,
P229A60012-96 and P229A6007). However, the contents do not necessarily represent the policy of the Department
of Education, and one should not assume endorsement by the Federal Government.
Editorial Board (continued)

Paige Ware
Cynthia White
Bonnie Youngs
Southern Methodist University

Massey University
Carnegie Mellon University
Editorial Staff
Dorothy Chun
University of California, Santa Barbara
Mark Warschauer
University of California, Irvine
Trude Heift
Simon Fraser University
Philip Hubbard
Stanford University
Rick Kern
University of California, Berkeley
Marie-Nolle Lamy
The Open University
Lara Lomicka-Anderson
University of South Carolina
Carla Meskill
State University of New York, Albany
Glenn Stockwell
Waseda University
Managing Editor
Mnica Vidal
Web Production Editor
Carol Wilson-Duffy
Book & Multimedia Review Editor
Paige Ware
Southern Methodist University
Action Research Column Editor
Greg Kessler
Ohio University
Emerging Technologies Editor
Robert Godwin-Jones
Virginia Commonwealth University
Social Media Director
Chin-Hsi Lin
Copy Editors
Mauricio Aldana
Universidad de los Andes
Nicole Bruskewitz
Kyle Crocco
University of California, Santa Barbara
Justin Cubilo
Reginald Gentry, Jr.
Kobe Womens University
Christopher Hunter
Anne Jund
Jake Kletzien
Sookmyung Womens University
Valerie Meier
Ricardo Nausa
Editors
Associate Editors

The contents of this publication were developed under a grant from the Department of Education (CFDA 84.229,
P229A60012-96 and P229A6007). However, the contents do not necessarily represent the policy of the Department
of Education, and one should not assume endorsement by the Federal Government.
Language Learning & Technology

http://llt.msu.edu/issues/february2014/emerging.pdf
February 2014, Volume 18, Number 1

pp. 110
EMERGING TECHNOLOGIES
TOWARDS TRANSPARENT COMPUTING:
CONTENT AUTHORING USING OPEN STANDARDS
Robert Godwin-Jones, Virginia Commonwealth University
With today's rapid developments in digital technologies, technical obsolescence can occur
much faster than in the past. Who would have predicted five years ago that Adobe's Flash
would have seen the rapid decline it has experienced as a development environment?
Using open, internationally accepted standards for materials development is no absolute
guarantee of longevity, but it does increase the likelihood that content will continue to be
usable and that, if needed, conversion tools will be available. In this column I will be
discussing approaches to the development of electronically delivered language learning
materials that I believe are the least likely to face short-term obsolescence. I will be
arguing in favor of multilaterally developed, open standards, supported by major industry
players and educational standards bodies. Specifically, I will be looking at approaches for
delivering learning materials through Web browsers or e-book readers (e-readers), so that
created content works seamlessly across devices and platforms.
APA Citation: Godwin-Jones, R. (2014). Towards transparent computing: Content
authoring using open standards. Language Learning & Technology 18(1), 110. Retrieved
from http://llt.msu.edu/issues/february2014/emerging.pdf
Copyright: Robert Godwin-Jones
WEB DELIVERY
For the vast majority of language learning projects, the delivery mechanism of choice is likely to be the
World Wide Web, accessed through a Web browser, whether that be on a desktop computer or on a
mobile device. There are use cases that may dictate a different delivery strategy, such as a compiled
application (for an intelligent language tutorial, for example), or a mobile app (if a particular device is
being targeted). It's more and more likely, however, that no matter what the programming environment is
on the backend, the client-side delivery will be (and should be) Web-based. As I've argued in a previous
column (PDF), that holds true for mobile delivery as well (Godwin-Jones, 2011). In fact, developing for
the Web with current technology standards makes that content most widely useable. This doesn't mean
that the created Web pages will have the identical look and feel in all environments, but it does mean that
the content and functionality of the pages will essentially be the same. This is made possible through the
most recent version of the Web programming language, HTML5 (by convention capitalized with no space
before version number), which, since December, 2012 is a "Candidate Recommendation" of W3C, the
standards body for the World Wide Web. HTML5 has features that enable capabilities not available
before (or only available through proprietary plug-ins), many of which are of substantial interest to
language professionals.
At the Intel Developer Forum in 2012, a language learning application was demonstrated (video clip) that
illustrates both the versatility of HTML5 and its interactive features. The demonstration ran on an iPad
(using Apple iOS) and on a Windows 8 laptop. This interoperability would be no challenge for typical
Web pages. This example was, however, more akin to an application than normal Web pages, as it made
rich use of multimedia, animations, and interactivity. This would typically be considered a "Web app".
The media playback and interactivity were accomplished without the use of plug-ins such as Microsoft
Silverlight or Flash, neither of which are supported on the iPad. The app featured synchronized audio with
animated character renditions, character writing with instant feedback (including verification of stroke
Copyright 2014, ISSN 1094-3501
Robert Godwin-Jones
Towards Transparent Computing
order), adaptive assessments, self-correcting exercises, and integrated cultural videos. The activities
included a drag and drop vocabulary drill, an animated interactive timeline of Chinese history, and active
manipulation of 3-D graphics (world map). New characters were introduced with a rich set of contextual
examples, easily shown or hidden, all displayed without page refreshes, as the needed data were preloaded from a server. The video clips played instantly (also due to preloading) and featured custom
playback controls.
The app was demonstrated as an example of "transparent computing", the idea that the consumer should
not have to be concerned with which connected device is being used and can easily switch
devices/platforms and have a similar user experience, with automatic data syncing. This kind of
interoperability is only possible through the adoption and implementation of common standards. In
contrast, many development environments today, particularly in the mobile space, represent a world of
walled gardens with no connected pathways. A mobile app developed natively for Apple devices (using
Objective-C) will not run in Android or Windows devices, or even on Mac computers. This pattern holds
true for development on other platforms as well. This means that developers must target a particular
mobile OS or, alternatively, create separate apps for each.
Any data generated by the user often remains within the walled garden. Through sending that data to a
server ("cloud services"), syncing is possible, but not necessarily across different platforms. In any case,
the cloud data is reserved for use by that app (or its siblings on other platforms) and is not accessible in
other ways. This may be understandable in terms of privacy and security, but it serves to inhibit the
development of open cloud-based services for language learning, which could be built on top of such data.
Having available user data including achievement milestones, proficiency levels, and interest areas could
offer the possibility of creating individualized learner profiles, forming the basis of personalized learning
in intelligent language learning environments. This is a vision of the future only achievable if the stored
data is encoded and structured in a non-proprietary format.
HTML5 FEATURES
The Chinese language-learning app from Intel uses a variety of HTML5 features. Animations, for
example, are created in Canvas, a graphics-rendering environment which uses scripting to draw and
manipulate images. Canvas elements can be quite simple and static or extraordinarily rich, complex and
animated. HTML5 also supports SVG (Scalable Vector Graphics), which draws graphics in the XML
language (Extensible Markup Language). These graphic environments have advantages over other
formats in that they do not lose quality when zoomed or re-sized, and each drawn element can be
animated and scripted. Both are widely used in Web-based game programming, an area of increasing
interest in language learning. One of the caveats that should be mentioned here, and that applies to the
graphics capabilities, as well to some other HTML5 features, is that they are not universally usable in all
desktop and mobile browsers. Canvas, for example, works across current versions of all the major
browsers, but not in older browsers (whose use might be necessitated by older OS versions). It's important
to be aware of the implementation status of any new HTML5 features which are crucial to the
functionality of a Web app. Sites such as html5test.com (evaluates browser in use) or caniuse.com
(support table) can provide help with that process. As always in content development, it's a good idea to
implement HTML5 using progressive enhancement, i.e., assuring basic content delivery and functionality
for all browsers, while enabling a more advanced, interactive view for users of supported browsers. This
is easiest implemented with a JavaScript library such as modernizr.com.
Embedded media are intended to be played in HTML5 without the need for plug-ins, through native
support for the new "audio" and "video" tags. Given the importance of multimedia in language learning,
this development is welcome, as it means better delivery options, particularly for video. Traditionally,
video could be delivered through 1) directly clicking on a link to the video file and waiting for it to
download, 2) using a commercial streaming service, or 3) using a proprietary plug-in which may not be
Robert Godwin-Jones
available on all platforms. The audio and video elements feature auto-buffering, allow custom controls to
be used, and, since they are normal elements of the page's DOM (Document Object Model), can be
manipulated by other elements of the page. The playback of a particular segment of a video, for example,
could result from user actions such as answering a question. For particular effects, video and Canvas
elements can be used together, as in a cartoon video featuring a Canvas-drawn background. There is a
W3C working group (Device APIs) that is working on a standard for capturing a user's video camera, if
available, and if permission is granted (info and example, supported currently in the Chrome browser).
One could also filter the video from a webcam using Canvas.
The main hiccup currently in using HTML5 video is the lack of support among browsers for a common
video codec (video compression format), with some browsers supporting h.264 (MP4) and others VP8
(WebM), making it currently advisable to serve up both encodings. A promising aspect of HTML5 video
for language learning is its use of WebVTT (Video Text Tracks), to display subtitles, captions, chapter
headings, or screen reader descriptions. The ability to switch a caption or subtitle on or off, or to switch
languages on the fly, and to do so programmatically, is a great boon to language learning applications. For
universal support of this feature currently, using the free video.js (open source HTML5 video player) or a
similar program is needed.
The user actions which could result in video playing, jumping to a scene, or turning on subtitles, could be
generated in a variety of ways. It might involve filling in a form element, such as a short answer or
multiple choice question, or possibly moving items on the page, such as arranging pictures illustrating
events in a story. HTML5 offers new native support for drag-and-drop. Designating "draggable" as an
attribute to a page element makes it "live". Intel's Chinese language learning app uses drag and drop to
have the user select an appropriate image to answer a question. This function may, indeed, seem to target
image manipulation, but drag and drop could involve text as well, as in this example. Ordering items,
sentence construction from words or phrases, matching, are some of the possible uses in language
learning. Drag and drop can also be used for uploading files, allowing, for example, images or text files to
be dragged from the desktop to be uploaded.
It used to be the case that text entry and text manipulation were only possible in HTML form fields.
HTML5 supports the "contenteditable" attribute, normally used in text fields, which now can be used for
any page element, allowing for rich text editing anywhere on the page. An example of a page element
made "editable" also illustrates an additional welcome feature of HTML5, support for storing name/value
pairs in the browser memory. This used to necessitate using HTTP cookies, which are hampered by size
and usage restrictions. Web storage allows for temporary, session-based storage of values
("sessionStorage") or persistent storage, after the browser is closed ("localStorage"). This more robust
form of browser storage of variable values is crucial for Web apps to be able to bring users back to finish
items only partially completed, with completed items remembered. This functionality is also needed for
Web browsers to recall scores or other data from one page to another. Web storage is supported widely in
both Web browsers and e-readers. Some developers prefer storage options that are offer more of a
database-style format such as WebSQL or IndexedDB, but they are not universally deployable (or
supported as W3C standards).
Other new HTML5 features in terms of interactivity include a multitude of new attributes for Web forms
and form elements. Pattern matching in text fields (also called use of "regular expressions"), for example,
allows for expanded options for evaluating user input. This enables testing for the presence of a core item
with the added ability to analyze verb endings or inflections, and provide individualized feedback.
Another attribute of possible use in grammar exercises or in writing help is the use of "datalist" controls,
which suggest input to be used (inflections, for example), based on a supplied list, providing a kind of
auto-complete functionality (example). Also new, but not yet widely supported, is a speech input field
(demonstration) which uses speech recognition to convert speech to text. Among other text manipulation
features, one that may be of interest for Asian languages is support for ruby annotations. These are
Robert Godwin-Jones
glosses used with logographic languages to indicate tones, or to provide other help in pronunciation.
They're widely used in Japanese (furigana) in which kanji are glossed with hiragana. Sites such as
html5demos.com or others listed in the reference list provide numerous demonstrations of HTML5, which
may provide ideas for applications in language learning.
E-TEXTS
HTML5 is the basis for what has become the standard format for electronically delivered books, EPUB 3
(by convention capitalized with space before version number). This is the standard created and promoted
by a nonprofit trade and standards organization, the IDPF (International Digital Publishing Forum), with
members from 35 countries. EPUB 3 (EPUB stands for electronic publication) was approved as a "Final
Recommended Specification" in October, 2011, and is now in wide use by publishers and e-readers. The
"big six" U.S. publishers format their book content now in house as EPUB 3 (rather than in XML). The
EPUB 3 format can be thought of as a Website in a box, as the content is encoded in HTML5 or XHTML.
As is the case with HTML5, a typical EPUB 3 will also include media files (for images, audio/video),
CSS files for formatting (Cascading Style Sheets; EPUB 3 supports CSS3), and .js files (JavaScript). That
content is accompanied by a manifest file, providing a complete list of files as well as navigational
information (i.e., a chapter list), with a mime type declaration ("application/epub+zip"), and a file
("container.xml") pointing to the name and location of the manifest file. Metadata is included in the
manifest file and can also be contained in separate, linked files. EPUB 3 incorporates standard Dublin
Core metadata but also supports use of fine-grained metadata, essential for cataloging and searching. The
folder containing all files for the EPUB 3 is zipped and given an .epub file extension. The IDPF provides
a complete description of what comprises a valid EPUB 3 package.
If an EPUB 3 package is basically Web-formatted content, why not just display that content as HTML in
a Web browser? In fact, some have argued for that. There are however a number of compelling use cases
for EPUB 3. One of the key advantages to EPUB is that once it is loaded onto a device, that content is
available whether the device is connected to the Internet or not. HTML5 content can also be cached so as
to be available off-line, but that process is cumbersome. Optimized text formatting is another key feature.
EPUBs are designed for reflowable content, so the text display is optimized for the device on which it is
being used, as contrasted, for example, with the fixed layout of PDF. Text in a Web browser too can be
displayed on different sized screens, and techniques of responsive Web design can automatically adjust
for different devices. However, these techniques involve adding extra code to the HTML and are
infrequently used. With EPUB 3 the text flows to adapt to the size of the display with automatic
pagination and navigation specific to the device or app used. Page breaks and pagination on an iPad and
on an Android phone, for example, will be adjusted for those devices. E-readers make it easy to adjust
font characteristics (size, color, and face), to change background color, or to adjust page turning
mechanisms. Additionally, EPUB 3 supports bookmarking, note-taking, and automatic dictionary lookups.
For content that is primarily text -- especially if it is of considerable length -- the EPUB format, with its
built-in book-friendly features, is preferable to straight HTML5.
EPUB 3 was designed with accessibility in mind. HTML5 supports some semantic markup such as tags
for article, aside, dialog, or summary. EPUB 3 adds more semantic tags, through the "epub:type" attribute,
which was designed specifically for accessibility. Additional tags can specify such text elements as
preface, chapter, or sidebar. This allows for tagging items so as to make the text structure clearly
understandable to e-readers. It also enables logical reading order to be defined for those using assistive
technologies. EPUB 3 provides full support for the ARIA specification (Accessible Rich Internet
Applications) of the Web Accessibility Initiative. Also supported is text to speech. The semantic tags in
EPUB 3 are not close to being as complete and comprehensive as they are in the TEI specification (Text
Encoding Initiative), but they can be helpful in defining the basic structure of documents, a process which
benefits all readers. Tagging footnotes, for example, allows for e-readers to display that information in a
Robert Godwin-Jones
user-friendly way.
The extensive semantic mark-up is a feature that greatly facilitates EPUB versions of textbooks. This is
one aspect of EPUB 3 that has led to the considerable interest we are seeing today among publishers. The
International Publishers Association in March, 2013 endorsed EPUB 3 as its preferred global publishing
standard. The Association of American Publishers has begun an EPUB 3 implementation project, which
calls for the release of a large number of EPUB 3 titles in 2014. Publishers who want to make their
content available electronically generally insist on some form of embedded digital rights management
(DRM), which is much easier to implement in EPUB 3 than in HTML5. Most commercial e-texts are only
available in e-readers supporting DRM, whether they be dedicated reading devices (Kindle, Barnes &
Noble Nook) or apps supporting DRM (iBooks, Kindle app, Adobe Digital Editions).
EPUB 3 AND LANGUAGE LEARNING
EPUB 3 incorporates features that will be of particular interest for language learning. It offers robust
support for non-Latin writing systems. Supported, for example, are vertical writing and right to left page
progression (sample for Japanese). EPUB 3 allows use of non-ASCII characters in names such as HTML
anchors or filenames (as used in sample referenced above). Emphasis dots and lines can also be used,
where the equivalent in western script would be italics. Sample e-books in Arabic, Hebrew and
Devangar illustrate text layout and character display in those languages. O'Reilly's EPUB 3 best
practices (highly recommended) provides additional examples of global language support. It's also
possible to embed a font in an EPUB 3, which allows for character display in virtually any writing system.
Another benefit for language learning in the EPUB feature set is synchronized audio narration, in which
text is highlighted as read, as in this Japanese manga example (direct link to EPUB 3 file; needs to be
viewed in one of the e-readers in reference list). This is a feature of obvious interest in children's books
(Dickens' Christmas Carol audio e-book; EPUB 3 file), but could be quite useful in language learning as
well, not to mention literary study (Dante's Divina Commedia with audio). This is implemented through
the use of media overlays. Overlays allow the user to switch easily and instantly from one reading
modality to another. One might start reading at home, then switch to audio mode in the car. To create a
media overlay, the typical process is to record the audio with a tool such as Audacity, mark and export the
synchronization points, then merge this data with the text. Some sample code from the Moby Dick e-book
illustrates SMIL use (Synchronized Multimedia Integration Language) to create this kind of "read along"
functionality.
EPUB packages consist of a single zipped file. This can be constructed by hand by writing the needed
manifest files, making sure content is valid HTML5 or XHTML, using the required naming conventions,
zipping the folder, and branding it as EPUB. Validators such as the IDPF EPUB tool can pinpoint errors
in the code. There are tools available for creating EPUB 3 e-books or converting from other formats, for
those who prefer not to code manually (see reference list). A widely used cross-platform conversion
utility is Calibre, which also functions as a desktop e-book reader. It can read and write in a variety of
formats, including the proprietary KF8, used by Amazon's Kindle readers. As EPUB 3 becomes more
widely supported, it's likely we will see more and better tools for formatting content as EPUB 3.
Particularly welcome would be tools which make it easier to take advantage of advanced features such as
audio synchronization.
While it's obvious how using EPUB 3 for delivery of long-form texts supplies a number of benefits, it
may appear that aside from interactive multimedia capabilities, the format is largely static and quite
different from the dynamic environment of HTML5, with its support for the use of highly interactive Web
apps. While it's certainly the case that complex, graphics-intensive Web apps would not be the best
candidates for the EPUB format, scripted interactions are in fact supported in the EPUB 3 specs. Scripting
support is not required, although most current e-readers include JavaScript support, although,
Robert Godwin-Jones
unfortunately, they do not have any script debugging. Scripted interactivity is normally created using
JavaScript and embedded in the HTML or included in external .js files -- in other words in the same way
as it would be for Web browser delivery. Typically, textbooks in EPUB 3 format include assessments at
the end of each chapter. A sample EPUB 3 (EPUB 3 file) from Infogridpacific illustrates a variety of
scripted activities similar to what one finds in electronic textbooks. Although e-texts formatted in EPUB 3
allow scripted interactions, connecting to external Web resources, although permitted, can be problematic.
In such situations, one is better off using straight HTML in a Web browser. This is also the case for
integration of social media, which is also limited within EPUB 3.
OUTLOOK
For a project constructed in HTML5, it is not a major extra effort to create an EPUB 3 version as well.
Doing so adds some potential book-related features, but the greater benefit is the additional delivery
options it provides. This could be of particular interest in schools with tablet initiatives, that are looking to
distribute content that will be usable at home without Internet access. Creating an EPUB 3 for use in
iBooks on an iPad or iPhone (or now on Mac OS computers as well) allows creators to bypass the lengthy
iBookstore approval process. In contrast, if one were to use Apple's e-book creation tool, iBooks Author,
the resulting product would be in a proprietary format, displayable only on Apple devices. Standard
EPUB 3s, on the other hand, can be shared easily with students in a variety of ways, including a simple
download link on a Web page.
There are some features missing from the EPUB spec that would be helpful in an educational setting. One
such feature is a standard way to create annotations and glosses, obviously an important feature for
language learners. Although it is possible to integrate into EPUB 3 look-ups from dictionaries or other
reference sources, there is as yet no widely supported method for annotations, despite the presence in the
specification of tags for glosses. Some e-readers have developed a means for achieving this function
while staying within the EPUB 3 specs. Apple iBooks, for example, relies on the footnote tag to create
pop-up overlays to display footnotes or annotations, rather than taking the reader to the end of the
document. For a true seamless e-book experience across devices, annotations should -- as is not the case
now -- transfer across reading apps and platforms, as should bookmarks and notes. Ideally, sharing such
data with others should be possible as well. The EDUPUB initiative is looking to deliver a way to do that.
Another important area for educational use is a robust and compatible scripting environment. Despite the
fact that most of the popular e-readers are based on WebKit, Apple's open Web browser engine, they
differ markedly in JavaScript execution. The IDPF has recognized this issue and is making its own
JavaScript engine (ReadiumJS) which will be made freely available to e-reader producers. IDPF is also
working with IMS Global, the education standards body, to support both the QTI specification (Question
and Testing Interoperability), widely used by publishers for creating test banks, and LTI (Learning Tools
Interoperability), which would allow for better integration of EPUB 3 into an LMS (Learning
Management System). A sample LTI and QTI integration demonstrates how that might be done. That
particular prototype is of interest as well in that it demonstrates how to create fallbacks should network
connections be unavailable, or should the connection be lost. An additional consideration in using EPUB
3 for content delivery is that revising content requires creating a new EPUB 3 file for downloading. In
contrast, with HTML, a text editor can be used to do revisions to a single page, without the necessity of
redistributing the entire unit of content in a package.
Following the development strategy and roadmap I have outlined here is not likely to appeal universally
to developers, particularly those targeting mobile delivery. Mark Zuckerberg famously announced in 2012
that his biggest mistake in developing mobile apps for Facebook was the decision to use HTML5, citing
slow performance and design constraints. These are common complaints from developers, particularly
those working on large-scale projects such as the mobile Facebook app. As one developer discusses,
development in such cases using a native development environment might be preferable. That necessitates,
Robert Godwin-Jones
however, creating compiled applications in different programming languages (for iOS, Android,
Windows 8), not a task to be undertaken lightly. A possibility for creating native apps is to port the
HTML5 code using a tool such as PhoneGap or Intel's XDK. It's also possible to create hybrid apps,
written in HTML5 with some native code integrated. Another concern for developers is the perceived
inability of Web apps to integrate deeply into mobile device hardware. It's not the case that creating an
HTML5-based Web app necessarily denies access to particular device features, but that access may be
different from device to device. In recognition of this issue, the W3C System Applications Working
Group is working to develop standards and methodologies for accessing such typical device features as
GPS, accelerometers, Bluetooth and NFC (Near Field Communication).
One of the other issues in developing EPUB 3 is the degree of support in e-readers for the EPUB 3
specification. Not all features are supported in all e-readers, so that developers need to proceed cautiously
in implementing advanced features such as scripted interactions or the recently developed fixed-layout
option. It's important to provide adequate fallbacks for features not supported. O'Reilly Media, one of the
first publishers to move to EPUB 3 for its publications, offers concrete examples of how that can be
achieved. The Book Industry Support Group offers a helpful EPUB 3 support grid for developers.
Currently the desktop e-readers which support the greatest number of features are iBooks (Mac), Azardi
(Linux, Mac, Windows), and Readium, an extension of the Chrome browser (Linux, Mac, Windows). For
mobile devices, iBooks (iOS) and Helicon Books EPUB3 reader (Android) are recommended.
At VCU this year I am serving on a university e-text task force, and I suspect many institutions have such
groups operating, looking at textbook affordability, the open access e-text movement, and the integration
of electronic texts into the rest of the university's digital infrastructure, especially its LMS. In fact, a
number of U.S. universities have engaged in pilot e-text initiatives, including a major collaborative
project sponsored by Educause and Internet2. That project is using Courseload for delivery of e-texts,
which formats commercial textbooks with proprietary DRM which is then distributed to authenticated
users from its servers. Similar services are available from VitalSource and Inkling. The project has gotten
mixed reviews, with a major problem being that Courseload did not until recently support use of portable
devices. The e-text distribution services currently in use in the U.S. tend to start with HTML5 or EPUB 3
content, then add proprietary elements. Inkling, for example, uses its own markup system, S9ML, for
interactive elements.
Many faculty members are seeking more open options for implementing e-texts and are looking to freely
available e-texts from sources such as Wikibooks or the Internet Archive. This bypasses the need for any
server authentication handshake, and makes it more likely that a wide number of devices and
environments can be supported. A number of open access language e-texts are available (see reference
list), which range from full textbooks to lessons or simple activities. As is always the case with open
education resources (OER), it's not always easy to find appropriate materials and, if found, they may not
meet specific needs. Usually OER come with a Creative Commons share-alike license, enabling content
to be tailored to need and to be combined with other materials, not something possible with traditional
commercial textbooks, even if they are available in a digital format. Another concern in OER materials is
quality. Examining the language e-texts available from OER Commons, for example, shows a wide range
of quality and scope. Using repositories which encourage peer review of content submissions, such as
Merlot, can be helpful in finding reliable content. For course-related open materials, the Open
Courseware Consortium is a good resource. There is likely to be considerably more activity in the near
future in this area, given U.S. federal funding for OER development for community colleges as well as
state initiatives in California, Florida, and Utah and in the Canadian province of British Columbia. South
Korea is transitioning to use electronic books exclusively in its schools by 2015.
The long-sought developer nirvana is "write once, run anywhere," enabling true interoperability and
transparent computing. It's too early to say whether HTML5 will fit this bill. Not long ago, it seemed that
Java would hold the keys to the kingdom, but that has not panned out, due to problems like slow
Robert Godwin-Jones
performance, memory issues, and significant platform and plug-in differences. Java applets running in a
browser also, like Flash, require a plug-in. In contrast, HTML5 is the native environment of the Web
browser and, unlike Java, does not need to be compiled into an executable program. While there are
plenty of arguments in favor of open standards, there are also powerful forces working in the opposite
direction. Many companies have a strategic and financial interest in exclusivity and in preventing
interoperability. Both Apple and Amazon, for example, use proprietary versions of EPUB 3 for their ebooks. For educational institutions, interoperability through open standards is crucial, so that the ability to
change institutional software platforms (such as the LMS) is maintained. HTML5 development is also
less costly with a shallower learning curve, important considerations in educational settings. Additionally,
HTML5 is the most likely development platform to be compatible with future OS's or devices, such as the
upcoming Firefox OS or Tizen. No one can see into the future, but it does seem a reasonable bet that the
mobile environment will continue to be fragmented, hence the added importance of interoperable
standards in developing learning materials to be accessed from tablets and phones.
REFERENCES
Garrish, M. and Gylling, M. (2013). EPUB 3 best practices: Optimize your digital books. Sebastopol,
CA: O'Reilly Media.
Godwin-Jones, R. (2011). Mobile apps for language learning Language Learning & Technology 15(2), 2
11. Retrieved from http://llt.msu.edu/issues/june2011/emerging.pdf
RESOURCE LIST
HTML5 Info
HTML 5 Official specs from W3C

HTML5 Introduction - What is HTML5 Capable of From WebDesigner
HTML 5 Doctor Helpful site on HTML 5 compatibility issues
HTML5 Demos
HTML 5 Demos and Examples Includes browser support info

HTML5 Website Showcase: 48 Potential Flash-Killing Demos HTML5 Canvas demos By
Kevin Roast
9 Mind-Blowing Canvas Demos
Gaming your way to language learning Example of an HTML5 language learning game (from
Shai Shapira)
HTML5 Hackathon - building a Language Learning App YouTube video
BlahBlahLearning takes the winning prize at Onswipe's HTML5 Hackathon
Mobile Development: HTML5 versus Native Apps
HTML5 rocketing in popularity, finds Developer Economics report From Telefnica Digital
Debunking five big HTML5 myths From Telefnica Digital
Intel commits to "fabulous" HTML5 From PCPro
Developing a Cross-Platform HTML5 Offline App From Grinning Gecko
To HTML5 or not to HTML5, that is the mobile question From Webdesignerdepot
HTML5 Vs. Native Mobile Apps: Myths and Misconceptions From Forbes
Here's why HTML-based apps don't work Argument in the developer debate
Native vs. HTML5 -- looked at objectively, the debate is over Another argument in favor of
native apps
Robert Godwin-Jones
Best Of Both Worlds: Mixing HTML5 And Native Code
EPUB 3 Info
EPUB 3 Overview From International Digital Publishing Forum

EPUB 3 links Nice list from epubtest.com
PIGS, GOURDS, AND WIKIS Liz Castro's blog, excellent source for information
EPUB Resources and Guides From O'Reilly Media
Digital Publishing and the Web From A List Apart
E-reader Applications with Support for EPUB 3
Comparison of e-book readers Wikipedia

EPUB 3 Support Grid Good reference for updated information on e-readers
iBooks Mac only
Readium Plug-in Google Chrome browser (Linux, OS X, Windows)
Azardi Linux, Mac, Windows
Calibre Linux, Mac, Windows, clunky interface & limited EPUB 3, but able to convert from
and to many formats
Mobile E-reader Apps with Support for EPUB 3
Apple iBooks iOS

Helicon Books EPUB3 reader Android
Sample E-books
EPUB 3 Sample Documents

Sample ePub 3 books From Azardi
EPUB 3 Samples From Infogrid Pacific
EPUB3 with embedded QTI assessments
eBooks: A Language Learner's Best From Langology
Language Learning eBooks
Two examples of EPUB3 Manga
Dante's Comedy as EPUB 3 Sample w. Media Overlays From Smuuks
300 Enhanced eBooks Published for Language Learning From the Samsung Learning Hub
Language e-books List for over 400 languages
EPUB 3 Development
Epub Validator From idpf

EPUB 3 Best Practices Book from O'Reilly Media
Create Your Own eBooks Review of different tools by Richard Byrne
Build a digital book with EPUB How to create an epub 3 by hand
Adobe InDesign Windows/Mac
Apple iWork Pages Mac only
oXygen XML Editor Windows/Mac/Linux
EPUB 3 Language Info
Requirements for Japanese Text Layout W3C Working Group

Richer Internationalization for eBooks Workshop from W3C
Suprasegmental Phonological representation of Tone in Kikuyu Example of ruby markup
Robert Godwin-Jones
What are situations with western languages where you'd use HTML 5's Ruby element? From
stackoverflow
Acessibility Info
Sample accessible (test) books in EPUB 3 format LIA Project

Accessibility and eBooks: International Endorsement
Tips for Creating Accessible EPUB 3 Files DIAGRAM Center website
EPUB 3: Accessibility Guidelines IDPF
Accessible EPUB 3 free publication on O'Reilly site
EPUB 3 Accessibility forum IDPF website
10

http://llt.msu.edu/issues/february2014/action1.pdf

pp. 1122
CAN A WEB-BASED COURSE IMPROVE

COMMUNICATIVE COMPETENCE OF FOREIGN-BORN NURSES?
Eileen Van Schaik, Talaria, Inc.
Emily M. Lynch, Talaria, Inc.
Susan A. Stoner, Talaria, Inc.
Lorna D. Sikorski, LDS & Associates
Key words: Web-Based Instruction, Second Language Acquisition, Speaking,
Pronunciation, Online Teaching & Learning, Culture, Language For Special
Purposes
APA Citation: Van Schaik, E., Lynch, E. M., Stoner, S. A., & Sikorski, L. D, (2014). Can
a web-based course improve communicative competence of foreign-born nurses?
Language Learning & Technology, 18(1), 1122. Retrieved from
Received: January 25, 2013; Accepted: May 18, 2013; Published: February 1, 2014
Copyright: Eileen Van Schaik, Emily M. Lynch, Susan A. Stoner, & Lorna D. Sikorski
INTRODUCTION
In the years since World War II, the United States has grown increasingly dependent on foreign-born
healthcare personnel at all levels of the healthcare system. Currently, 1.46 million immigrants account for
15% of all healthcare providers in the United States. (Clearfield & Batalova, 2007). Current estimates
indicate that 17% of the total U.S. nursing workforce are foreign-born, a dramatic increase from 1998
when they were approximately 4.5% of the total (Brush, Sochalski, & Berger, 2004; Spry, 2009).
Recruiting foreign-born nurses has been recommended as a solution to the U.S. shortage of nurses (Xu,
2003; Gamble, 2002; Neal, 2002) and as a means of diversifying the nursing work force to meet the needs
of an increasingly multicultural patient population (Abriam-Yago, Yoder, & Kataoka-Yahiro, 1999). The
American Nurses Association (ANA), however, questions the ethics of recruiting nurses for immigration
to the United States. The ANA objects to the resulting brain drains in less wealthy nations as well as the
failure to remedy the root causes of the nursing shortage in the United States (Trossman, 2002; Buchan &
Aiken, 2008). Nevertheless, shortages in the U.S. health-care labor force are not expected to disappear
anytime soon (McMahon, 2004; Buerhaus, Auerbach, & Staiger, 2009; Ellenbecker, 2010) and reliance
on immigrant professionals is likely to continue.
Foreign-born professionals, whether educated in the United States or abroad, face tremendous challenges
adjusting to differences in language, culture, and healthcare practices in the United States (Doutrich,
2001; McMahon, 2004; Yi & Jezewski, 2000; Xu & Davidhizar, 2004). Foreign-born nurses report that
while they may feel clinically competent, they often feel unprepared for the use of English in the
healthcare setting (Davis & Nichols, 2002; Guttman, 2004). Immigrant health professionals experience
communication difficulties with patients and coworkers that are easily exacerbated in a healthcare setting
where situations can quickly become emotionally charged and stressful. Their U.S.-born peers identify a
number of difficulties in working with foreign-born nurses, including lack of communication skills and
differences in decision making, behavioral norms, role expectations, and attitudes (Yi & Jezewski, 2000).
Patients may express distress at being unable to understand physicians who are not native English
speakers, and, whether their complaints reflect prejudices or not, they are undermining for the immigrant
practitioner (McMahon, 2004).
The Commission on Graduates of Foreign Nursing Schools (CGFNS) conducted focus groups with
11
Eileen Van Schaik, Emily M. Lynch, Susan A. Stoner, & Lorna D. Sikorski
Communicative Competence of Foreign-born Nurses
foreign-born nurses, identifying major challenges around language, culture, and the practice of nursing
(Davis & Nichols, 2002). These challenges point to the need for competencies in four aspects of
language use: 1) intelligibility (phonology and syntactic patterns), 2) vocabulary, 3) conversation, and 4)
context-dependent speech (discourse).
Participants in CGFNS focus groups report feeling unprepared for using English in the healthcare setting.
They had difficulty with colloquial expressions and abbreviated medical terms and found using the phone
especially problematic. Participants identified difficulties with medical and pharmacological terminology,
the U.S. system of weights and measures, and the variety of medications U.S. patients take. Participants
were too embarrassed to ask about procedures they did not understand. Additionally, they felt that
patients and coworkers perceived them as unskilled rather than as professionals. Participants suggested
that instruction would be helpful in basic communication techniques in the healthcare setting and in
conversational English. Nurses from more homogenous countries also reported that they were
unaccustomed to the cultural diversity among U.S. patients. They lacked knowledge of diverse dietary
and religious needs and communication styles. They also found aspects of U.S culture challenging,
including exposure to substance abuse, unfamiliar sexual orientations, and religious customs.
Maintaining a nonjudgmental stance was a challenge for them in many situations (Davis & Nichols,
2002).
Growing numbers of educators and employers are providing orientation courses for newly immigrated
healthcare workers, incorporating content on American English and culture (McMahon, 2004; Guttman,
2004; Xu et al, 2010). It may, however, take years for foreign-born healthcare professionals to feel fully
comfortable with American English, culture, and healthcare (Yi & Jezewski, 2000) and the initial
measures must be continued and reinforced to ensure that cultural and linguistic competence are
consistently implemented in the healthcare setting (Guttman, 2004, p. 265).
The communicative approach has been dominant in the field of second language instruction in the U.S.
and abroad since the 1970s (Nunan, 1991; Hiep, 2007). This approach emphasizes the importance of
meaningful dialogue and frequent interaction for second language acquisition in the belief that the
individual will become fluent in English if provided with sufficient meaningful communication
opportunities. While the communicative approach encourages conversations in naturalistic settings, it
does not provide a framework for the improvement of pronunciation and grammar, nor does the
communicative approach promote the mastery of specific vocabulary sets (Levis, 2005; Sikorski, 2005a).
Many foreign-born adults who have learned English using the communicative approach acknowledge
gaps in their grasp of English and further indicate that these gaps frequently put them at a disadvantage in
the workplace (Sikorski, 2005a). Specifically, these adults experience difficulties with speech
intelligibility, or the amount of information a listener can obtain from a spoken message. Derwing &
Munro argue that mutual intelligibility is the paramount concern for second language learners who must
be understood in a context where native speakers are the majority and when they wish to integrate
socially in the native culture (2005, p. 380).
To address the need of foreign-born nurses for continued language and cultural support, we developed
and evaluated the first phase of a multimedia, Internet-based educational tool, the Intercultural
Communication Workshop (ICW), designed to improve the overall communicative competence of
foreign-born nurses. The ICW is not designed to erase accents. Rather, it expands the users
knowledge of intonation (appropriate stress and pitch) and phonological rules specific to American
English, enhancing their oral communication skills and improving listener understanding (Sikorski,
2005a; Derwing & Munro, 2005). Participants in the ICW learn a three-fold set of rules for improving
their speech intelligibility: consonant clarity, vowel accuracy, and appropriate intonation (Sikorski,
2005a). Instruction in the ICW is tailored to the healthcare workplace whenever possible and includes
contextual information about communication, culture, and healthcare in the United States, thereby
avoiding the one size fits all approach to second language instruction rejected by Derwing & Munro
12
(2005).
The Intercultural Communication Workshop is designed as a supplement to formal in-person instruction
and orientation programs. When such programs are lacking, this e-learning course can stand alone as
self-paced instruction. While some may argue that the ICW cannot take the place of face-to-face
interactions in a well-designed and executed ESL course, the e-learning environment offers several
advantages: it is self-paced, private (thus less threatening), and provides continuous opportunities for
practice and reinforcement.
RESEARCH METHODS
The Course
The first version of the ICW is a web-based course that integrates instruction in speech intelligibility with
contextual information on communication and culture in U.S. healthcare, along with instruction on
common idioms and figures of speech relevant in the healthcare setting. Instruction in the ICW is
centered on multimedia recording exercises that encourage learners to apply "insider information" about
the phonological and prosodic rules that govern American English pronunciation and intonation. This
distinctive approach enables users to progress quickly by making patterned changes to their speech.
Whenever feasible, exercises are built on words and phrases that are likely to occur in conversations
nurses have with patients and their families and with other members of the healthcare team.
The core instruction in speech intelligibility was adapted from an established curriculum (Sikorski, 2004)
in collaboration with the author. Lessons in the ICW are grouped into three chaptersintonation, vowels,
and consonantscorresponding to the most important components of speech intelligibility. The
intonation chapter explains that English requires stress timing (at odds with the syllable-timing rules of
many FBNs native languages). The focus is on typical intonation patterns at the word, phrase, and
sentence level and the unique ways that intonation patterns change meaning. The vowels chapter teaches
the vowel inventory of American English, where and how to make these sounds, and common changes in
vowel sounds based on word stress. The final chapter on consonants teaches the consonant inventory of
American English, how and where to make these sounds, and common changes to consonant sounds that
must occur in conversational English. Additional examples of speech intelligibility topics included in
ICW are provided in Table 1.
Table 1. Examples of Topics Included in ICW Speech Intelligibility Lessons
Intonation
Vowels
Consonants
Pitch
Typical patterns for words and
phrases
Special patterns for words and
phrases
Stress
Word reductions
Inflection
Falling inflection
Statements
WH-questions
Front vowels
Similar sounds (minimal pairs,
e.g. cot v. cat)
Vowel length
Vowel reductions in unstressed
syllables
Vowel additions
Linking
/t/ variations (such as the /t/

sound in butter)
Initial aspiration
/p/ v. /b/ and /k/ v. /g/ (e.g.
minimal pairs pin v. bin)
Unreleased stops
Linking
Consonant voicing
Endings with s and ed
Consonant reductions
Throughout the course, users also view hundreds of tips about American English communication,
language practice strategies, and idiomatic healthcare expressions. Tip boxes are inset on most pages in
13
the course and allow users to read succinct information with examples on the social and cultural context
of healthcare in the United States. Examples of the types of tips included in ICW are provided in Table 2.
Table 2. Examples of Tips for Speaking and Practicing Provided in ICW
Healthcare
Medical talk is a new language, too / Medical terminology is a new language

for everyone. Your patients may not know the meanings of many medical words
that you use every day!
Everyday Talk
(idioms)
A frog in ones throat / Meaning: Hoarseness or phlegm in the throat; unable

to speak clearly until you give a slight cough. / Ahem, excuse me, I seem to
have had a frog in my throat!
Mr. Formal vs.

Informal Guy
Offering Help / Formal & Polite: How may I help you? / Informal: What do you
need?
U.S. English
Are you bored? / If you don't use a significant pitch rise, listeners may think
you are bored or want to get away!
Mirror
You may want to use a mirror as you practice the contrasts in this section.
Providing immediate, tailored feedback to learners is one of the most significant challenges in the web
environment. To meet the need for feedback, we designed the listen-record-compare object (hereafter
LRC), which allows users to listen to a model speaker, record their own speech, and then compare their
recorded utterances to the model. The ICW presents learners with 86 LRC lessons (with 10 to 20 target
utterances each) for a total of 1,932 opportunities to compare their speech with a native speaker. The
usefulness of the LRC is limited for those users who cannot distinguish between their utterances and the
models; however, it is expected to aid intonation and pronunciation of U.S. vowels and consonants for
the majority of learners.
Research Design
We evaluated the feasibility and usability of the ICW and used a single-group, pre- post-test design to
conduct a preliminary evaluation of usefulness of the ICW in improving speech intelligibility. All
procedures were reviewed by the Western Institutional Review Board, in accordance with requirements
for the protection of human subjects in all research funded by the National Institutes of Health.
Participants received an incentive of $125 for participation in the study.
Measures
The following screening and testing measures were completed online by all study participants:
Demographic Questionnaire: includes questions about participants age, gender, education,

native language, and employment, as well as questions about their use of American English.
Test of English Proficiency Level - Online (TEPL-Online): a placement tool challenging the full
range of American English language skillslistening comprehension, oral expression, reading
of correct structures, reading comprehension, and writingthat is scored on a 7-level scale (A
= Survival English G = Advanced English) with results shown for each skill area. Field tests
with adult students have shown it to be an appropriate placement and proficiency instrument.
The TEPL-Online was administered at baseline (Rathmell & Sikorski, 2006).
Knowledge Test: a test consisting of 30 items, 15 true/false and 15 multiple choice questions
based on information presented in the course (e.g., Your best clues for how to pronounce
words comes from American English spelling.; The word comfortable is typically
14
pronounced as a three-syllable word.; Saying the words below aloud, which word contains
the /ae/ vowel: a) Last b) Calm c) Mammogram).
Proficiency in Oral English Communication Screen (POEC-S): a two-part screen of auditory

discrimination and verbal production of key vowel, consonant, and intonation variations unique
to American English (Sikorski, 2005b). POEC-S scores have been strongly correlated to
perceptual ratings made by Speech Language Pathologists (SLPs) and SLP graduate students on
accent, articulation, intonation, naturalness, and intelligibility (Morton, Brundage, & Hancock,
2010).
Satisfaction and Usability Questionnaire: a 22-item questionnaire using a 5-point Likert scale
on which participants rated their agreement.
Active, unencumbered participation in an online training program demands good visual language skills.
For this pilot study, we used the written (nonverbal) portions of the TEPL-Online to determine examinees'
ability to actively participate in an online language learning program. The two written portions of the
TEPL-Online, "reading of correct structures" and "reading comprehension," are multiple choice and can
be scored automatically in the web environment. All components of the current POEC-S were adapted to
the web environment. Cuing and prompts were adapted for the multimedia environment, but no changes
in content were necessary. The Demographic Questionnaire and Knowledge Test were created
specifically for this project. The Satisfaction and Usability Questionnaire was adapted from the System
Usability Scale (SUS) (Brooke, 1996) with items tailored for the ICW.
Participants
The participants were 22 foreign-born nurses and nursing students recruited from nine institutions,
including schools of nursing, nursing associations, and websites for nurses via fliers (hard copies and
online) and e-mail announcements distributed by directors and staff at the sites. All individuals who
expressed interest in the study were eligible for participation. Twenty-two participants enrolled in the
study and received practice tools (described below). Due to time constraints, 3 individuals did not
continue with the course, leaving 19 participants who were invited to complete follow-up measures.
Procedures
The entire study took place online. Participants completed all screening measures and the written
components of the baseline measures. Participants were then sent a mirror and a headset with a
microphone for use in the study, along with a letter instructing them how to return to the website. Upon
receiving their practice tools, participants logged in and completed the auditory components of the
baseline measures. Auditory discrimination sections were automatically scored by the software, and
scores were included in the participant database for later analysis. Participants verbal answers were
recorded and stored on a secure server for scoring by trained POEC-S coders. When the baseline
measures were complete, participants were mailed a payment of $50.
Participants were given 3 months to complete 8 hours of practice time in the online course. They were
asked to practice a minimum of three times a week for 15-20 minutes and no more than 20 minutes at a
time for a total time of 1 hour per week (for a minimum of 8 weeks). During the study period, study
personnel contacted participants by telephone and/or emails, according to the participants preference, at
least every 3 weeks. These check-ins were intended to provide support to participants and to learn
about their experiences in the course. The check-ins also provided an opportunity to answer
participants questions and solve any technical difficulties.
After 8 weeks, participants completed the follow-up measures. When these were completed, the study
research assistant conducted an exit interview by telephone or e-mail, depending on the participants
15
preference, to allow participants to discuss their experience using the course. The research assistant had a
BA in anthropology and was familiar with collecting and analyzing qualitative data. The interview guide
included 13 open-ended questions about the participants experience in the course and covered topics
such as what they found most useful and anything they would like to see added. Participants were then
mailed a payment of $75 for a total of $125 for their participation in the study. Notes taken during phone
interviews and the content of e-mail responses were analyzed for similarities and differences in content.
FINDINGS
Participants
All participants (N = 19) were women, ranging in age from 22 to 67 years (M = 39.2, SD = 12.3).
Additional participant characteristics are summarized in Table 3, including education, native languages,
current employment, the reported percentage of time spent in English-only communication. More than
one-third (39%) of participants reported previous training in language/grammar improvement and onethird (33%) reported cross-cultural training.
Table 3. Characteristics of Study Participants
Native Language
Degrees Held*
Employment Position
English-Only
Communication (% of day)
Tagalog / Ilocano
BSN
Staff Nurse
<20%
Chinese
RN
Student
20%-40%
Thai
MSN
Project Director
40%-80%
Russian
CNA
Research Technician
80%-100%
Danish
LPN
Research Assistant
Korean
Other
[None reported]
Note.*Some participants held multiple degrees.
Table 4. TEPL-ONLINE Scores

TEPL Scores Structure Subscale Reading Comprehension Subscale
G
Missing
As anticipated, participating nurses scored in the mid to upper ranges of the TEPL-Online placement tool
(Table 4). The TEPL-Online is scored on a 7-level scale (A = Survival English G = Advanced English)
with results shown for each skill area. Scores for the TEPL-Online correspond to the seven-level system
used to place ESL students in classes according to their English proficiency. Upon review of the TEPLOnlines Seven Instructional Level Scoring system, the authors chose Level E (Intermediate) as the
Participant Placement minimum. This placement level matches the Readability Index (vocabulary
16
sophistication, sentence structure, etc.) of the course content. Additionally, the authors felt that
intermediate English skills would minimally impact participants speed and comprehension of the
information presented, allowing them to move through the course at an appropriate pace.
Participants reported communication challenges in the following five areas: difficulties with
dictation/reporting (n=6), problems understanding clinical lectures (n=5), problems communicating with
patients and family members (n=4), difficulties with rapport building (n=3), and problems communicating
with colleagues (n=3). Nearly three-quarters (71%) of the participants indicated that they had been in
clinical situations where their English language affected patient care, and 58% felt that they had been
perceived or treated differently by patients, colleagues, family members, or others because of
language/accent or cultural differences.
Main Analyses Quantitative Data
Outcome Results: The Knowledge Test and the POEC-S were administered before and after completion
of the online workshop. Paired t-tests were conducted to evaluate within-subject effects of the course.
As shown in Table 5, participants made significant gains from pretest to posttest in several domains. The
largest effect (d = 1.27) was observed for the Knowledge Test, suggesting that users were learning the
rules which govern American English. Participants also showed significant gains on all subscales of the
POEC-S Verbal Performance Test with a medium-to-high effect size for the overall verbal performance
scale score (d = 0.66). These gains are particularly remarkable given the relatively short duration of the
course. No significant gains were observed for the POEC-S Auditory Discrimination Performance Test.
Table 5. Means, Paired t-tests, and Effect Sizes for the Pre-Post Study Measures
Pretest
Mean
Knowledge Test
16.2
Proficiency in Oral English
112.1
Communication Screen Total
Auditory Discrimination
24.1
Performance Test Total
Single Word Discrimination
5.7
Word Discrimination within Sentences*
6.6
Sentence Completion Discrimination
7.6
Word Stress Discrimination
4.2
Verbal Performance Test Total
98.3
Vowel Survey
49.9
Intonation Survey
13.7
Articulation Variations Survey Total
36.3
SD
Posttest
Mean
SD
df
3.2
36.5
20.4
121.9
3.5
39.9
-5.33
-1.64
14
18
<.001
.118
1.27
0.26
2.7
23.6
2.6
0.80
17
.433
-0.19
0.7
1.0
1.8
0.9
18.7
13.8
2.4
7.2
5.4
6.5
7.6
4.1
111.3
56.9
15.1
41.1
0.9
1.0
1.6
0.7
20.5
8.9
3.0
8.6
0.72
-0.14
0.90
-4.77
-2.53
-2.35
-4.13
17
-17
17
15
15
13
15
.481
-.889
.381
<.001
.023
.035
.001
-0.29
-0.08
-0.01
-0.20
0.66
0.60
0.52
0.60
Note. *All variables were normally distributed, with the exception of Word Discrimination within Sentences, which was skewed
and kurtotic. Pre and post scores on this measure were compared using a Wilcoxon Signed Rank Test, which was not significant.
User Experience
Usage Data: Participants varied in their use of the ICW. Most logged into the course every 2 to 5 days for
approximately 20 minutes to 1 hour while others participated every 1 to 2 weeks for 1- to 3-hour sessions.
17
Some used the course twice in a single day in short, 15- to 30-minute intervals. The majority of users
proceeded linearly through the course, although several returned to previously finished sections for
additional practice. One user repeated the course in its entirety.
Exit Interviews: Participants (N = 10; 3 by email and 7 by phone) reported generally positive experiences
with the course, and several noted that they had already recommended the course to their friends and/or
colleagues. The majority of respondents found the ability to play back their own recorded voice for
comparison with a model speaker a helpful tool. While one participant had difficulty using the LRC,
seven found it instructive and said that it provided a unique opportunity to hear specific features of their
own speech that needed improvement. Typically, these participants found that it was helpful, Being able
to hear yourself right away. You know, being able to hear your voice recorded and compared.
Ten participants emphasized the benefits of instruction in the mechanics of speech while nine commented
on the usefulness of the cultural information. They reported improvement in stressing syllables and
sentence elements with some stating that they were able to correct the pronunciation of words that had
long been stumbling blocks. The following comments reflect the features participants found most useful:
Everyday talk was the most helpful. Also how 'stress timing' is done. How to stress certain
syllables and words in sentences.
I really liked the part where, at the end of each page, you were shown how people talk in
America. Either how to pronounce things properly or what particular sayings mean.
This course in general was helpful to understand why some people dont understand foreigners
sometimes. Although, I am pretty sure that mostly its their inability to listen.
When I learned how to pronounce things correctly in English (for example, not pronouncing
every syllable like I would normally do), Filipino nurses saying you are speaking 'fake English'.
Like you're trying too hard and it doesn't sound right. But now I know it IS right!
Participants described two key benefits from the cultural information: 1) a better understanding of
common American English idioms and 2) ways to speak in the most professional and inoffensive manner.
Many acknowledged developing awareness of cultural norms in the United States that they were
previously unaware of, and several noted that they were already using the tips practical advice for
communicating in a more professional style in their workplaces and daily lives. Other respondents stated
that they would have liked more of this information, including a greater focus on terms specific to
healthcare professionals.
Participants were enthusiastic about the tip boxes, offering comments, such as the following:
The cultural information was helpful. I also liked the parts about informal/formal speaking.
I really liked these! They were very helpful and showed me how to speak 'American. I could
learn new sentences (sayings).
I liked learning about how words may have two meanings and what certain phrases mean.
I think these are very helpful. There are things you don't understand unless you understand the
culture around it.
I learned that saying I was assisting a crippled patient can be offensive, so I would say
disabled.
Most participants found the quizzes and tests helpful and motivating, and used the feedback to identify
areas where they needed more practice. One participant offered the following recommendation, Also, I
found the quizzes very helpful. I think you should have more quizzes.
18
Participants noted that it would be very useful for healthcare workers to complete the course before
coming to the United States. Several noted the absence of any similar educational programs in their
countries of origin. Many participants thought the course would also serve as a useful refresher for
foreign-born healthcare workers who are currently in the United States.
Asked what they would like to see added to the course, participants offered a number of recommendations
such as these:
It should be required because if it is optional, people won't do it. Not because they are lazy
but because they are so busy in school!
I think you should have more practice with consonants. I am from Taiwan, not the Philippines,
and we have more trouble with consonants like 'L'.
More practice on communication. If you could add short stories about the habits and culture
in America, I think we'd find that useful.
More video! This is more helpful than the graphs and descriptions that show people how to
speak. I would rather see someone's tongue placement, jaw movement, etc. than have it
described or just listen to it.
DISCUSSION
Scores on the TEPL-Online screening measure confirmed that the study population had the written
fluency needed to benefit from an Internet-based course.
Changing an accent is a slow and difficult process (Corrigan, 2010), and we did not expect to see
improvements on the POEC-S over the short duration of the study period. As expected, participants
scores on the POEC-S for auditory discrimination did not improve significantly but were relatively high
at pretest. On the other hand, participants scores on the POEC-S for verbal performance improved
significantly, suggesting that the practice presented in this early version of the course was beneficial.
Moreover, participants made significant gains on the Knowledge Test from pretest to posttest, suggesting
that they were learning the rules that native speakers of American English employ unconsciously.
There are several limitations to this study. First, the study sample was small and lacked a control group.
Second, 8 hours of practice time over a 3-month period is not sufficient for most speakers to show major
improvement. Third, users received telephone calls or emails to support their participation in the course,
a procedure that will not always be possible outside a research environment. Fourth, as mentioned earlier
in this paper, there may be some would not find computer-based practice an acceptable substitute for inperson instruction. Additionally, the LRC exercises will not be helpful for participants who do not hear
the difference between their own recordings and the model speakers. However, in the exit interviews,
participants frequently reported that the comparison with a model speaker was most helpful for
identifying areas for improving their own speech intelligibility.
In the next phase of development, we will add several features to support users, including a personalized
homepage where they will set personal goals and view an automated record of their progress in the course.
Content will be presented in small units suited to 20 minutes of practice in one sitting, and the personal
progress report will encourage daily practice. We will also incorporate social network support, including
automatically generated emails offering tips and encouragement and a discussion forum where learners
can share their experiences with other foreign-born nurses. We will evaluate the effectiveness of the
complete ICW in a two-group randomized controlled trial in which participants have access to the course
for 6 months. We will attempt to recruit nurses who are preparing to emigrate while they are still in their
native countries as well as newly arrived and recently employed nurses in the United States.
When complete, the ICW will address several of the challenges presented in developing web-based
19
training for adult learners, particularly those working in healthcare. Highly contextualized for nursing
and other healthcare professionals, the course is immediately relevant and useful, two crucial criteria for
adult learners. With automatic feedback, learners can readily see the areas where they need improvement
and are reinforced for their effort and progress. Learning strategies that generally work for adult learners,
visual memorization or self-tutoring, wont improve speech intelligibility. A self-directed course with
ample opportunities to practice speaking can be effective and should be worthwhile and engaging for
busy adults.
Healthcare professionals work under tremendous time constraints, making speech improvement classes at
community colleges, for example, unattractive. Speech trainers are often reluctant to offer courses at
hospitals because absenteeism is such a chronic problem. Yet the stakes are high for ineffective
communication in the stressful healthcare environment. Web-based training will appeal to healthcare
professionals who are already savvy with computers and who appreciate the convenience and flexibility
of working online at their own pace. Furthermore, technology that makes it possible to record speech to a
remote server and receive immediate feedback offers learners a dynamic practice environment free of
embarrassment in front of colleagues or burden on patients.
The ICW is a self-directed, rather than instructor-driven, course and is dependent on how self-directed
users are. It could be integrated into instructor-driven courses, either classroom or web-based, where it
would provide opportunities for review and practice with guidance from an instructor who is not always
present. Busy professionals could benefit from the structure and interaction provided in an instructordriven course and take advantage of the flexible practice time and the anonymity of a virtual community
of learners in the ICW.
CONCLUSION
This early field test of the ICW demonstrated that it is feasible to deliver speech intelligibility training
over the web. While computer-based speech intelligibility instruction is no substitute for in-person,
human feedback, an online course can provide a low-stakes environment for learners who wish to work
on the readability of their speech at their own pace.
In the next phase of development of the course, we intend to:
Expand the speech intelligibility content to meet the training needs of native speakers of the
major language groups represented in the nursing work force;
Develop additional content on vocabulary, communication practices, and culture in the U.S.
healthcare workplace, including interactive role play simulations;
Restructure the navigation to simplify the learning process for users and increase the length of
time users spend in the course; and
Add social networking features to promote learners commitment and deepen their engagement
by participating in a learning community.
ABOUT THE AUTHORS

Eileen Van Schaik is a Senior Research Scientist at Talaria, Inc. and a Clinical Assistant Professor,
Biobehavioral Nursing and Health Systems at the University of Washington. She is a medical
anthropologist and develops e-learning on culture and communication for healthcare professionals.
Additional research interests include healthcare disparities and end-of-life care.
20
Emily Lynch has degrees in linguistics, cultural anthropology and library and information science. She is
a freelance writer and consultant as well as an on-call librarian for the Seattle Public Library.
Susan Stoner is a clinical psychologist with expertise in research design and broad interests related to
health psychology and healthcare provision. She holds current appointments as a Research Scientist at
Talaria, Inc., and Affiliate Instructor in Anesthesiology & Pain Medicine at the University of Washington.
Lorna Sikorski is a speech pathologist and founder of LDS & Associates, which conducts research,
authors products and offers communication programs for adult English learners. She has authored
products, published and presented on: intonation, accent assessment/instruction, communication/learning
style. LDS Online Learning Center delivers professional education and client training.
REFERENCES
Abriam-Yago, K., Yoder, M., & Kataoka-Yahiro, M. (1999). The Cummins model: A framework for
teaching nursing students for whom English is a second language. Journal of Transcultural Nursing, 10,
143149.
Brooke, J. (1996). SUS: A quick and dirty usability scale. In Jordan, P.W., Thomas, B., Weerdmeester, B.
A., & McClelland, A.L. (eds.) Usability evaluation in industry. London, UK: Taylor and Francis.
Brush, B. L., Sochalski, J., & Berger, A. M. (2004). Imported care: Recruiting foreign nurses to U.S.
health care facilities. Nurse Migration, 23, 7887.
Buchan, J., & Aiken, L. (2008). Solving nursing shortages: A common priority. Journal of Clinical
Nursing, 17, 33623268.
Buerhaus, P. I., Auerbach, D. I., & Staiger, D. O. (2009). The recent surge in nurse employment: Causes
and implications. Health Affairs, 28(4), w657w668.
Clearfield, E., & Batalova, J. (2007, February 1). Foreign-born health-care workers in the United States.
Migration Policy Institute. Retrieved from http://www.migrationpolicy.org/article/foreign-born-healthcare-workers-united-states-0/
Corrigan, P. (2010). Whats that you said? Sometimes its not the words, its the accent. Health Progress,
91(4), 3437.
Davis, C. R., & Nichols, B. L. (2002). Foreign-educated nurses and the changing U.S. nursing workforce.
Nursing Administration Quarterly, 26, 4351.
Derwing, T. M., & Munro, M. J. (2005). Second language accent and pronunciation teaching: A researchbased approach. TESOL Quarterly, 39(3), 379397.
Doutrich, D. (2001). Experiences of Japanese nurse scholars: Insights for U.S. faculty. Journal of Nursing
Education, 40, 210216.
Ellenbecker, C. H. (2010). Preparing the nursing workforce of the future. Policy, Politics, & Nursing
Practice, 11(2), 115125.
Gamble, D. (2002). Filipino nurse recruitment as a staffing strategy. Journal of Nursing Administration,
32, 175177.
Guttman, M. S. (2004). Increasing the linguistic competence of the nurse with limited English proficiency.
The Journal of Continuing Education in Nursing, 35, 265269.
Hiep, P. H. (2007). Communicative language teaching: unity within diversity. ELT Journal, 61(3), 193
201.
Levis, J. M. (2005). Changing contexts and shifting paradigms in pronunciation teaching. TESOL
21
Quarterly, 39(3), 369378.

McMahon, G. T. (2004). Coming to America: International medical graduates in the United States. New
England Journal of Medicine, 350, 24352437.
Morton, E. S., Brundage, S. B., & Hancock, A. B. (2010). Validity of the proficiency in oral English
communication screening. Contemporary Issues in Communication Science and Disorders, 37, 153166.
Neal, S. L. (2002). Hiring foreign nurses in the U.S.A. International Journal of Nursing Practice, 8, 173
174.
Nunan, D. (1991). Communicative tasks and the language curriculum. TESOL Quarterly, 25(2), 279295.
Spry, C. (2009). Between two cultures: Foreign nurses in the United States. Global Perspectives, 89, 593
595.
Rathmell, G. & Sikorski, L.D. (2006). Test of English Proficiency Level (TEPL) (Electronic ed.). Santa
Ana: LDS & Associates, LLC.
Sikorski, L.D. (2004). Mastering effective English communication CD series (5th Ed.). Santa Ana: LDS &
Associates.
Sikorski, L.D. (2005a). Foreign accents: Suggested competencies for improving communicative
pronunciation. Seminars in Speech and Language, 26, 126131.
Sikorski, L.D. (2005b). POEC Screen, Electronic Edition (Proficiency in Oral English Communication
Screen) (3rd ed.). Santa Ana: LDS & Associates, LLC.
Trossman, S. (2002). The global reach of the nursing shortage: The ANA questions the ethics of luring
foreign-educated nurses to the United States. American Journal of Nursing, 102, 8587.
Xu, Y. (2003). Are Chinese nurses a viable source to relieve the U.S. nurse shortage? Nursing Economics,
21, 269274.
Xu, Y., Bolstad, A. L., Shen, J., Colosimo, R., Covelli, M., Torpey, M., & Jorgenson, M. (2010). Speak
for success: A pilot intervention study on communication competence of post-hire international nurses.
Journal of Nursing Regulation, 1(2), 4248.
Xu, Y., & Davidhizar, R. (2004). Conflict management styles of Asian and Asian American nurses
implications for the nurse manager. The Health Care Manager, 23, 4653.
Yi, M., & Jezewski, Y. M. (2000). Korean nurses' adjustments to hospitals in the United States of
America. Journal of Advanced Nursing, 32, 721729.
22


pp. 2337
EXPLORING HOW COLLABORATIVE DIALOGUES FACILITATE

SYNCHRONOUS COLLABORATIVE WRITING
Hui-Chin Yeh, National Yunlin University of Science and Technology
Collaborative writing (CW) research has gained prevalence in recent years. However, the
ways in which students interact socially to produce written texts through synchronous
collaborative writing (SCW) is rarely studied. This study aims to investigate the effects of
SCW on students writing products and how collaborative dialogues facilitate SCW.
Following an initial analysis, 54 students were divided into 18 groups; six groups with
higher proportions of collaborative dialogue (HCD), six groups with median proportions
of collaborative dialogue (MCD), and six groups with lower proportions of collaborative
dialogue (LCD). The data collected includes the students three reaction essays, their
transcripts of text-based collaborative dialogues, and their writing process logs. The results
showed that there were significant differences between the LCD, MCD, and HCD groups
in terms of fluency and accuracy of their reaction essays. Through collaborative dialogues,
students benefitted from text-based synchronous communications, such as clarifying their
linguistic misconceptions, and receiving immediate feedback to help resolve their writing
problems. The findings suggest that students could be provided with more opportunities
for collaborative dialogues during the entire writing process, including the stages of
generating ideas, writing reaction essays, and editing.
Keywords: Collaborative Learning, Writing, Collaborative Dialogues
APA Citation: Yeh, H.-C. Exploring how collaborative dialogues facilitate synchronous
collaborative writing. Language Learning & Technology 18(1), 2337. Retrieved from
Received: April 9, 2013; Accepted: August 30, 2013; Published: February 1, 2014
Copyright: Hui-Chin Yeh
INTRODUCTION
Writing is an important skill for college students, as professional and academic success in all disciplines
depends, at least in part, upon writing skills (Cho & Schunn, 2007, p.409). For many English as a
Foreign Language (EFL) teachers, developing students writing skills has proved challenging. Teachers
often have a limited understanding of students writing processes. An understanding of this process is
necessary for them to help students develop into proficient writers (Florio-Ruane & Lensmire, 1990). EFL
teachers writing pedagogy is often derived from their own learning experiences as a pupil, rather than
from the evaluation and observation of students actual writing processes (Florio-Ruane, 1989). For many
teachers, understanding the difficulties that students have in writing is influenced by personal experience
and may not address the specific needs of every student. These problems are often exacerbated by too
much emphasis on corrections of grammatical errors in students writing. While this is often common
practice in L2 writing instruction, these teaching approaches neglect the global view of writing, such as
the incorporation of brainstorming ideas and organizing writing into logical paragraphs (Lee, 2004;
Sotillo, 2000). As a consequence, in recent years many researchers and educators have been looking for
an effective approach to enhance students writing (e.g. Elola & Oskoz, 2010; Kessler & Bikowski, 2010)
Collaborative Writing
The prevalence of collaborative writing (CW) research in recent years attests to the potential contributions
of CW in different aspects, such as higher quality of writing (Beck, 1993; Storch, 1999; Storch, 2005), a
better understanding of the reader-writer relationship (Leki, 1993), and the acquisition of writing
23
Hui-Chin Yeh
How Collaborative Dialogues Facilitate Collaborative Writing
knowledge, including grammar, vocabulary usage, and text structures (Kowal & Swain, 1994; Swain &
Lapkin 1998). CW is defined as collaborators producing a shared document, engaging in substantive
interaction about that document, and sharing decision-making power and responsibility for it (Allen,
Atkinson, Morgan, Moore, & Snow, 1987, p. 70). CW focuses on the social interaction process, where
two or more people, through discussion, work together to construct written documents, reach consensus
on resolutions of questions and quality of work, and coordinate individual contributions on various
aspects of writing (e.g. Elola & Oskoz, 2010; Kessler & Bikowski, 2010; Storch, 1999; Storch, 2005). In
the social interaction process, students contribute their particular ideas or expertise, while taking into
account others perspectives, in order to collaboratively complete a writing task (Speck, Johnson, Dice, &
Heaton, 1999).
Some potential Web 2.0 technologies for CW include Wiki and Google Docs. Wiki refers to an
asynchronous networking tool where users have equal opportunities to asynchronously organize,
compose, and revise content (Kessler, 2009; Lee, 2010; Elola & Oskoz, 2010) at any time. One potential
use of Wiki is its tracking tool, History. It documents the collaborative writing processes so that
teachers can recognize the changes students have made on texts, and trace who makes the changes.
Google Docs refers to a synchronous and asynchronous networking tool that allows writers to share and
access written documents over the internet in real time. Students can either synchronously or
asynchronously create, edit, or revise written documents, with synchronous communication supported by
chat rooms. The tracking tool is provided in Google Docs to manage different versions of written
documents, as well as record the time and date they are modified. While asynchronous communication
provides a useful method of communicating with peers, synchronous methods permit the immediate
addressing of key topics across a potentially wide audience.
Researchers have recognized some of the benefits of text-based synchronous communication, which (a)
focuses on meaning rather than on form (Kessler, 2009), (b) improves fluency and accuracy of
communication (Elora & Oskoz, 2010; Lee, 2001), (c) values the chance to share ideas and provide
feedback (Ware & ODowd, 2008) and (d) enhances language learning motivation in general (Cononelos
& Oliva, 1993; Oliva & Pollastrini, 1995). Students also benefit from text-based synchronous
communication by immediately having their linguistic misconceptions and writing problems addressed.
They receive timely constructive feedback from peers which helps them make their writing more
meaningful and comprehensible to others (Lee, 2002; Webb, 1989). For example, Lee (2002) designed
collaborative writing activities to enhance students writing proficiency through a synchronous discussion
forum in Blackboard, which acts as an online communication tool allowing students to have synchronous
interactions and consultations with others, in order to collaboratively accomplish writing tasks. In
observing real-time synchronous collaborative writing processes, students are exposed to linguistic input
alongside the vocabulary or sentence structures from written documents that they can co-construct and coedit (Lee, 2002). As a result, students may apply collective linguistic input to self-correct or edit texts. A
SCW tool offers a text-based synchronous forum for students to carry out collaborative dialogues and
obtain immediate feedback in congruence with face-to-face (F2F) collaborative dialogues (Blake, 2000;
Kessler & Bikowski, 2010; Smith, 2003). Swain and Lapkin (2002) perceive collaborative dialogues as an
externalization of thoughts which can be scrutinized, questioned, reflected upon, disagreed with,
changed, or disregarded (p. 286). Since text chats provide an avenue for students to reflect and negotiate
meanings with peers on the basis of collaborative dialogues in written forms, they also allow students to
elaborate on their ideas more clearly and attend to linguistic output so that students can better understand
the comments and feedback that lead to L2 improvements (Koschmann, Kelson, Feltovich, & Barrows,
1996; OSullivan, Mulligan & Dooley, 2007). For example, Wells and Chang-Wells (1992) explored the
effectiveness of text-based synchronous collaborative dialogues upon argumentative writing. Their results
showed that the text-based synchronous collaborative dialogues fostered literate thinking development.
That is, students performed much better when elaborating their ideas in written argumentative essays,
24
Hui-Chin Yeh
while given the opportunity to perform text-based synchronous collaborative dialogues.

Research Gap in CW Research
The social interaction process, namely how students produce written texts through CW, is difficult to
conceptualize and observe empirically, and so it is not well understood (Swain, 2000). According to
sociocultural theory, the social interaction process cannot be disregarded, since language learning always
occurs in the process of social interaction rather than in writing products (Donato, 1994; Lee, 2004b).
Studies have shown that the social interaction process can provide valuable information which may not be
directly observed from writing products (Masoodian & Luz, 2001). Understanding what factors may
affect the quality of the social interaction process when incorporating Web 2.0 tools is valuable. For
example, Lee (2004a) examined the social processes of networked collaborative interaction by using
Blackboard. The results showed that students language proficiency, computer skills, and ages, are the
core factors that determined the success of online negotiation and influenced students learning
motivation. Brodahl, Hadjerrouit, and Hansen (2011) also pointed out key factors, such as learning tasks,
course content, perceptions toward tools, and prerequisite knowledge, which may result in different levels
of collaboration and learning outcomes in CW.
Currently, the effectiveness of SCW remains relatively unexplored. Many scholars (e.g. Lowry, Curtis, &
Lowry, 2004; Storch, 2005) have addressed compelling needs for more studies to look into social
interaction processes in synchronous modes. Only a few L2 studies have attempted to investigate
collaborative dialogues in SCW (e.g., Digiovanni & Nagaswami, 2001; Jones, Garralda, Li, & Lock,
2006). Among these attempts, collaborative dialogues have been limited to the co-editing stage, where
students edit and provide feedback on peers texts in order to produce a final writing product (Storch,
2005). Nixon (2007) similarly argues that most of the conditions under which students are given
opportunities in the classroom to engage in dialogues are concerned with only one part of the entire
writing process, (p. 6) namely co-editing. Jones, Garralda, Li, and Lock (2006) also examined
collaborative dialogues of L2 students in both online and F2F in co-editing, based on two types: initiating
moves (e.g., offer, directive, statement, and question) and responding moves (e.g., clarification,
confirmation, acceptance, rejection, and acknowledge). Results showed that students raised more
questions and made more comments online than through F2F in the co-editing process. They also
reported that collaborative dialogues in co-editing focus much more on the micro-level feedback (e.g.,
vocabulary and grammar) in the F2F session, and are more concerned with the macro-level feedback
(e.g., content, organization, and topic) in online co-editing.
Collaborative dialogues in co-editing seemed to result in superficial levels of writing in which students
only concentrated on identifying either micro-level or macro-level writing problems (Jones et al., 2006).
When students are only engaged in the final stage of writing, namely co-editing, they might lose sight of
the entire writing process for generating insights into the deeper meaning of the writing (Hirvela, 1999).
To discourage the collaborative dialogues from centering on superficial CW processes of simply finding
and fixing errors, and raising the process to a comprehensive level for the total extent of writing, L2
instructors and researchers are encouraged to design the tasks which involve the entire process of CW.
Research Questions:
The core objectives of the current study, scheduled to run for one semester, were to investigate the effects
of SCW upon writing products and how collaborative dialogues facilitated SCW. Based on the research
purposes, the research questions included:
1. Do highly collaborative groups produce higher quality writing products?
2. How do the collaborative dialogues facilitate SCW?
25
Hui-Chin Yeh
Research Background
In this current study, 18 groups, comprised of three students each, were required to write three reaction
essays synchronously and collaboratively. The researcher, in the role of instructor, assisted the 54 EFL
students in writing the reaction essays using EtherPad (EP). Like Google Docs, EP is a word processing
software application that includes similar functionalities to support synchronous writing, editing, and text
chatting, and documents historical records of writing processes and different versions of written products.
EP is free open-source software that users can use without license restrictions. Writers can collaboratively
write and edit a shared document in real time with text chats embedded in the same page (see Figure 1).
Each author in EP is provided with a unique text highlighting color and authors can save the different
versions of drafts or edit them at any time. Periodically saved documents are historically recorded for
writers to visualize the developmental process using a flash form in Time Slider (see Figure 2). To date,
EP has yet to be examined as a potential venue for SCW in L2 writing classrooms (Chu, Kennedy, &
Mak, 2009).
Bettys editing
with color coded
Records of
collaborative dialogues
Written texts
Figure 1. Real time collaborative writing in EtherPad

In writing the reaction essay, the students were required to read an article assigned to them by the
instructor. After reading, the students collaborated with their group members to write down how they
reacted to the assigned reading. The students were given one-hour out of a two-hour class on three
occasions during a semester, to complete the writing of three reaction essays. The students were also
required to continue to complete the SCW tasks after class in their own time. Through collaborative
writing, students could exchange their understanding and knowledge about writing conventions, the
reader-writer relationship, and genre rules to produce high quality reaction essays (Kowal & Swain, 1994;
Leki, 1993; Swain & Lapkin, 1998).
While writing the reaction essay, the students were encouraged to use text chats in the right column of EP
to negotiate meanings with each other throughout the entire SCW process. After completing and saving
the reaction essay in the EP system, the students obtained a hyperlink to their written text, which was
generated automatically by the EP system. Each group then posted the hyperlink in an online course
management system, E-Campus, for the instructor and other students to read.
26
Hui-Chin Yeh
Save different
written versions
View saved different

versions of text
developments
Figure 2. Time Slider as the tracking tool to view different saved versions
METHODOLOGY
Participants
The 18 groups consisted of 54 non-English major students registered in a two-credit college course,
Vocabulary and Reading, at a university in central Taiwan. These non-English majors were first-year
students from the Colleges of Management, Design, and Engineering. The 54 students had been learning
EFL for more than 10 years and were not proficient in English writing, which required higher cognitive
levels in application, organization, and integration of existing English writing knowledge. Individual
students from the 18 groups took the writing section of TOEFL to measure their writing proficiency
during the first week of the semester. Their writing proficiency scores ranged from 11 to 22 on a 30-point
scale (M = 15.9).
The students confirmed their willingness to participate by signing a consent form informing them of the
research scope and data to be collected. They were also informed that they could withdraw from the study
at any time. All names were displayed as pseudonyms in written communications to ensure the
participants anonymity.
Data Collection and Analysis
Data sources included (a) the students three reaction essays, (b) the transcripts of text-based collaborative
dialogues, and (c) the students writing action logs from Time Slider, the system function providing
historical records of the student writing processes using EP. The three essays and collaborative dialogues
were analyzed to examine the effects of the SCW upon the fluency and accuracy of the writing products
and how collaborative dialogues facilitated SCW.
27
Hui-Chin Yeh
At the initial analysis stage, the 18 groups were classified into groups with high proportions of
collaborative dialogue (HCD groups), median proportions of collaborative dialogue (MCD groups), and
low proportions of collaborative dialogue (LCD groups). To determine the HCD, MCD and LCD groups,
the researcher first counted the frequency of collaborative dialogues in each of the 18 groups. The
minimal analytical unit to count the frequency of the collaborative dialogue was turns. Each time a person
typed some text in a text chat counted as one turn. The HCD groups were in the top 33% of the collective
mean of the 18 groups collaborative dialogue frequency, the MCD groups were in the median 34%, and
the LCD groups were in the bottom 33%.
The English writing proficiency of the HCD groups ranged from TOEFL writing scores of 13 to 22 (M =
16.5). The range of the MCD groups writing proficiency was 13 to 20 (M = 16.1). The LCD groups
writing proficiency was 11 to 19 (M = 15). A one-way analysis of variance (ANOVA) showed nosignificant differences (F = 0.398, p >.05) between the HCD, MCD, and LCD groups in terms of
students writing proficiency. In other words, the HCD, MCD, and LCD groups were evaluated as having
equivalent writing proficiency.
The rubrics, which include the major generic elements for reaction essays, were used to rate the three
essays. The T-units as one main clause plus whatever subordinate clauses happen to be attached to or
embedded within it (Hunt, 1966, p. 735) served as the references to measure the text quality of student
essays in terms of fluency (Frequency of T-unit use), and accuracy (the percentage of error-free T units)
(Polio, 1997; Spelman Miller, 2006). Meetings were held regularly between the researcher and the
research assistant to cross-examine the reliability of the analysis of fluency and accuracy. The inter-rater
reliability of the text fluency and accuracy were determined as .88 and .84 respectively.
A multivariate analysis of variance (MANOVA) was conducted to investigate whether the proportions of
collaborative dialogues had any effect on Essay 1, Essay 2, and Essay 3 in terms of fluency and accuracy
respectively. The researcher selected one HCD groups reaction essay and one LCD groups reaction
essay to serve as sample texts to more closely analyze whether highly collaborative groups produced
higher quality writing products. The selection strategy was typical case sampling (Patton, 1990), as it is
often used to describe and illustrate a program or a participant to people not familiar with the program
(p. 173). As Patton indicates, cases could be selected from a demographic analysis of averages or other
programmatic data that provide a normal distribution of characteristics from which to identify "average"
examples (p. 174). The sample essays of the HCD and LCD groups were selected from essays where
fluency and accuracy scores were close to the LCD and HCD groups collective mean scores.
The collaborative dialogues were analyzed using content analysis (Weber, 1990; Patton, 1990) to explore
how the collaborative dialogues facilitated SCW. The researcher and the research assistant first coded the
transcripts of the dialogues into 627 meaningful turns. Then, the researcher and assistant categorized the
meaningful statements into different thematic units for the three writing stages. The collaborative
dialogues, centering on the stages of (a) generating ideas, (b) writing reaction essays, and (c) editing the
reaction essays, were analyzed. The inter-rater reliability of the coding for the three stages was
determined as .86, .82, and .87. The examples of collaborative dialogues were illustrated for different
stages. The main ideas were interpreted by offering examples, drawing conclusions, and making
inferences. One of the HCD groups was selected by the typical case sampling method to explain how the
students used the collaborative dialogues to help (a) generate ideas, (b) write the reaction essays, and (c)
edit the reaction essays during the entire SCW process. The action logs served as the supplementary
source of information that triangulates the research findings.
RESULTS
RQ1: Do highly collaborative groups produce higher quality writing products?
A MANOVA was conducted to determine whether the proportions of collaborative dialogues had any
28
Hui-Chin Yeh
effect on the three reaction essays in terms of fluency and accuracy respectively. The results showed that
the proportions of collaborative dialogue in the different groups had a significant effect on Essay 1, Essay
2, and Essay 3 (Pillais Trace = 1.73, F = 12.14***) in terms of fluency and accuracy. The post hoc
analysis with Sheff Test showed that there were significant differences between the LCD, MCD, and
HCD groups in the three essays in terms of fluency and accuracy (Table 1). The results showed that
highly collaborative groups can produce high quality reaction essays.
Table 1. MANOVA analysis of the LCD, MCD and HCD groups essays in terms of fluency and accuracy
MANOVA
ANOVA
DV
Pillais Trace
Post hoc
Essay 1-fluency
1.73
12.14***
19.40***
MCD>LCD
MCD<HCD
MCD>LCD
MCD<HCD
MCD>LCD
MCD<HCD
MCD>LCD
MCD<HCD
MCD>LCD
MCD<HCD
MCD>LCD
MCD<HCD
Essay 2-fluency
18.38***
Essay 3-fluency
17.48***
Essay 1-accuracy
14.74***
Essay 2-accuracy
13.29***
Essay 3-accuracy
19.23***
Note: ***<0.01
To examine how the collaborative dialogues facilitated students writing performance, one HCD groups
reaction essay and one LCD groups reaction essay were selected as sampling texts to explain how their
collaborative dialogues helped improve their writing products to different extents (see Table 2 & 3).
Table 2. Examples of the Reaction Essays Selected from the HCD Group and the LCD Group
Group
Examples of the main idea
Examples of supporting ideas
Fluency
Scores
Analysis of reaction
essays
LCD
We do not think it is fair.

Because his prosthetic legs
are less flexible and
cooperative than normal.
So, we think disabled athletes

should not compete with ablebodied athletes at same time.
41
No supporting ideas
are used to support
the main idea
HCD
Although it is difficult to
make the competition fair
in the official game, here
are some methods to aid
you on how to be fair in
the competition.
First, you can find a

prestigious person to preside
over the game. Second, it is
also good to make some rules
in advance. Each player would
be required to follow them to
make the competition fair.
Finally, no un-sportsman like
conduct, every player should
have fun.
90
Three concrete
supporting ideas are
used to strengthen
the main idea
29
Hui-Chin Yeh
The MCD groups reaction essays were not presented, as the essays displayed similar findings to the ones
produced by the HCD groups. As shown in Table 3, the HCD groups main idea was presented in the first
sentence of the reaction essay, noting Although it is difficult to make the competition fair in the official
game, here are some methods to aid you on how to be fair in the competition. To support this main idea,
the HCD group proposed three supporting ideas, marked by the transition words first, second, and finally.
Table 3. Examples of the Reaction Essays Selected from the HCD Group and the LCD Group
Group
An excerpt from the reaction essays
Analysis of reaction essays
LCD
We do not think that it is fair. Although his story makes us feel

impressive. Overall, we believe that the normal person would
have more advantages. Because his prosthetic legs are less
flexible and cooperative than normal. So, we think disabled
athletes should not compete with able-bodied athletes at same
time.
Incorrect use of
subordinating
conjunctions
HCD
Although it is difficult to make the competition fair in an

unofficial game, here are some methods to make the official
game fair. First, you can find a prestigious person to preside over
the game. Second, it is also good to make some rules for each
player to follow to make the competition fair. Finally, no unsportsman like conduct, every player should have fun
Correct use of
subordinating
conjunctions and
transition words
The LCD group, however, wrote their reaction essays without providing supporting ideas and examples in
the reaction essays. Table 2 shows that, in a selective reaction essay from one of the LCD groups, the
supporting ideas of the topic sentence were not elaborated upon before reaching the conclusion. As a
result, the reaction essays of the LCD groups were rated less fluent (a score of 41), when compared with
the essays submitted by the HCD groups (a score of 90).
The results suggest that students active engagement in collaborative dialogue with their peers could have
had a positive impact on the fluency and accuracy of the essays. The HCD groups made fewer
grammatical errors compared with the LCD groups when writing the reaction essays (see Table 2, 3, and
4), the LCD groups work often contained grammatical errors such as when using pronouns and
subordinating conjunctions. In the LCD groups it was often hard to locate the antecedent of the pronoun.
For example, the LCD group started writing the reaction essay with the passage We do not think that it is
fair. Although his story makes us feel impressive. Overall, we believe that the normal person would have
more advantages. Because his prosthetic legs are less flexible and cooperative than normal in which the
antecedent of the possessive pronoun his remains unclear and may refer to either the noun phrase normal
person or the disabled person in the text. This example also shows that the LCD group had difficulty
using conjunctions correctly.
Table 4. Proportions of the Collaborative Dialogues During SCW
Group
Generating Ideas
HCD
241
(28%)
537
(61%)
96
(11%)
874
MCD
201
(28%)
434
(60%)
87
(12%)
722
LCD
152
(29%)
304
(58%)
66
(13%)
522
Total
594
(28%)
1275
(60%)
249
(12%)
2118
Writing reaction essays
Editing reaction essays
Total
30
Hui-Chin Yeh
RQ2: How do the collaborative dialogues facilitate SCW?

All of the students were encouraged to employ collaborative dialogues to discuss how to (a) generate
ideas, (b) write reaction essays, and (c) edit the reaction essays. The results showed that collaborative
dialogues often took place during the process of writing reaction essays. Sixty per cent of the
collaborative dialogues centered on how to write the reaction essays, 28% of the collaborative dialogues
centered on how to generate ideas, and 12% centered on how to edit the reaction essays (see Table 4). The
proportion of collaborative dialogues that centered on how to write reaction essays was about twice the
proportion of collaborative dialogues that centered on how to generate ideas and about five times that
centered on editing the reaction essays. The collaborative dialogues from one of the HCD groups were
selected as an example to demonstrate how collaborative dialogues facilitated students to (a) generate
ideas, (b) write the reaction essays, and (c) edit the reaction essays during the entire SCW process.
The students began writing the reaction essays collaboratively by using collaborative dialogues to
generate writing ideas, as shown in Table 5. The students initiated the conversations by addressing their
concerns on the writing topic.
Table 5. Examples of how collaborative dialogues facilitate the HCD group to generate ideas
Interaction process
Collaborative dialogues
Formulating the
writing topic
Tim: There are many directions we may go about for this essay. Our
reading is about the handicapped should not compete against able-bodied
athletes
Amy: What do we want to write about?
Tim: Maybe we can write about fairness.
John: Our topic could be In an informal competition, how to judge
whether it is fair or not?
Eliciting ideas
Amy: Do you have any ideas?

Tim: Yes. We can discuss our ideas first. We may start with main ideas
and supporting ideas first.
Contributing ideas
Amy: First, enjoy the competition without cheating. Second, look for a
prestigious person to host the competition.
John: Third, set up the rules. Anything else?
Tim: But these came from our reading. We should have our own opinions.
Amy: I am still thinkingHow about cultivate good sportsmanlike
conduct? Is that appropriate?
Iterative seeking
clarification
John: I am not sure. Is there any relation between good sportsman like
conduct and holding a fair competition?
Explaining
Amy: Yes, if you have good sportsmanlike conduct, you will enjoy
yourself at the game. You do not care about the contest results
John: I dont quite understand.
Iterative clarifying
Amy: By good sportsmanlike conduct, I mean joining in the contest for

self-accomplishment not for competition with others.
Tim: O.K. it will be a direction for us to write about. To discuss the
purpose of being in a competition.
As shown in Table 5, Tim proposed that, There are many directions we may go about for this essay and
they started to discuss how to write and organize their ideas in an essay. Cognitive conflicts occurred
31
Hui-Chin Yeh
when John was confused with Amys statements and asked Amy to provide further elaborations to clarify
the relation between ideas. Through constant meaning negotiations, the group members and Amy
resolved emerging confusion, reached a consensus on how to use the main and supporting ideas to
strengthen their reaction essays, and determined a writing topic together.
After generating ideas, the students used the proposed ideas to write their reaction essays. When writing
the reaction essays, the collaborative dialogues assisted the students in clarifying any misunderstanding of
English vocabulary and helped them understand how to write and develop ideas. John and Amy started
the writing task by listing the developed writing ideas they had generated in English. As shown in Table 6,
while constructing the English sentences, John was hindered by his limited English vocabulary and asked
Amy about the meaning of competition by saying Does competition mean the person who joins in the
contest? Based on Amys responses, John recognized competition as the noun of the verb to compete,
and used player to identify people who compete with each other in contests. However, the other group
member Tim did not agree with Johns English translation of the word player, as people who join in a
contest, and provided the noun competitor, as the other English translation of people who compete in a
contest. By receiving feedback from Tim, John expanded his vocabulary and usage of player and
competitor.
After listing the developed writing ideas in English, Amy discussed with Tim and John about how to
integrate the developed writing ideas into a passage by noting We could first discuss how to integrate the
content. Tim then responded, We can write down the main idea, followed by providing examples.
John also suggested that some transition words (such as first, second, and finally) could be used to help
connect the ideas between sentences for readers to follow the arguments.
Table 6. Examples of how collaborative dialogues facilitate the HCD group to write the reaction essay
Interaction process
Seeking clarification
John: Does the word competition refer to people who enter a contest?
Providing feedback
Amy: Probably not. It is the noun for the verb to compete.

John: Fine, then we can use the word player.
Providing alternate
feedback
Tim: It is not appropriate at all. I looked it up in the dictionary, and it

shows that competitor is more suitable.
Clarifying/formulating
response
John: Is player inappropriate?

Tim: It is fine, but it seems that the word competitor is more correct
than player
Proposing alternatives
Amy: Then, shall we start writing our summary? We could first discuss
how to integrate the content.
Providing solution
direction
Tim: We can write down the main idea, followed by providing

examples.
Providing concrete
example
John: We can connect the sentence with a sequence marker like First,
Second, and Finally. By doing so, we can clarify each statement.
Once the students finished their writing, they collaboratively edited the reaction essays. In editing the
reaction essays, the collaborative dialogues not only allowed students to expand their knowledge of
grammar and vocabulary, but also exerted collaborative efforts to correct incorrect grammatical sentences
and revise semantically incorrect sentences. As shown in Table 7, Tim found a grammatical error in the
32
Hui-Chin Yeh
sentence With these rules, everyone has to follow it written by Amy. Then, Tim negotiated with Amy
about the grammatical use of pronouns them. Through the collaborative dialogues with Tim, Amy
reconstructed her understanding of the pronoun them. They thus expanded their grammatical knowledge
through collaborative dialogues. Another example is how they negotiated meanings with each other to
edit an appropriate phrase that reflected their intended meaning. Through collaborative dialogues, they
changed the word solemn into take more care at the game. They explored the possibilities for using
different words or phrases to express their meanings.
Table 7. Examples of how the collaborative dialogues facilitate the HCD group to edit a reaction essay
Original writing
texts
Revised
writing texts
Analysis of
collaborative
dialogues
With these rules,

everyone has to
follow it.
Tim: Rules is the plural form, isnt it? We

should use it, rather than them
Amy: But, them only represents humans,
doesnt it?
Tim: Them also can stand for the object.
Amy: WellLets revise it.
With these
rules,
everyone has
to follow
them.
Expanding
grammatical
knowledge
about pronouns
Because it can let

player more
solemn.
Tim: I think it is not appropriate to put the

word solemn here.
John: solemnmeans take something
seriously?
Tim: Right, any other word that we can alter?
John: How about substituting the solemn
with care
Tim: I think it is somehow weird. Let me think
twice.
John: How about revising it to be take care at
the game?
Tim: Fine.
John: Ok, I revise it.
Because it
can let player
take more
care at the
game.
Expanding
knowledge
about
vocabulary and
phrases
CONCLUSIONS
When engaged in collaborative dialogue in SCW, students are exposed to bountiful linguistic sources of
exposure to language input and output which in turn contribute to L2 writing development. Students can
exchange their understanding and knowledge about vocabulary, sentence structures, and idea organization
to collaboratively produce written texts. The findings suggest that highly collaborative groups can
produce high quality reaction essays in terms of fluency and accuracy. It was suggested that students
should be provided with more opportunities to exchange information or ideas during the entire writing
stages, including generating ideas, writing reaction essays, and editing. The findings responded to
previous research, in that students benefitted from text-based synchronous communications, realized their
linguistic misconceptions, and received immediate feedback to help resolve their writing problems (Lee,
2002; Webb, 1989). The results also showed that students often centered collaborative dialogues on how
to write the essays more than how to generate ideas and edit the texts to complete reaction essays. Such
findings echo some previous studies, in that students participating in the entire SCW process often placed
more emphasis on the meaning negotiation of the content than on the language forms (Kessler, 2009).
33
Hui-Chin Yeh
In the social interaction process, collaborative dialogues provide an avenue for students to use language
and reflect on their language use. Collaborative dialogues can be regarded as a form of mediation for
facilitating L2 writing, particularly when students work together with peers to co-construct a written
document. Collaborative dialogues often occur when students notice linguistic problems and work
together to solve them, leading to L2 knowledge construction. As a result, Swain (2000) specially referred
this concept to dialogues with which speakers are engaged in problem solving and knowledge building,
and dialogues heighten the potential for exploration of the product (p. 102). In other words, through
collaborative dialogues, students can construct new or half-developed ideas, since students in a group
collaboratively generate and refine the ideas. Students may elicit challenging questions and provide
constructive feedback to promote knowledge development. The feedback in turn often motivates students
to monitor and evaluate their learning processes. Through monitoring and evaluating in the social
interaction process, students can obtain a broader scope of their own problems to adjust their writing.
Using EP, teachers could save different versions of students written documents, and then use the Time
Slider feature, to monitor students writing processes and examine students developmental processes of
text quality. Moreover, EP also historically recorded all the transcripts of students collaborative
dialogues in the chat room. By observing the transcripts of students collaborative dialogues, teachers can
understand how students develop writing products, negotiate meanings, and resolve any
misunderstandings with peers during the SCW process.
While this research contributed to the effectiveness of SCW in EFL students writing products, limitations
remain. Because this research was conducted in an EFL context to investigate the effects of SCW and
collaborative dialogues on writing, the results might not be generalized for other English teaching
contexts, such as ESL. In addition, the research is limited in its one-semester implementation timeline.
Longitudinal studies could be conducted to investigate whether time is a key factor for successful SCW.
ACKNOWLEDGEMENTS
This research was partially supported by the Ministry of Science and Technology of Taiwan under grant
NSC 101-2410-H-224-021 and NSC 102-2410-H-224-011.
ABOUT THE AUTHOR

Hui-chin Yeh is currently an associate professor in the Graduate School of Applied Foreign Languages at
National Yunlin University of Science and Technology in Taiwan. She received her Ph.D. in Language
Education at Indiana University-Bloomington. Her research interests center on EFL teacher education,
computer-assisted language learning, and EFL reading and writing. She has published many articles on
these topics in Language Learning & Technology, Computer Assisted Language Learning, Teaching and
Teacher Education, British Journal of Educational Technology, Educational Technology Research &
Development, ReCALL, Educational Technology & Society, Australasian Journal of Educational
Technology, Asia-Pacific Education Researcher, and Asia Pacific Education Review. She has received
several awards for her work, among them a distinguished young researcher award from the Ministry of
Science and Technology, a 2010 teaching excellence award, and a 2011 excellence in mentoring award.
Her Multimedia English course has been certified as a quality e-learning course from the Ministry of
Education in Taiwan. She hopes that her efforts in different aspects of her work can make significant
contributions to her academic field of study.
34
Hui-Chin Yeh
REFERENCES
Allen, N. J., Atkinson, D., Morgan, M., Moore, T., & Snow, C. (1987). What experienced collaborators
say about collaborative writing? Journal of Business and Technical Communication, 1(2), 7090.
Beck, E. E. (1993). A survey of experiences of collaborative writing. In M. Sharples (Ed.): Computer
supported collaborative writing (pp. 87-112). London: Springer-Verlag.
Blake, R. (2000). Computer mediated communication: A window on L2 Spanish interlanguage. Language
Learning & Technology, 4(1), 120136. Retrieved from http://llt.msu.edu/vol4num1/blake/default.html
Brodahl, C., Hadjerrouit, S., & Hansen, N.K. (2011). Collaborative writing with Web 2.0 technologies:
Education students perceptions. Journal of Information Technology Education: Innovations in Practice,
10, 73103.
Cho, K. & Schunn, C. D. (2007). Scaffolded writing and rewriting in the discipline: A web-based
reciprocal peer review system. Computers & Education, 48, 409426.
Chu, S., Kennedy, D., & Mak, M. (2009). MediaWiki and Google Docs as online collaboration tools for
group project co-construction. Proceedings of the 2009 International Conference on Knowledge
Management. Hong Kong, Dec 3-4, 2009.
Cononelos, T., & Oliva, M. (1993). Using computer networks to enhance foreign language/culture
education. Foreign Language Annals, 26(4), 527-534.
DiGiovanni, E., & Nagaswami, G. (2001). Online peer review: an alternative to face-to-face? ELT
Journal, 55(3), 263272.
Donato, R. (1994). Collective scaffolding in second language learning. In J. P. Lantolf & G. Appel (Eds.),
Vygotskian approaches to second language research (pp. 33-56). Norwood, NJ: Ablex.
Elola, I., & Oskoz, A. (2010). Collaborative writing: Fostering foreign language and writing conventions
development. Language Learning & Technology, 14(3), 5171. Retrieved from
http://llt.msu.edu/issues/october2010/elolaoskoz.pdf
Florio-Ruane, S. (1989). Social organization of classes and schools. In M. C. Reynolds (Ed.), Knowledge
base for the beginning teacher (pp.163-172). Oxford: Pergamon.
Florio-Ruane, S. & Lensmire, T. (1990). Transforming future teachers' ideas about writing instruction.
Journal of Curriculum Studies, 22(3), 277289.
Hirvela, A. (1999). Collaborative writing: Instruction and communities of readers and writers. TESOL
Journal, 8(2), 712.
Hunt, K. (1966). Recent measures in syntactic development. Elementary English, 43, 732739.
Jones, R. H., Garralda, A., Li, D. C. S., & Lock, G. (2006). Interactional dynamics in on-line and face-toface peer-tutoring sessions for second language writers. Journal of Second Language Writing, 15, 123.
Kessler, G. (2009). Student-initiated attention to form in wiki-based collaborative writing. Language
Learning & Technology, 13(1), 7995. Retrieved from http://llt.msu.edu/vol13num1/kessler.pdf
Kessler, G., & Bikowski, D. (2010). Developing collaborative autonomous learning abilities in computer
mediated language learning: attention to meaning among students in wiki space. Computer Assisted
Language Learning, 23(1), 4158.
Koschmann, T., Kelson, A. C., Feltovich, P. J., & Barrows, H. S. (1996). Computer-supported problembased learning: A principled approach to the use of computers in collaborative learning (pp.83124). In T.
Koschmann (Ed.), CSCL: Theory and practice of an emerging paradigm. Mahwah, NJ: Lawrence
35
Hui-Chin Yeh
Erlbaum.
Kowal, M., & Swain, M. (1994). Using collaborative language production tasks to promote students
language awareness. Language Awareness, 3(2), 7393.
Lee, L. (2001). Online interaction: Negotiation of meaning and strategies used among learners of Spanish.
ReCALL, 13(2), 232244.
Lee, L. (2002). Enhancing learners communication skills through synchronous electronic interaction and
task-based instruction. Foreign Language Annuals, 35(1), 1624.
Lee, L. (2004a). Perspectives of nonnative speakers of Spanish on two types of online collaborative
exchanges: Promises and challenges. In L. Lomicka & J. Cooke-Plagwitz (Eds.), Teaching with
technology (pp. 221248). Boston: Thomsen & Heinle.
Lee, L. (2004b). Learners perspectives on networked collaborative interaction with native speakers of
Spanish in the U.S. Language Learning & Technology, 8(1), 83100. Retrieved from
http://www.llt.msu.edu/vol8num1/pdf/lee.pdf
Lee, L. (2010). Exploring wiki-media collaborative writing: A case study in an elementary Spanish
course. CALICO Journal, 27(2), 260276.
Leki, I. (1993). Twenty-five years of contrastive rhetoric: Text analysis and writing pedagogies. In S.
Silberstein (Ed.), State of the art TESOL essays (pp. 350370). Virginia: Teachers of English to Speakers
of Other Languages.
Lowry, P. B., Curtis, A., & Lowry, M. R. (2004). Building a taxonomy and nomenclature of collaborative
writing to improve interdisciplinary research and practice. Journal of Business Communication, 41(1),
6699.
Masoodian, M., & Luz, S. (2001). COMAP: A content mapper for audio-mediated collaborative writing.
Usability Evaluation and Interface Design, 1, 208-212.
Nixon, R. M. (2007). Collaborative and independent writing among adult Thai EFL learners: Verbal
interactions, compositions, and attitudes. Thesis (Ph.D.) University of Toronto.
Oliva, M., & Pollastrini, Y. (1995). Internet resources and second language acquisition: An evaluation of
virtual immersion. Foreign Language Annals, 28(4), 551563.
OSullivan, D., Mulligan, D., & Dooley. L. (2007). Collaborative information system for university-based
research institutes. International Journal of Innovation and Learning, 4(3), 308322.
Patton, M. Q. (1990). Qualitative evaluation and research methods. Thousand Oaks, CA: Sage.
Polio, C. G. (1997). Measures of linguistic accuracy in second language writing research. Language
learning, 47(1), 101143.
Smith, B. (2003). Computer-mediated negotiated interaction: an expanded model. The Modern Language
Journal, 87(1), 3857.
Speck, B. W., Johnson, T. R., Dice, C. P., & Heaton, L. B. (1999). Collaborative Writing: An Annotated
Bibliography. Westport, CT: Greenwood Press.
Spelman Miller, K. (2006). The pausological study of written language production. In P. H. Sullivan and
E. Lindgren (Eds.), Computer keystroke logging and writing: Methods and applications (pp. 1139).
Amsterdam: Elsevier.
Storch, N. (1999). Are two heads better than one? Pair work and grammatical accuracy. System, 27(3),
363374 .
36
Hui-Chin Yeh
Storch, N. (2005). Collaborative writing: Product, process, and students reflections. Journal of Second
Language Writing, 14(3), 153173.
Sotillo, S. M. (2000). Discourse functions and syntactic complexity in synchronous and asynchronous
communication. Language Learning & Technology, 4(1), 82119. Retrieved from
http://llt.msu.edu/vol4num1/sotillo/default.html
Swain, M., & Lapkin, S. (1998). Interaction and second language learning: Two adolescent French
immersion students working together. The Modern Language Journal, 82(3), 320337.
Swain, M. (2000). The output hypothesis and beyond: Mediating acquisition through collaborative
dialogue. In J. P. Lantolf (Ed.), Sociocultural theory and second language learning (pp. 97114). Oxford:
Oxford University Press.
Ware, P. D. & O'Dowd, R. (2008). Peer feedback on language form in telecollaboration. Language
Learning & Technology, 12(1), 4363. Retrieved from http://llt.msu.edu/vol12num1/pdf/wareodowd.pdf
Webb, N. M. (1989). Peer interaction and learning in small groups. International Journal of Educational
Research, 13, 2139.
Weber, R. P. (1990). Basic content analysis (2nd ed.). Newbury Park, CA: Sage.
Wells, G., & Chang-Wells, G. L. (1992). Constructing knowledge together: Classrooms as centers of
inquiry and literacy. Portsmouth, NH: Heinemann.
37

http://llt.msu.edu/issues/february2014/news.pdf

pp. 3841
NEWS FROM SPONSORING ORGANIZATIONS

Sponsors
University of Hawaii National Foreign Language Resource Center (NFLRC)
Michigan State University Center for Language Education and Research (CLEAR)
University of Hawaii National Foreign Language Resource

Center (NFLRC)
The University of Hawaii National Foreign Language Resource Center engages in research and materials
development projects and conducts workshops and conferences for language professionals among its
many activities.
LANGUAGE FOR SPECIFIC PURPOSES SUMMER INSTITUTE
July 7-11, 2014 University of Hawaii at Mnoa
Language for specific purposes (LSP) courses and programs focus on developing learner communicative
competence in a particular professional or academic field (e.g., Korean for business, Japanese for health
care providers, etc.).
This summer institute will provide accepted participants with training and experience in developing LSP
courses (including doing needs analysis, setting goals and objectives, assessing, developing materials,
teaching, and evaluating LSP) for their home institution. Participants are expected to develop some aspect
of a real LSP course as part of this institute. Selected projects will be published by the NFLRC as a
Network (available online for teachers around the world). Partial travel funding is available for eligible
accepted participants. The application deadline is March 31, 2014.
STAY IN TOUCH WITH SOCIAL MEDIA
Did you know that the NFLRC has its own Facebook page with over 1,600 fans? Its one of the best
ways to hear about the latest news, publications, conferences, workshops, and resources we offer. Just
click on the Like button to become a fan. For those who prefer getting up-to-the-minute tweets, you
can follow us on our Twitter page. Finally, NFLRC has its own YouTube channel with a growing
collection of free language learning and teaching videos for your perusal. Subscribe today!
38
News from Our Sponsoring Organizations
NEW NFLRC PUBLICATIONS

The National Foreign Language Resource Center released three new titles in June:
Noticing and second language acquisition: Studies in honor of Richard
Schmidt
by Joara Martin Bergsleithner, Sylvia Nagem Frota, &
Jim Kei Yoshioka, (Eds.)
(2013)
374pp
Download/view the table of contents
This volume is a collection of selected refereed papers presented at the Association of Teachers of
Japanese Annual Spring Conference held at the University of Hawaii at Mnoa in March of 2011. It not
only covers several important topics on teaching and learning spoken and written Japanese and culture in
and beyond classroom settings, but also includes research investigating certain linguistic items from new
perspectives.
Practical Assessment Tools for College Japanese
by Kimi Kondo-Brown, James Dean Brown, & Waka Tominaga (Eds.)
(2013)
162pp
Download/view table of contents
Practical Assessment Tools for College Japanese collects 21 peer-reviewed assessment modules that were
developed by teachers of Japanese who participated in the Assessments for Japanese Language
Instruction Summer Institute at University of Hawaii at Manoa in summer 2012. Each module presents a
practical assessment idea that can be adopted or adapted for the readers own formative or summative
assessment of their Japanese language learners. For ease of use, each module is organized in
approximately the same way including background information, aims, levels, assessment times,
resources, procedures, caveats and options, references, and other appended information.
39
Check out our many other publications.

Save the trees! Check out our other two online journals:
Language Documentation & Conservation is a

refereed, open-access journal sponsored by NFLRC
and published by University of Hawaii Press.
LD&C publishes papers on all topics related to
language documentation and conservation, as well
as book reviews, hardware and software reviews,
and notes from the field.
Reading in a Foreign Language is a refereed

international journal of issues in foreign language
reading and literacy, published twice yearly on the
World Wide Web and sponsored by NFLRC and
the University of Hawaii College of Languages,
Linguistics, and Literature.
40
Michigan State University Center for Language Education

and Research (CLEAR)
CLEARs mission is to promote the teaching and learning of foreign languages in the United States.
Projects focus on materials development, professional development training, and foreign language research.
RICH INTERNET APPLICATIONS FOR LANGUAGE LEARNING
One of CLEARs most successful projects is our suite of free online language learning and teaching tools
called Rich Internet Applications (RIAs). We have been busily working on retooling the user interface for
the RIAs and look forward to launching the new website this fall. New features will include improved
documentation of all the RIAs, screen flow videos and demonstrations of each app, a gallery of sample
RIA activities created by language teachers, and printable start-up guides for both teachers and students.
CONFERENCES
CLEAR exhibits at local and national conferences year-round. We enjoyed seeing many of you in
Hawaii at the CALICO conference hosted by our sister center, the University of Hawaii National
Foreign Language Resource Center. We look forward to seeing you at MIWLA and ACTFL later this
year, and the Central States Conference in the spring.
PROFESSIONAL DEVELOPMENT
CLEAR hosted four well-attended professional development workshops in July and August 2013.
Covering topics including writing in the foreign language classroom, the teaching of vocabulary, online
tools for language learning, and assessing speaking skills, the workshops offered hands-on experience and
lots of ideas for language educators. Watch our Web site in October for the announcement of topics and
dates for summer 2014.
NEWSLETTER
CLEAR News is a free bi-yearly publication covering FL teaching techniques, research, and materials.
Download PDFs of back issues and subscribe at http://clear.msu.edu/clear/newsletter/.
41

http://llt.msu.edu/issues/february2014/review1.pdf

pp. 4245
REVIEW OF RESEARCHING ONLINE FOREIGN LANGUAGE

INTERACTION AND EXCHANGE: THEORIES, METHODS, AND
CHALLENGES
Researching Online Foreign Language Interaction and
Exchange: Theories, Methods and Challenges
Melinda Dooly & Robert ODowd (Eds.)
2012
ISBN: 978-3034311557
US
347 pp.
Peter Lang
New York, US
Review by Linda Bradley, Chalmers University of Technology

Researching Online Foreign Language Interaction and Exchange, edited by Melinda Dooly and Robert
ODowd, is the third book in a series by Peter Lang on telecollaboration in education. This third edited
volume maps out topical aspects of computer-mediated communication (CMC) in online exchange
research in three sections: theoretical approaches, key areas of research, and research methods. The three
sections consist of chapters authored by specialists in each field. The book provides overviews of topical
areas of interest for telecollaboration research, and it serves as valuable input for anyone interested in
research in online foreign language interaction and exchange environments.
In this time, when online interaction and exchange in foreign language (FL) education has increased,
researchers are faced with a range of decisions to make regarding research frameworks, methods, task
designs, and ethical issues, as well as with conceptual questions related to defining constructs such as
intercultural competence and multimodality. This volume provides insights into prevailing theoretical
approaches, together with key areas of research and many of the methods used to investigate FL
interaction and exchange in online environments. This area is relatively new within the field of CALL due
to what Dooly and ODowd describe as a combination of various developments. These developments are
related to technical aspects, such as the increase in user-friendly technologies in educational contexts,
cheaper computer hardware, and widespread access to internet connections, but other reasons also factor
into recent growth. They outline the first reason as the importance of acknowledging culture in FL
learning, particularly in cases in which online intercultural interaction plays an important role. A second
reason is the growth of the sociocultural processes whereby learning is understood as the meaningful use
of language in interaction. The third reason they attribute to both the growing importance of online
technologies in shaping the ways we work and learn in global networks, and to how FL competence and
e-literacies have emerged as components within the new set of skills required of individuals in response to
changing labor markets.
42
Linda Bradley
Review of Researching Online Foreign Language Interaction and Exchange
The book provides a comprehensive historical overview of the research to date on online language
learning. Dooly and ODowd describe this history as first appearing as collections of accounts of
classroom practice and anecdotal research, but then moving to more in-depth empirical studies. They
describe three general categories of online interaction and exchange for the purpose of FL learning. The
first category is in-class interaction, which refers to online networks for students to interact in one
particular class through synchronous communication in chats, MOOs and local area networks. Many of
these in-class interaction studies focus on the interactionist perspective to FL education and
psycholinguistic theories of SLA. The second category, class-to-class interaction, often termed
telecollaboration, came with improved online connectivity. Both asynchronous communication tools
evolved such as discussion forums, blogs, wikis, synchronous oral communication, and multimodal
technologies. With these studies, the sociocultural perspective of learning plays a predominant theoretical
framing role. The third category of class-to-world interaction describes contexts in which learners enter
into contact with others globally, but not in communication organized formally by the instructor. Rather,
learners collaborate in specialized interest communities or environments outside of the classroom context,
which opens up a blurring of boundaries of the traditional classroom with other communicative
environments.
Section I. Theoretical Approaches to Researching Online Exchange
In the first of three chapters in this section, Jonathon Reinhardt gives an overview over the function of the
interactionist approach and socio-cognitive perspectives. These two frameworks are examined by
focusing on methodological approaches and the areas of overlap they share, including the concept of
interaction and negotiation of meaning. He offers examples of key findings from each approach to
illustrate the underlying concepts and to discuss critiques of these approaches. As a response to some of
the issues in this debate, Reinhardt examines the ecological approach that can offer new insights into
SLA. He elaborates on the discussions of acquisition and learning within applied linguistics and foreign
language acquisition and development. He emphasizes that interaction must be understood broadly since
negotiation of meaning is not bound to an input-interactionist framework. One way forward is adopting an
ecological approach for understanding learning. Such an approach transcends our most recent
conceptualizations of technology from originating in mainframes and social processing units but also in
the individual use of personal computers as among the many tools available in todays distributed,
networked, and ubiquitous gadgets.
In the next chapter, Introducing Cultural Historical Activity Theory (CHAT) for Research CMC in
Foreign Language Education, Franoise Blin gives an overview of CHAT and discusses the basic
concepts and principles by drawing on examples from studies with a connection to CMC. Activity theory
aims to understand human beings in their everyday life circumstances, seeing individuals as contextbound. The chapter brings up activity theory, built on interconnected relationships in the form of
triangles that display the relationships between students (subjects), the object that they are engaged in,
and the larger context of community. All connections are made by tools and artifacts, rules and division
of labour. Blin suggests that we have reached the third generation of activity theory. In this chapter,
activity systems are illustrated with examples of how telecollaboration transforms activity systems and
provides ethnographic data.
In the final chapter of the first section, Paige Ware and Brenna Rivas investigate mixed methods
approaches to analyzing data. They suggest a combination of qualitative and quantitative data approaches
to investigating learning and show that much of the research on telecollaboration has been examined with
a qualitative lens that, appropriately, relies on close contextualization of such projects. They suggest that
mixed methods might offer a unique way to examine online exchanges in the particular context of
secondary education where different types of institutional constraints are in place. They illustrate the
unique challenges of the secondary context in the final part of their chapter through a case study of a
43
Linda Bradley
telecollaboration between adolescent learners in Spain and the US.

Section II. Key Areas of Research: Tasks, Culture, Multimodality, and Virtual Worlds
In the first chapter out of four in this section, Melinda Dooly and Mirjam Hauck emphasize the need to
embrace multimodal communicative competence (MCC) in telecollaborative language learning research.
The central question for researchers is deciding what may be considered researchable in multimodal data.
For example, they show how exchanges can rely on a variety of constellations of asynchronous and
synchronous online tools. Interestingly, in analyzing such data, they argue that written and spoken
language are still the two most prominent areas of interest, whereas modalities such as intonation and
gestures are more often described in relation to spoken and written language and not as separate meaning
making and communication modes in their own right. According to Dooly and Hauck, exploring the
MCC field is challenging but promising for future researchers willing to explore expanding parameters
of communication that is opening up, exponentially, with each new generation of language learners (p.
154). The chapter discusses the complexity involved in multimodal data collection and analysis and the
challenges of achieving transparency in the research process.
In the second chapter of this section, The Classroom-Based Action Research Paradigm in
Telecollaboration, Andreas Mller-Hartman gives an account of action research and activity theory
connected to telecollaboration. His focus is on teachers competence development within telecollaborative
projects, such as their development of Intercultural Communicative Competence (ICC), multimodal
competence, and teaching competence. Since telecollaborative studies are mostly conducted by
researchers who are also practitioners, the case study approach is suggested as a frame for the research
context. In such situations, instructors do not come to the contexts as outsiders. Such approaches allow
studies to consist of complex interrelations of environments and their agents. For the case study approach,
he argues that activity theory can help researchers to understand the complex processes taking place and
to gain the perspectives of the participants. He illustrates three different forms of competence:
multiliteracy competences, task-based language teaching competences (TBLT), and intercultural
communicative competences (ICC) by drawing on a sample case study.
The next chapter, by Luisa Panichi and Mats Deutschmann, gives a balanced review of learning in virtual
worlds. They argue that virtual worlds can be used in telecollaborative activities as environments for
explorations because the synchronous aspects of virtual worlds resemble face-to-face meetings. This realtime effect also affects such aspects as designing and monitoring tasks. They outline the affordances for
communication offered by virtual environments and tackle one of the main challenges of
telecollaboration, that of designing meaningful activities. They show how it is possible to build virtual
environments and to use avatars to communicate through non-verbal cues. They discuss the complexity of
this new type of environment with a focus on the emergence of new ethical and copyright issues.
In the final chapter of this section, Intercultural Competence in Computer-Mediated-Communication: An
Analysis of Research Methods, Martina Mllering and Mike Levy offer insight specifically into the
intercultural side of online communication. They provide an overview of what the intercultural turn has
meant for research and outline themes that emerge from selected studies that use sociocultural theory to
frame ICC. They explore underlying constructs of culture, including dimensions such as elemental, group
membership, contested, and individual. They claim that research in ICC is multifaceted because the
concept of culture itself is quite complex.
Section III. Research Methods for Online Interaction and Exchange
In the first chapter out of the two in this last section describing methodological challenges and potentials,
Nina Vyatkina shows how corpus analysis methods and tools can be used when examining
telecollaborative discourse in her chapter: Applying the Methodology of Learner Corpus Analysis of
44
Linda Bradley
Telecollaborative Discourse. Using CMC learner corpora in telecollaborative studies implies accessing
data that is automatically saved and thus immediately available for research. In addition, native-speaker
and non-native speaker contrasts are built into the corpus for further elaboration. The researcher can
thereby easily get an insider view on the study context, which enriches the ethnographic dimension. This
added information enhances the ecological validity of the learner corpus as well as the findings based on
its data. Vyatkina also emphasizes the advantage of telecollaboration corpora in offering a wide array of
discourse types and linguistic features. Her chapter highlights an example of the application of the
Telekorp corpus that is based on exchanged emails and chats, and she explores the use of German modal
particles as used by both learners and by native speakers of German. Her conclusion discusses analytical
tools for telecollaboration corpus research design, including both proprietary corpus software as well as
open source corpus software.
In the final chapter of the book, Using Eye-Tracking to Investigate Gaze Behaviour in Synchronous
Computer-Mediated Communication for Language Learning, Breffni ORourke brings up the relatively
new research area in the application of eye-tracking to synchronous, text-based computer-mediated
communication for language learning. In this chapter, three different analyses of eye-tracking devices are
reported in which native speakers of English were learning a foreign language by interacting in a textbased virtual environment. As eye-tracking tools become more accessible and easier to use, he argues that
it is likely that they will be used more for telecollaboration research.
In Researching Online Foreign Language Interaction and Exchange, Dooly and ODowd present some of
the most topical theoretical and methodological research trends within telecollaboration in education. This
book contributes with its insights into applications of approaches for both initiated researchers but also for
novices in the field of CALL. It maps out existing research as well as suggestions of where the field is
heading. The interaction and exchange aspects of online studies are increasing since mobility is spreading.
Although there is no specific chapter attributed to mobile learning in the book, this topic comes up in
some of the chapters. The two previous books in the series have focused on more specific areas of
telecollaboration: Guth and Helm discussing the concept of telecollaboration 2.0 and Sadler examining
virtual worlds. Dooly and ODowds book offers a comprehensive overview of contemporary notions of
the most recent practices in telecollaboration. In sum, this book attracts anyone interested in research in
online foreign language learning interactions and exchanges. It offers valuable input regarding current
aspects of CMC and telecollaboration.
ABOUT THE REVIEWER

Linda Bradley recently finished her PhD within the area of web-based technology and learning at the
University of Gothenburg in Sweden. Her research interests include investigating student collaboration,
communication and intercultural learning in digital environments in language learning and specifically
within English for Specific Purposes (ESP) in higher education.
E-mail: linda.bradley@chalmers.se
45


pp. 4648
REVIEW OF LANGUAGE LEARNING WITH TECHNOLOGY: IDEAS FOR

INTEGRATING TECHNOLOGY IN THE CLASSROOM
Language Learning with Technology:
Ideas for Integrating Technology in the
Classroom
Graham Stanley
2013
ISBN: 9781107628809
215 pp.
Cambridge University Press
Cambridge, UK
Review by Nancy Montgomery, Southern Methodist University

Language Learning with Technology: Ideas for Integrating Technology in the Classroom by Graham
Stanley is a handbook designed for both new and veteran teachers who wish to improve their knowledge
of how to integrate technology in the curriculum. One of the many strengths of this handbook is its
centralizing of learning goals, rather than of specific technologies. Each chapter has a specific learning
goal around which the discussion of technologies takes place. This useful organizational strategy
emphasizes a sequence of presentation that begins with the learning focus, moves through to lesson
preparation, and finally suggests technical requirements and pedagogical possibilities. The overall effect
of the handbook thereby provides useful and interesting activities in which technology provides added
value to language learning activities across all language levels that are outlined by goals, level, and time
estimates.
Chapter 1 outlines activities that help students become acquainted with the use of technology in the
classroom. Each activity includes an explanation of the exact procedure teachers can follow to create
successful exercises for their students, including ideas for variations of each activity to meet the needs of
learners across a variety of language levels. Examples from this chapter include several suggestions for
simple and efficient diagnostic tools that teachers can use to understand their students attitudes about
using technology in the classroom. For example, the activities of Technological Survey and Favorite
Website both elicit information about general attitudes and Internet use patterns that can provide teachers
with class profiles of student background interests with technology.
Chapter 2 focuses on activities that help foster a classroom learning community. Stanley anchors his view
of community as "a group of people with shared values, a common purpose, and similar goals" (p.25).
Since language is constructed within such social contexts, the building of a community in language
46
Nancy Montgomery
Review of Language Learning with Technology
learning classrooms is extremely important. Technology offers the additional advantage of extending the
classroom community even beyond the physical classroom. Through the use of various communication
tools such as blogs, email, and threaded forum discussions, the results can become a community of
practice (Lave & Wagner, 1991, p. 98) that occurs when people with a common interest regularly interact
to learn with each other over an extended period of time. There are several community-building
strategies in chapter 2, many of which are geared for a wide range of language levels, and they involve a
range of technologies including blogs, wikis, Facebook, and Twitter that can be used to support
independent learning and also keep a social presence after the class has ended. Also of importance in this
chapter are activities that explicitly focus on online safety.
The next seven chapters are organized in a conventional way familiar to language teachers, as each
chapter is devoted to a particular language skill area. Chapters 3 and 4 focus on vocabulary and grammar,
respectively. For vocabulary, Stanley emphasizes extending word fields by exposure to and participation
in rich contexts, and he offers ideas for fostering learner autonomy. The activities in this chapter are
arranged according to the language level of the students, from simple word games for the beginning
language learner to more advanced techniques for advanced learners that draw on a range of technologies.
The focus of Chapter 4 is on grammar and how technology can support teachers in helping their students
gain a greater understanding of underlying language structures. The activities are organized according to
the different levels of language, and most are relatively brief, taking between 20 and 30 minutes. Many of
the activities focus on authentic grammar in use, rather than on contrived grammatical constructions, as
illustrated by an activity called Authentic Word Clouds, which is designed to increase the awareness of
authentic written text that contains complex sentences. One weakness of this chapter is the paucity of
activities for the beginner level, although skilled teachers can modify many of those aimed at higher
language levels.
The role of technology, as Stanley explains in Chapter 5, is to help students achieve better listening skills
through activities that teach them various aspects, including discerning the gist of text, listening for
details, and developing inference skills. Many activities in this chapter make use of authentic listening
materials available on the web that expand the types of registers and accents beyond the packaged voices
of curricular materials. Many ideas in this chapter also rely on basic technology tools, which allow
teachers easy integration without having to invest too much time learning new tools.
Chapters 6 and 7 are dedicated to developing stronger reading and writing skills, and Stanleys view of
reading involves being proficient in a range of information and communication literacies. Some of the
sub skills targeted are skimming, scanning, reading for the gist, activating schema, summarizing, and
developing writing fluency. Many activities integrate reading and writing, such as Word-Cloud Warmer,
which taps into prior knowledge about a topic. Students can make predictions about what is in the text
and share through reconstruction of the text what they have learned with another student through a blog or
email to bring together the reciprocal skills of reading and writing. Although the chapters are named
specifically reading and writing, the author gives activities that involve the use of both skills.
Technology has offered several tools to support speaking, from early computer-mediated communication
(CMC) to the more current use of Skype and voice recognition software that allow students to have
authentic speaking experiences when collaborating with other students and when practicing their
speaking. Learners can record themselves speaking and develop more autonomy in determining areas in
need of improvements. Activities for speaking in this chapter focus on fluency, accuracy, pronunciation,
and learner autonomy and include technology tools such as voice recorders, audio websites, audio
journals, and mobile learning tools. The chapter includes activities that encourage speaking in and out of
class, scripted and unscripted dialogue, as well as interactive speaking. The activities are easy to modify
across language levels and offer students the chance to take speaking out of the classroom into real-world
experiences through cell phones, audio websites, computer-based simulated environments, and audiovoice forums.
47
Nancy Montgomery
Review of Language Learning with Technology
The activities in Chapter 9 center on raising awareness of phonetics, common pronunciation errors,
connected speech features, minimal errors, stress, and intonation. Some of the activities in this chapter
help students understand how they can use recorders, head phones, podcasting, and mobile phones to
extend pronunciation practice outside the classroom. An example of an interesting activity is Rerecording Speeches and Scenes, an activity that raises awareness of the intonation of connected speech
and gives students opportunities to record their speech to see how well their pronunciation and intonation
match those of the speakers pre-recording. Many of the activities in this chapter leverage the use of
integrating more than one technology, such as the use of podcasts with wikis in the development of a
class vocabulary audio notebook.
One of the most powerful chapters of this book is Chapter 10, Project Work, because it offers
collaborative activities and illustrates many specific ways to use technology for group work. Learners
become active participants in doing meaningful tasks within authentic settings. In this chapter, Stanley
focuses on collaboration through activities such as the creation of e-books, online magazines, recipe
books, and filmmaking. These activities can motivate students to become more aware of how they are
developing integrated skills. Students can conduct these projects in and out of class both on their own and
in collaboration with others. The skill levels in this chapter span all levels of proficiency.
In Chapter 11, Assessment and Evaluation, many activities help the teacher using technology to diagnose
and evaluate what the student has learned and where they are in the learning process. The activity of EPortfolio Archive and Showcase, for example, is focused on formative assessment and is for all levels of
learners. It helps the learner understand how portfolios work and gives them ownership for determining
their best work to showcase. Comparing Placement Tests is another assessment activity that encourages
learners to reflect on their development. Through web-based organizational archives, including blogs and
course management systems, teachers and students can collect documents or artifacts that are text-based
and multimodal. In this way, through technology integration teachers are shown a wide range of options
for both formative as well as summative assessments.
This book is an excellent tool for any teacher who wants to integrate technology in the curriculum to
support student learning. The activities place pedagogy firmly over technology. The appendices also
provide explicit definitions of the technologies in the book and suggested software tools. Teachers do not
have to be technology gurus in order to use this book effectively and efficiently in a classroom, and as
such, the book is particularly useful to those who are new to technology integration.
ABOUT THE REVIEWER

Dr. Nancy Montgomery is an Assistant Clinical Professor at Southern Methodist University and has
taught all ages of English language learners, from elementary students to adults, in public and private
institutions in the United States and in Indonesia. She has also served as an administrator and has
presented at many state, national, and international conferences on the topic of supporting refugee
learners from Southeast Asia and Africa.
E-mail: nmontgomery@smu.edu
REFERENCE
Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge, UK:
Cambridge University Press.
48


pp. 4952
REVIEW OF ENCOUNTERS: CHINESE LANGUAGE AND CULTURE

Encounters: Chinese Language and Culture (Student Book 1)
Encounters: Chinese Language and Culture (Character
Writing Workbook 1)
Cynthia Y. Ning; John S. Montanaro
2012
ISBN: 978-0-300-16162-5
978-0-300-16170-0
US $94.99; US $29.99
336 pp.; 256 pp.
Yale University Press; China International Publishing Group
New Haven and London
Review by Yaqiong Cui, Michigan State University

In the past two decades, great value has been placed on communicative and cultural competency in
language classrooms. This has led to a burgeoning of textbooks focusing on learners communicative
skills and their ability to use the language in the target culture. However, Chinese textbooks often lack
key communicative criteria and often do not require learners to be able to use Chinese in authentic, reallife settings. Some Chinese textbooks on the market, such as Chinese Link (Wu, Yu & Zhang, 2007) and
Integrated Chinese (Liu, 2008), seem to place great emphasis on learners reading and writing skills and
introduce a large amount of vocabulary and grammar rules in each chapter. Few activities are devoted to
learners communicative skills, and instructions are sometimes unclear, which require teachers to invest
much time preparing for class. Culture might only be presented as facts within limited space at the end of
each chapter. Thus, teachers of Chinese might be intrigued by Encounters: Chinese Language and
Culture, because it is promoted as a truly communicative and task-based textbook. I am glad to report that
it is highly communicative, has authentic materials, and most importantly, is, as it is promoted, culturally
rich and delightfully engaging (p. xvii). However, teachers who use it will still need to supplement parts
of the book to enhance its potential to promote task-based learning.
Integrated, Authentic, Practical, and Engaging
The Encounters program provides an integrated series of learning materials: student books, character
writing workbooks, companion website, CD-ROMs (dramatic video episodes, instructional videos on
Chinese culture, information about class testing and sample exams, etc.), and annotated instructor editions.
One of the most distinctive features of the Encounters program is that it includes a dramatic storyline that
was filmed in six different locations across China with a cast of nine characters from different areas and
with different cultural and social backgrounds. These characters discover themselves and others as they
explore the language and culture of China. This is very similar to Sol Viente (VanPatten, Lesser &
Keating, 2011), a Spanish textbook, in that it also provides a video series particularly made for the
textbook. This genre of textbook has been gaining popularity in language education in the United States.
49
Yaqiong Cui
Review of Encounters: Chinese Language and Culture
An additional bonus is a series of video culture notes. For example, people on the street talk about their
views on topics such as different forms of greetings in Chinese and different methods of bargaining in
street markets. For the culture notes, the videographers recorded a mix of non-actors (passers-by), actors
tasked with certain communicative speech acts (these appear to be mainly unscripted), and teachers, who
comment on the speech acts and explain them through a Chinese cultural lens. Learners thereby become
engaged in both urban and rural settings in China. Also importantly, through watching these videos,
learners get a chance to hear a range of native speakers accents.
The title of each unit is written in English, Pinyin, and Chinese characters. Interestingly, unlike the other
textbooks in which the titles are simply Lesson X, specific expressions that are related to the topics in
the unit are presented as the titles. These are often four-character Chinese idioms or widely used proverbs.
For example, Unit Five concerns family and friends, with title (qnpnghoyu, family and
friends); Unit Nine deals with shopping and bargaining; the title is, then, (y fn
jiqin y fn hu, you get what you pay for). In addition, on each units introductory page, students can
garner a clear idea of what they will learn in the unit from a list of skills and topics that will be covered in
each lesson. Introducing a unit in this way is not only appealing to learners but also culturally rich and
authentic. It corresponds with the authors idea that students should be immersed in the target language
and culture immediately, from the start of Chinese learning. When overviewing the complete table of
contents, I sensed that this textbook might be task-based, in that the chapter topics are based on what
students will need to know to communicate in Chinese.
The units in the textbook contain supplementary or additional information through three types of colored
boxes: Grammar Bits (blue), FYI (green), and Cultural Bits (red). Vocabulary, grammar, culture,
and other information are, in this way, not presented as isolated sections as seen in many other textbooks.
Rather, they are interspersed throughout the units, interwoven with the storyline, and embedded in the
audio and video materials. Grammar Bits boxes explicitly instruct the grammar needed to convey
meaning related to the unit topics and activities. The authors created these grammar lessons as
incidental; however, one might contextualize them better as reflecting a planned focus on form, as the
grammar forms highlighted anticipate the language needed to complete the communicative tasks at the
heart of the lessons. The grammar explanations in these boxes are presented in English, which might
negate the need for teachers to focus precious in-class time explaining sometimes complicated grammar
points, which further allows teachers to only talk in Chinese during class. FYI boxes provide factual
information (for example, that China has only one time zone) and study tips. The Cultural Bits boxes
often have questions for discussion related to the cultural aspects of the video and only provide, as the
authors note, a jumping-off point (p. xxvii) for investigations into Chinese society and culture. In other
words, teachers have plenty of room to supplement the instruction with their own culturally-focused tasks
here. But the authors should be commended for not leaving culture to the very end of each unit and for
not simply listing cultural facts. With this textbook the Chinese teaching field is closer to the notion that
culture is inseparable from language. It helps students explore how Chinese culture is different from or
similar to their own by providing thought-provoking questions.
With a great emphasis on practicality, lessons in Encounters deal with up-to-date topics. For instance,
when studying Unit Six, in which professions and careers are discussed, students learn about the changes
taking place in workplaces in modern China, which leads to a further discussion of the relationship
between education and careers in contemporary Chinese society. Another example is Unit Nine, which
deals with shopping and bargaining. Different from other textbooks in which students only learn about
some expressions used when shopping, this unit embeds language learning within the art of bargaining
and offers useful tips for bargaining in China. Another enjoyable aspect about Encounters is that
numerous authentic texts are presented throughout. By incorporating authentic materialsbusiness cards,
advertisements, newspapers, signs, and hand-written notesthe lessons provide practical information that
enables students to better understand Chinese culture.
50
Yaqiong Cui
Unlike some other Chinese textbooks, which can be monotonous in design, Encounters features a colorful
layout. It has authentic photographs, attractive illustrations, and organizing icons. These bells and
whistles likely keep learners motivated because they are visually pleasing. To encourage students to learn
Chinese, entertaining rap songs are presented to review the core vocabulary and expressions in each unit.
A Recap (wrap-up) section appears at the end of each unit. Each of these sections includes a summary
of grammar points, a list of vocabulary, and a checklist of can-do skills that students should have
mastered after learning the unit. Those skills span listening, speaking, reading, writing, and particularly,
understanding culture. The Recap, which corresponds to the preview section on the introductory page
of that unit, also helps students monitor their progress, identify gaps in their learning, and appreciate their
accomplishments. All of these features may help deepen learners understanding of the language, the
culture, and the people of China, and also make the learning process more organized and enjoyable. This
is particularly important to help build self-regulating and autonomous learners, ones who will be more
likely to continue to learn Chinese to the advanced level.
A Variety of Activities
Another characteristic of Encounters is the provision of a wide range of activities. Each unit is comprised
of several Encounters in which real-life topics are presented, and activities are designed in such a way
that students start from more form-focused practices and then progress to more meaning-based
communicative exercises. For example, in Unit Four in which nationalities are discussed, the first
encounter is expressing nationality. To begin, students are introduced to vocabulary concerning
nationalities through various form-focused exercises. Once they become familiar with the expressions,
they start doing communicative activities. For example, students are asked to take notes on short
conversations with classmates about their home cultures. The next encounter focuses on talking about
places students have lived, in which students, again, learn vocabulary and expressions through exercises,
and then move to a mingling communicative activity, in which they are instructed to ask several
classmates where they were born, where they grew up and where they live now. The following encounters
follow a similar pattern. The last encounter of each unit is always Reading and Writing, in which the
focus shifts to reading and writing skills so that students can develop a range of communication abilities.
However, within this variety, I found the various activities to be somewhat unrelated at times. For
example, in Unit Four, students first read various window signs from China and a postcard from Taiwan,
all of which they then translate into English, and then they are instructed to write an introductory note to a
new friend, explaining a bit about oneself. It is perhaps a shame that the various activities are not
integrated more, with, for example, the post-card writing example being extended into a postcard writing
task. The content of the postcards could have also been related to the text presented in the window signs.
These missed opportunities are unfortunate and would be worth revisiting in a subsequent edition. While
the speaking and writing activities often provide a nice shell for the communicative activities, teachers
will need to thoughtfully consider how to make the activities more meaningful and more developed into
complex tasks with outcomes that can be assessed. The annotated instructors edition provides useful tips
and suggestions on how to use the materials. Teachers may want to take those tips a step further and
create links between the activities in order to design more robust and meaningful speaking and writing
task.
Use of Multimedia and Technology
An outstanding feature of Encounters is that the program has a companion website which offers an array
of materials and activities, including interactive exercises, streaming video and audio content, and other
resources for practicing speaking, listening, reading, and writing in Chinese. Online media enhances
learning and teaching by providing a powerful yet intuitive tool to engage language learners and
instructors (Cairncross & Mannion, 2001; Evans & Gibbons, 2007). The use of music also provides
51
Yaqiong Cui
students a lighthearted and friendly environment to engage in Chinese learning.

Other Notes
What I also like about the Encounters program is that the character writing workbook presents the stroke
order of each character and contextualizes each character to help learners understanding. Strategies for
remembering the characters are also suggested. In addition, the workbook shows the evolution of
characters from ancient to modern Chinese in both traditional and simplified forms. Indeed, the inclusion
of both traditional and simplified characters may help learners who have different learning purposes;
however, for beginning learners with no previous experience with the Chinese writing system, it might be
confusing. Also, unlike many other textbooks in which the exercises are written solely in characters,
Encounters provides both Pinyin and character forms, with instructions in English. This may help
beginning learners better recognize Chinese characters; however, it also involves the potential problem
that learners may overly rely on Pinyin as they go through the units.
Despite those minor flaws, Encounters provides learners with meaningful, authentic, and engaging
contexts. More importantly, the use of multimedia and technology helps learners explore China and
Chinese culture. As the authors claim, Encounters masterfully guides learners along a well-prepared path
toward intercultural communication and understanding, a path that also leads to fuller participation in the
modern global community (p. xvii). There is no doubt that this textbook can bring significant
contributions to the field of Chinese language teaching and inspire new innovations in the development of
Chinese teaching materials.
ABOUT THE REVIEWER

Yaqiong Cui is a doctoral student and research assistant in the Second Language Studies Ph.D. Program
at Michigan State University. Her primary research interest is second language acquisition and the
processing of Chinese from a psycholinguistic perspective. She has taught Chinese for three years in the
Department of East Asian Languages and Cultures at the University of Illinois at Urbana-Champaign. She
is now working as a language facilitator at the Center for Language Teaching Advancement at MSU.
E-mail: cuiyaqio@msu.edu
REFERENCES
Cairncross, S., & Mannion, M. (2001). Interactive multimedia and learning: Realizing the benefits.
Innovations in Education and Teaching, 38(2), 156164.
Evans, C., & Gibbons, N. J. (2007). The interactivity effect in multimedia learning. Computer &
Education, 49(4), 11471160.
Liu, Y. (2008). Integrated Chinese (3rd ed.). Boston, MA: Cheng & Tsui.
VanPatten, B., Lesser, M. J., & Keating, G. D. (2011). Sol Viente (3rd ed.). Columbus, OH: McGraw-Hill.
Wu, S., Yu, Y., & Zhang, Y. (2007). Chinese link: Zhongwen Tiandi. Upper Saddle River, NJ: Prentice
Hall.
52


pp. 5356
REVIEW OF TECHNOLOGY IN INTERLANGUAGE PRAGMATICS

RESEARCH AND TEACHING
Technology in Interlanguage Pragmatics Research and
Teaching
Edited by Naoko Taguchi & Julie M. Sykes
2013
ISBN: 978-90-272-13136
US $143.00 (hardcover), $54.00 (paperback)
276 pp.
John Benjamins (North America)
Philadelphia, PA
Review by Feng Xiao, Carnegie Mellon University

This book is a collective endeavor of exploring the role of technology in interlanguage pragmatics (ILP)
research, instruction, and assessment. It starts with an introduction (Chapter 1) and ends with a
commentary and a prologue, both of which envision future directions of using technology in the field of
ILP. The included studies are divided into two parts: Part I (Chapters 2 to 6) which addresses issues in
ILP research, and Part II (Chapters 7 to 10) which addresses issues in ILP instruction and assessment.
In Chapter 1, Naoko Taguchi and Julie Sykes (both editors) first give a brief review of the field of ILP
and then highlight four major ways that technology has advanced ILP studies: (a) expanding the construct
of pragmatic competence (the processing speed); (b) digitizing learners online performance for analysis
(e.g., speed rates, pauses); (c) creating digital spaces where pragmatic functions are frequently performed;
and (d) quantifying and visualizing textual data (e.g., learners writing corpora) through concordancing
programs and grammatical tagging. In addition, the authors provide brief summaries of studies included
in this book.
In Chapter 2, Naoko Taguchi synthesizes the findings of her previous studies on comprehending
implicatures in second language (L2) English. Using a computerized pragmatic listening task, she
measures L2 learners pragmatic comprehension at two different levels: accurate understanding of
implied intentions and the speed of processing pragmatic information (measured by response time).
Findings showed that when conventionality increased, learners increased their comprehension speed and
even gained more of it over time compared with their comprehension accuracy, Taguchi attributes to
conventional implicates relying on fixed linguistic forms to deliver implied intentions. Other findings
showed that learners in the target language country (taking ESL classes) had gained more in
comprehension speed than in accuracy, while their counterparts in the domestic instructional context
(EFL), on the other hand, experienced the opposite and gained more accuracy.
In the following chapter, Shuai Li investigates the effects of different types of practice on American
learners levels of accuracy and speed in their recognition and production of requests in L2 Chinese.
During four consecutive days, the two experimental groups (the input-based and output-based groups)
53
Feng Xiao
Review of Technology in Interlanguage Pragmatics Research and Teaching
received input-based and output-based practices separately after receiving the same metapragmatic
instruction on target request forms, while the control group received instruction and practice in L2
Chinese reading comprehension. The data of the computerized pragmatic recognition and production
tasks showed that the two experimental groups outperformed the control group at both levels of accuracy
and speed in their pragmatic recognition and production. The magnitude of gain at the level of speed,
however, was smaller than at the level of accuracy, regardless of the types of practice.
In Chapter 4, Julie Sykes investigates how multiuser virtual environments (MUVEs) affect the production
of apologies in L2 Spanish. Learners were engaged in Croquelandia, the first MUVE developed for
learning requests and apologies in L2 Spanish. In Croquelandia, learners interacted with non-player
characters (NPCs) by choosing appropriate utterances according to situations. The data of pre- and postdiscourse completion tasks (DCTs) showed a moderate shift from using speaker-oriented to heareroriented apology strategies. In addition, pre-and post-surveys and interviews showed a self-perceived
increase in using appropriate apology strategies. These findings suggest a positive role of MUVEs in
pragmatic instruction. In the next chapter, Adrienne Gonzales examines an L2 Spanish learners use of
conversation closings in a text-based synchronous computer-mediated communication (CMC). The data
consisted of learners text-based conversations with native speakers in a digital space (i.e., Livemocha).
Conversation analysis revealed a tendency to use closing sequences as rapport management strategies,
suggesting that digital spaces such as Livemocha can mediate L2 pragmatic development. In addition, the
interview data showed a positive orientation towards using digital spaces to learn pragmatics.
In the last chapter of Part I (Chapter 6), Alfredo Urza uses automated corpus-based techniques (e.g.,
concordancing programs and grammatical tagging) to investigate the use of subjective pronouns and
possessives (self-positioning strategies) in L2 English writing. The corpus consisted of essays written by
two groups of ESL learners who took sequential writing courses over two semesters. Longitudinal
analysis of the corpus showed that as time passed, learners tended to avoid using you and I, but favored
we in terms of normed frequency. In addition, the essays from four randomly selected learners in each
group were analyzed separately, showing differences in choosing self-positioning devices between the
two groups. The study showed the advantage of using corpus-based techniques to seek patterns of using
different linguistic devices in order to convey pragmatic information.
In Part II, four pedagogically-oriented studies focused on technology-based feedback (Chapter 7),
assessment (Chapter 8), learning (Chapter 9) and analysis (Chapter 10) of pragmatic features. In Chapter
7, Christopher Holden and Julie Sykes investigate four types of feedback on L2 pragmatics via Mentira, a
place-based mobile game of learning Spanish. In Mentira, learners interacted with NPCs by choosing
appropriate utterances to seek information in order to profess their innocence over a murder in a local
Spanish community. As a part of a curriculum of a college-level Spanish class, Mentira required learners
to visit the real local community where the game was situated in order to find clues for game playing. The
authors found that in the three iterations of Mentira, learners received implicit feedback on L2 pragmatics
via interactions with NPCs, and what they learned in the game was enhanced through environmental
feedback (communicating with people in the real local community). Peer and instructor feedback,
however, was lacking in terms of pragmatic information. In Chapter 8, Yumi Takamiya and Noriko
Ishihara examine the role of blogging in improving pragmatic awareness and production. The study
documented how blogging mediated learning of refusals in L2 Japanese from a sociocultural perspective.
Three learners wrote blogs reflecting on taught speech acts. In addition, they were asked to add a short
open-ended questionnaire consisting of DCTs to collect speech act data from their native speaker
partners. The data also included their background surveys and audio-recordings of all classes, individual
meetings and course evaluations. One case was extensively discussed to show that asynchronous
interaction via posting and responding to blog entries can gradually increase L2 learners pragmatic
awareness and in turn, facilitate their pragmatic production (e.g., refusals in this study).
In Chapter 9, Carsten Roever discusses issues on practicality and reliability of a computer-based test of
54
Feng Xiao
L2 pragmatics in English. Roevers test includes three sections: a multiple-choice section of

comprehending implicatures, (N=12) a multiple-choice section of situation-bound routines, (N=12) and a
brief DCT section for requests, apologies, and refusals. The data of 335 test takers were subject to
statistical analysis. Findings revealed that degrees of computer familiarity did not have a significant effect
on scores, and the overall use of vocabulary aids decreased with increasing proficiency. Moreover, the
computer-based test showed strengths in practicality and improving reliability because of the instant
digitization of score information.
In the last Chapter, Helen Zhao and David Kaufer introduce the potential of DocuScope to facilitate L2
pragmatic analysis. DocuScope is text-visualization and genre analysis software. The prescribed codes
used by this software can analyze texts according to the appropriateness of pragmatic functional clusters
of each genre. The study used DocuScope to analyze Chinese EFL learners written text in three genres:
descriptive, narrative, and informative. Their writing was analyzed by the software based on pragmatic
functions specified by each of the three genres. Findings showed that, overall, learners were able to write
different types of essays even with no training in English genres. These findings showed the possibility of
using this type of technology to examine pragmatic functions at the discourse level.
Following Chapter 10, a commentary by Andrew Cohen presents detailed comments on each study. In his
closing words, he confirms the significance of studies included in this book and emphasizes the
importance of implementing technology in general L2 instruction. In the prologue, the two editors share
their thoughts on the possible role of technology in future ILP research and instruction. For ILP research,
they talk about how more advanced technology can provide various digitized data of L2 learners
pragmatic performance such as texts generated from telecollaboration (e.g., asynchronous and
synchronous CMC). Moreover, it can provide more behavioral information when learners are completing
real-time tasks (e.g., eye-tracking and using backend database software). For ILP instruction, they discuss
how online authorship, social networking, mobile learning, and digital game playing offer venues for
teaching L2 pragmatics. The editors argue that technology is not just a tool that can help researchers
collect and analyze data, but also a mediating artifact that can expand the area of inquiry in the field of
ILP.
The volume has revealed the status quo of the interface between technology and ILP studies. The nine
core studies use various technologies in their research design, data analysis and context of pragmatic
learning; (e.g., digital spaces) for example, experimental lab software was used to measure response time
in pragmatic tasks such as PsyScope, SuperLab Pro in Chapter 2 and Revolution in Chapter 3. Text-based
tools with built-in linguistic codes were used to analyze pragmatic features in learners writing corpora
(Chapters 6 and 10). Web-development software was used to design web-based pragmatic tests (Chapter
9). These types of technology served as tools to expand the construct of pragmatic competence (Chapters
2 and 3), to analyze data (Chapters 6 and 10), and to change the interface of a standardized test (Chapter
9). They all showed strengths of technology-enhanced studies and practical reports; however, the specific
knowledge of these types of technology use is required in its implementation, which limits the
applications of these tools. Four other studies, on the other hand, explored the benefits of interactions
supported by technology in pragmatic teaching and learning (Chapters 4, 5, 7, and 8). With supporting
technology, L2 learners can interact with NPCs to learn pragmatic features such as with Croquelandia in
Chapter 4 and Mentira in Chapter 7. They can also develop their pragmatic competence through
synchronous (e.g., Livemocha in Chapter 5) and asynchronous CMC (e.g., blogging in Chapter 8). The
major function of CMC is to provide interactions that may involve pragmatic performance; however, the
pragmatic functions performed in daily face-to-face communication may be different from those
performed in text-based CMC. A tool that can provide face-to-face conversations, therefore, may be
beneficial for L2 learners, specifically in a domestic instructional context. The current Web 2.0 tools such
as videoconferencing software (e.g. Skype) can be used to conduct face-to-face communications. But no
such studies were included in this volume, showing paucity in existing literature on pragmatic
55
Feng Xiao
performance in oral computer-mediated interactions (Heins, at al., 2007; Yanguas, 2010). These studies
can shed light on how L2 learners pragmatic competence and development can be mediated by oral
computer-mediated interactions. Another future direction that the volume can take is discussing target
pragmatic features that are beyond utterances. Technology can allow researchers to analyze discourse
patterns (Chapter 10); therefore, researchers should take advantage of these tools to analyze pragmatic
functions at the monologic and dialogic levels (Rover, 2011).
In summary, the volume is a good collection of research studies that critically discuss the implementation
of technology in the field of ILP. The book is a timely addition to the growing body of work on the
interface between technology and L2 pragmatics. To the best of my knowledge, this publication is the
first book dedicated to this topic and is therefore not only beneficial for researchers interested in this
topic, but also for practitioners who would like to incorporate technology into the teaching of L2
pragmatics.
______________________________________________________________________________
ABOUT THE REVIEWER
Feng Xiao is a Ph.D. student in the Department of Modern Languages at Carnegie Mellon University. His
research interests include interlanguage pragmatics, interpersonal pragmatics, and second language
pedagogy.
E-mail: fxiao@andrew.cmu.edu
______________________________________________________________________________
REFERENCES
Heins, B., Duensing, A., Stickler, U., & Batstone, C. (2007). Spoken interaction in online and face-to-face
language tutorials. Computer Assisted Language Learning, 20 (3), 279295.
Roever, C. (2011). Testing of second language pragmatics: Past and future. Language Testing, 28 (4),
463481.
Yanguas, . (2010). Oral computer-mediated interaction between L2 learners: Its about time. Language
Learning & Technology, 14 (3), 7293. Retrieved from http://llt.msu.edu/issues/october2010/yanguas.pdf
56

http://llt.msu.edu/issues/february2014/chenyang.pdf

pp. 5775
FOSTERING FOREIGN LANGUAGE LEARNING THROUGH

TECHNOLOGY-ENHANCED INTERCULTURAL PROJECTS
Jen Jun Chen, National Sun Yat-sen University
Shu Ching Yang, National Sun Yat-sen University
The main aim of learning English as an international language is to effectively
communicate with people from other cultures. In Taiwan, learners have few opportunities
to experience cross-cultural communication in English. To create an authentic EFL
classroom, this one-year action research study carried out three collaborative intercultural
projects using web-based tools (online forums, weblogs, Skype, and email) in a 7th grade
EFL class. The projects were designed to improve students language skills and
intercultural communicative competence (ICC). To triangulate the findings, qualitative
and quantitative methods were used to collect the data; specifically, questionnaires,
interviews, and document analyses were used to investigate the learners responses and
learning processes. The results revealed that the participants had strong positive attitudes
towards technology-enhanced intercultural language learning (TEILI), which enabled the
learners to experience authentic language learning that fostered linguistic competence and
ICC. The findings suggest that TEILI approximates real-life learning contexts by allowing
students to use a language for the same purposes that they will use it outside school.
Keywords: Culture; Web-Based Instruction; Video
APA Citation:. Chen, J. J., & Yang, S. C. (2014). Fostering foreign language learning
through technology-enhanced intercultural projects. Language Learning & Technology
18(1), 5775. Retrieved from http://llt.msu.edu/issues/february2014/chenyang.pdf
Received: December 27, 2012; Accepted: March 7, 2013; Published: February 1, 2014
Copyright: Jen Jun Chen & Shu Ching Yang
INTRODUCTION
Globalization and the rapid spread of English have challenged traditional notions of Standard English and
language education practices (Shomoossi & Ketabi, 2008). Modern English language learners use English
to communicate with native speakers and, increasingly, with non-native speakers. The main aim of
learning English as an international language is to effectively communicate with those from other
cultures. English should therefore be taught as a means of cross-cultural communication (Erling, 2005;
Jenkins, 2006; Kilickaya, 2009; McKay, 2003).
One of the fundamental difficulties Taiwanese students of English as a foreign language (EFL) encounter
is the lack of opportunities to experience interactive cross-cultural communication in English (Liu, 2005;
Su, 2008). In classroom settings, non-native English-speaking teachers, often teaching in Chinese,
typically struggle to teach pragmatic competence. Lacking extensive knowledge of the English pragmatic
system, these teachers often focus their teaching on textbooks to help students perform well on their
exams. However, content analyses of the English textbooks used in junior high schools revealed that these
textbooks provide inadequate cultural information about Anglo-American cultures (Chen, 2007). In other
words, the inauthenticity of Taiwanese English language education cannot fulfill the demands for English
language competency in a globalized world.
Studies indicate that technology plays an important role in creating authentic language learning
environments (Thorne, 2005). ODowd provided evidence that telecollaborative activities have the
potential to support the development of students intercultural communicative competence (ICC) in a way
that traditional culture learning materials would not be able to achieve (2007, p. 146). To address
57
Jen June Chen and Shu Ching Yang
Technology-Enhanced Intercultural Projects
language-learning problems in the Taiwanese context, this action research study carried out three
technology-enhanced, collaborative intercultural projects. The aim of the projects was to demonstrate that
technology-enhanced, cross-cultural tasks could provide a larger and more realistic context of
communication for language learners, which is rarely possible to achieve with other instructional models.
Below, the relationships among the role of culture, the use of technology, and language teaching will be
discussed to elucidate the theoretical and practical frameworks of the instructional design.
LITERATURE REVIEW
The Role of Culture in Language Education
Language and culture are closely linked. Although linguistic accuracy is necessary for language users to
communicate effectively, when language is used in contexts of communication, it is bound up with
culture in multiple and complex ways (Kramsch, 1998, p. 3). Traditionally, English language education
has involved learning how English is used in native English-speaking countries. However, [t]he global
spread of English into diverse multilingual contexts has brought with it the development of many varieties
of English (McKay & Bokhorst-Heng, 2008). To increase authenticity in the teaching of English as an
international language (EIL), instructors need to re-emphasize the context of use, to re-define the
participants, and to reconsider the nature of EIL (Shomoossi & Ketabi, 2008, p. 182). English education
should consider the status of English in all of its varieties and functional ranges throughout the world,
prepare students to communicate across cultures, and create linguistic awareness through exposure to
different varieties of English.
In their standards for FL learning, the American Council on the Teaching of Foreign Languages (ACTFL,
1996) advocated for application of the five Cs of language learning: communication, culture,
connection, comparison, and community. Communication is the heart of language learning.
Understanding the cultural context of both the target language and the learners native language leads to
greater awareness of the interdependent relationship between languages and cultures. Connection refers to
interdisciplinary instruction, which provides learners with detailed information about the FL and its
cultures from multiple disciplinary perspectives. Comparison refers to increasing awareness of linguistic
elements and cultural concepts by comparing and contrasting the studied language and the native
language. Finally, community suggests that learners can use the language in an international setting and
actively participate in multilingual communities beyond the classroom. Clearly, the sociocultural
component is a significant feature of FL education (FLE). Students are expected to gain insight into and
awareness of cultural interactions in communication settings.
Recently, intercultural competence has been the central concern for instructors in EFL classrooms (Liaw,
2006). Intercultural communication is not just an encounter between cultures; it should be viewed and
analyzed as a complex process (Stire, 2006, p. 5). A range of intercultural communicative education
models have been proposed by researchers worldwide (Byram, 1997; Deardorff, 2006; Spitzberg, 2000;
Spitzberg & Changnon, 2009; Stier, 2006). The most exhaustive and influential is that of Michael Byram,
whose model incorporates holistic linguistic and intercultural competence and has clear, practical, and
ethical objectives (Byram, 1997). According to Byram, the aims of intercultural language teaching are:
to give learners intercultural competence as well as linguistic competence; to prepare them for
interaction with people of other cultures; to enable them to understand and accept people from
other cultures as individuals with other distinctive perspectives, values and behaviors; and to help
them to see that such interaction is an enriching experience (Byram, Gribkova, & Starkey, 2002,
p. 10).
Byrams model consists of five factors (see Figure 1). Critical cultural awareness is positioned centrally in
relation to the other four: knowledge, intercultural attitudes, interpreting and relating skills, and discovery
and interaction skills. Byram (2012, p. 9) insists that critical cultural awareness embodies the educational
58
dimension of language teaching and that skills, attitudes and knowledge, both linguistic and cultural,
should be centered on the dimension of critical awareness (p. 6).
Interpreting and relating skills
Ability to interpret a document or
event from another culture, explain it,
and relate it to ones own culture.
Knowledge
Knowledge of ones own culture,
that of ones interlocutor, and of
the general processes of societal
and individual interaction.
Critical cultural awareness
Intercultural attitudes
Ability to evaluate, both critically and Curiosity and openness, readiness to

on the basis of explicit criteria, the
suspend preconceptions about other
perspectives, practices, and products cultures and ones own.
of ones own culture and those of
other cultures and countries.
Discovery and interaction skills
Ability to acquire new knowledge of a
culture and cultural practices and the
ability to implement knowledge,
attitudes, and skills under the
constraints of real-time
communication and interaction.
Figure 1. Byrams model of intercultural communicative competence (Byram, 1997, 2012 & Byram et
al., 2002)
Teaching Intercultural Communication in the Digital Age
As noted above, authentic language teaching with an intercultural dimension helps students acquire the
linguistic and intercultural competence needed for communication. What matters most in this complex
interactive process is what teachers do to reach these goals. Byrams model only provides a link between
intercultural communication and FL teaching; teachers must formulate the best teaching strategies for
their own contexts (Byram, 1997 & Byram et al., 2002).
Traditionally, cultural learning in the classroom has been decontextualized and has borne minimal
resemblance to actual communication scenarios. Through telecommunications, the limitations of the
classroom can be overcome through the use of web-based tools to bring authentic texts and real
intercultural communication experiences into the classroom (Byram, 1997 & Byram et al., 2002). Web
2.0 technologies (blogs, Skype, and social networking sites) facilitate online practices that allow a
classroom to connect with the world (Peters, 2009). Additionally, online education communities, such as
the ePals Global Community and the International Education and Resource Network (iEARN), provide
collaborative projects that enable teachers and students to build authentic cross-cultural communication
pathways.
Studies show that by integrating technology into their curriculum, teachers enable students to experience
varied cultures and cultivate their language skills through meaningful learning situations relevant to reallife communicative events (Cunningham, Fagersten, & Holmsten, 2010; Cziko, 2004; Greenfield, 2003;
Kilimci, 2010; Lee, 2007; Richards, 2010; Smith, 2000; Wu & Marek, 2010; Wu, Yen, & Marek, 2011).
This allows learners to develop meaningful relationships with one another and to use the language they
are studying to do so (Thorne, 2005).
Research Questions
Based on the evidence presented above, this study designed 3 web-based projects to promote crossLanguage Learning & Technology
59
cultural communication and expose students to various English language contexts. The umbrella term
technology-enhanced intercultural language instruction (hereafter TEILI) is used in this study to
describe a cross-cultural FL instruction model mediated by technology tools. The study aims to illustrate
the possibilities and problems associated with TEILI, evaluate the influence of TEILI on FLE, and
examine the instructional challenges of TEILI by soliciting teaching reflections. The research questions of
this study are as follows:
1. How do learners respond to TEILI?
2. What benefits and challenges did the learners experience during TEILI?
3. What teaching-related challenges are associated with TEILI?
METHODOLOGY
Participants
The 15 participants were 7th grade students in a pull-out bilingual program. Students were removed from
their regular classes ten class periods per week to receive specialized instruction in Chinese and English.
These participants had to pass language tests in both Chinese and English before enrolling. Compared to
their classmates, these students were advanced learners with stronger language skills. A beginning-ofsemester survey revealed that all of the students agreed that English is an important tool for connecting
with the world. Nine of them had studied English in cram schools for over 6 years, and 4 had previously
used the Internet to learn English.
In the 2011 academic year, the first author taught a one-year English course in this pull-out program,
meeting with the students regularly for 45 minutes per week. The research comprised the records and
analyses of the classroom activities in which students were engaged throughout the course.
Instructional design
The one-year English course utilized TEILI to create a realistic language-learning environment. Three
projects were conducted in the course: Folk tales/storytelling: past and present in the first semester (16
weeks) and, in the second semester, Video conference: storytelling and cross-cultural discussion and Epal project (8 weeks each) (see Table 1). Since none of the participants had ever been exposed to TEILI,
the first two projects consisted of group work to allow students to become more comfortable with this
mode of instruction. The final project focused on one-on-one communication. At the beginning of the
course, participants were asked to complete a questionnaire that evaluated their language learning
experience. At the end of each project, a reflective questionnaire was administered to encourage students
to reflect upon their learning. Finally, a questionnaire was given at the end of the course to analyze the
participants perceptions and to obtain course evaluations. The students learning journals were
systematically collected and examined, and the curriculum and teaching was adjusted according to
students feedback. The project activities are detailed in the following sections.
Table 1. Theoretical and Procedural Framework of the Three Projects
Projects
st
1 semester
1st project
Folk tales/
storytelling:
Past and present
Description
Learning focus
ICT tools
Beginning-of-semester survey about students language learning backgrounds

Small-group learning: 16 weeks
Participants shared
Language skills: Reading English stories;
traditional folk tales and writing traditional folk tales and creative
creative stories in an
stories.
online forum.
Intercultural communication: Exchanging
ideas and feedback on stories with Dubai
Weblog
Online
forum
60
students.
Mid-term survey analyzing students perceptions of the project
nd
2 semester
2nd project
Videoconference:
Storytelling and
cross-cultural
discussion
3rd project
E-pal project
Small-group learning: 8 weeks

Participants told
traditional stories in a
puppet show and then
discussed customs and
daily life via
videoconferencing.
Language skills: Listening and speaking to Weblog

non-native speakers via video
Skype
conferencing; writing scripts for the puppet
show.
Intercultural communication: Live
discussion with Pakistani students about
folk tales and cultural similarities and
differences.
Individual learning: 8 weeks

Participants exchanged Language skills: Reading and writing
weekly emails through emails.
ePals.
Intercultural communication
Exchanging information about daily life
with American key pals.
Weblog
E-mail
End-of-term survey analyzing students perceptions of the whole course

A class weblog (see Figure 2) was set up to facilitate teaching and learning in the course. The blog
contained a collection of information and learning resources that gave step-by-step support to students as
they completed their project tasks. Each group had its own individual blog within the class blog in which
students wrote their project drafts and did peer-corrections. The researcher also answered students
questions and commented on students work through the class weblog.
Figure 2. Screenshot of the class weblog

Folk tales/storytelling: past and present
Folk tales/storytelling: past and present is an iEARN project in which students from different parts of the
61
world share stories through their preferred digital forum. iEARN, founded in 1988, is the world's largest
non-profit educational network, and each iEARN project has its own project forum that provides a safe
and structured online discussion environment. Our project partner was a class from Dubai containing 10
students who had been learning English for more than 8 years. We began with a preliminary project plan
that became more defined through experience with each schools schedule and the students English
language abilities. Figure 3 illustrates the 5-phase procedural framework of the project conducted in
Taiwan. Before the project, 2 technology lessons were offered to familiarize the students with the iEARN
forums and web-based tools that would be used in the project. During the interactive process, students
visited the project forum to read stories from other countries and exchange ideas with project participants.
Figure 4 shows the intensive interaction between the students and their foreign partners in the project
forum.
Figure 3. The procedural framework of the Folk tales/storytelling: past and present project
62
List of posts
showing
participants
nationalities
Figure 4. Screenshot of the folk tales project online forum

Video conference: storytelling and cross-cultural discussion
In the second semester, we had two 45-minute videoconferences via Skype. This was an extension
activity of the Folk tales project and involved sharing folk tales with two Pakistani partners. For this
activity, the students turned their stories into a puppet show. They worked collaboratively to adapt their
stories into scripts, practice reading the scripts aloud, make paper puppets, and draw the story
background. Along with these tales, the students introduced their partners to traditional customs and food.
Furthermore, to build intercultural awareness, the students were asked to read a book about a Taiwanese
womans experience in Pakistan, and a class discussion was held. Then, in the videoconference, the
participants asked each other questions to learn about the similarities and differences between their
cultures.
E-pal project: email exchange
After the first two collaborative projects, the E-pal project was used to build students independent
English language skills and to offer them opportunities to develop autonomous language learning skills.
Our partner class was American. Each student had his or her own American key pal, with whom they
exchanged weekly emails through ePals webmail. The students learned email formatting and netiquette
before the exchanges began. To relieve students anxieties and enhance their confidence in writing, this
was a free language exchange. Students could write about anything that they were interested in and
worked on their email-writing independently during class time. The instructor gave advice to help the
students decide upon their writing topics and solve their language problems.
Data collection
The research utilized both qualitative and quantitative methods of data collection and analysis, including
questionnaires, interviews, and document analysis. Five separate questionnaires were administered:
1. Beginning-of-course survey
The beginning-of-course questionnaire contained 6 questions relating to students language learning
background and self-evaluation on their language skills. The findings from the survey helped the
researchers to understand the participants and create an appropriate instructional design.
63
2. Mid-term survey
The mid-term survey, conducted at the end of the first semester, allowed students to make comments and
suggestions; it was helpful in evaluating the first-semester course and adjusting the curriculum in the
following semester. The first part of the questionnaire comprised 14 5-point Likert scale measuring the
students attitudes towards the project and the instructional design. In the second part, the students were
asked to express their thoughts about the course and their own learning process.
3. Two reflective surveys
At the end of projects two and three, reflective questionnaires were used to encourage students to reflect
upon their English language learning. Each reflective questionnaire contained two parts. The first part
solicited responses about learning attitudes, teaching activities, and learning results in a 5-point Likertstyle format. The second part, which employed open-ended questions, required students to report their
difficulties and learning gains in the project.
4. End-of-term survey
The final questionnaire, which had 5 open-ended questions, was created to provide an overview of
participants experiences in, perceptions of, and attitudes towards the whole course.
FINDINGS
Students attitudes towards TEILI
The one-year course utilized 3 projects with different language activities to enhance English language
learning and cross-cultural communication. Table 2 shows that the participants had strong positive
reactions towards TEILI. The learners enjoyed studying English through these intercultural projects and
especially enjoyed the E-pal project (M = 4.80). The students affirmed that the projects helped them to
learn English in an authentic learning context. In the interviews conducted at the end of the Folk tales
project, all of the learners but one said that they would like to have had more intercultural language
learning activities. Table 2 reveals the same responses at the end of the videoconference (M = 4.33) and epal exchanges (M = 4.40).
We utilized a weblog-assisted teaching model and a collaborative learning method to overcome the
students lack of experience with intercultural projects. In the mid-term survey, over 93% of the students
affirmed that the weblog-assisted teaching model was helpful for their projects. Table 2 shows that
collaborative activities can reduce learning pressure in the Folk tales (M = 4.60) and Video conference (M
= 4.00) projects. The instruction conducted during the projects was acceptable to the learners. The
learning tasks were not beyond the learners language abilities, and the E-pal project was considered the
easiest (M = 4.40).
Table 2. Learners Attitudes Towards TEILI and its Instructional Design
Learners attitudes
I like this new type of learning activity involving

communication with foreign students.
Folk tales
Video
conference
E-pal
project
(SD)
(SD)
(SD)
4.26
(.79) 4.27
(.88) 4.80
(.41)
The project enables us to use English in realistic situations. 4.33
(.61) 4.40
(.82) 4.33
(.72)
4.33
(1.04) 4.40
(.63)
(.72) 4.60
(.63) 4.73
(.50)
I hope that in the future I can keep participating in this type

of learning activity.
I appreciated the teachers instruction during the project; it
4.66
64
was clear and helpful.

Learning through collaboration can reduce learning
pressure.
4.60
(.63) 4.00
The learning activities in the projects were not difficult for

me.
4.00
(1.19) 4.00
(1.00)
(.85) 4.40
(.63)
In the end-of-term survey, all of the participants preferred TEILI to traditional classroom-based
instruction. The primary reason was because TEILI provided authentic, lively, practical, and interesting
learning experiences. In describing the benefits of TEILI, many of the participants criticized the limits of
traditional instruction in Taiwanese English classrooms, in which rote memorization of language
knowledge is emphasized over language use; two students used the term dead English to describe
learning in traditional classroom-based instruction (S2, S5) and preferred the living English of TEILI.
With traditional classroom teaching, I dont know how to use the English that I have learned in
real life. With TEILI, I can communicate with foreigners. I have more opportunities to use the
English that I have learned. (S6)
With traditional instruction, it takes us a lot of time to memorize vocabulary and grammar rules,
but my mind goes completely blank when I need to speak English. With TEILI, I can
communicate directly with English speakers. It is a more real experience and is more helpful.
(S7)
I prefer TEILI because it makes English class more interesting, vivid, and lively. With traditional
teaching, we only read textbooks. With TEILI, we can use English in a natural environment, just
as we use Chinese. (S13)
The above comments indicate that traditional instruction could not satisfy students learning needs. The
students were aware of the aim of learning English via TEILI and were eager to learn in a realistic
communicative context.
Students evaluations of TEILI
To improve the instructional design, students were asked to provide feedback about the curriculum and its
implementation in the end-of-term survey. All of the students expressed positive attitudes towards the
researchers instruction. They commented that the instructional practices were helpful and sufficient, with
no need for improvement (S1, S2, S5).
Interestingly, the students and the instructor differed in their views of the teachers role in TEILE. While
the instructor viewed herself as a facilitator, some students felt that the course instructor was strict
because the students were asked to complete each task on time. Students who could not meet the
deadlines had individual meetings to discuss their problems. Some students had positive attitudes towards
the requirements. One student commented, Teacher, keep your strict education, and then students will
learn they should work hard on their project tasks (S13).
Given that students bring differing perspectives to language learning, learners with passive orientations
require careful guidance and intervention from pedagogical applications to this approach. Strict
supervision can ensure that projects go smoothly and can help students gain awareness that learning
English through TEILI demands new learning strategies and self-directed learning. Strict project
management also assures successful interactive experiences, which is beneficial for building students
communicative confidence in cross-cultural projects.
Table 3 presents the students evaluation of the projects; the project that 66.7% of the students liked best
was the E-pal project. This was also the project that 80% of the students chose as the one with the easiest
65
writing tasks, consistent with the earlier analysis of the students attitudes toward TEILI, shown in Table
2. Unlike the other two projects, E-pal was a free writing exchange and also a one-on-one communication
project. This gave students a sense of novelty and motivated them to communicate. The students
commented that they were looking forward to having a pen pal (S1), enjoyed writing about anything they
liked to their foreign friends, and gained more knowledge about foreign life (S6, S10).
Table 3. Learners Attitudes Towards the Projects
The project
Folk tales
project
Video
conference
E-pal project
I liked most
6.7%
26.7%
10
66.7%
I liked least
12
80%
13.3%
6.7%
with the hardest writing tasks
46.7%
40%
3.3%
with the easiest writing tasks
20%
12
80%
Most of the students shared information about their everyday lives in their emails. For students, writing
about their personal experiences was easier than writing traditional stories. The students said that the Epal project allowed them to write about daily life as if they were chatting and without pressure, and it was
therefore easy to write a lot in a short time (S3, S8, S10).
There is one positive comment especially worthy of notice. A student mentioned the benefit of
asynchronous communication:
We dont have to write back to the key pals right away. I have enough time to think carefully
before writing. (S11)
S11 is a quiet, shy boy and group discussions and synchronous activities, such as video conferencing, put
him under pressure. In asynchronous activities, he was able to take time to think and work at his own
pace. His example reminds us that students have unique personalities and learning styles. Teachers need
to consider the advantages and disadvantages of technology-enhanced learning activities and provide
multiple tasks to support learners diverse learning needs.
In contrast, the project most students did not enjoy was the Folk tales project (n = 12, 80%), which was
also chosen as the project with the most difficult writing tasks (n = 7, 46.7%). The students enjoyed the
collaborative writing in the Folk tales project and Video conference; however, creative writing proved to
be a challenge for students with little English writing experience. The students made the following
comments:
Group writing relaxes my writing pressure, so I dont think it is hard to write stories. However,
it is a challenge to modernize traditional stories or adapt stories to scripts. (S4)
When writing a story, we have to pay attention to a lot of elements, such as vocabulary,
grammar, and even dialogues between characters. We need imagination and creativity. It is not
easy. (S8)
In addition, students also mentioned that they expected to have multi-aspect exchanges instead of
focusing exclusively on exchanges of traditional stories. S8 mentioned that there was less direct
communication with foreign partners in the Folk tales project than in the other two projects. The students
expected to have more interactions that would help them understand different cultures and foreign life
more deeply (S3).
Students learning benefits from TEILI

The students reported in their reflections and interviews that they received many learning benefits from
66
TEILI, including increased understanding their own and other cultures, increased English vocabulary,
improved language skills, experience using technology learning tools, experience with intercultural
communication, and improved collaboration skills. Table 4 shows that cultural understanding was the
main learning benefit reported for all three projects. The reflective surveys for the Video conference and
E-pal project also showed the students positive attitudes towards cultural learning; both the Video
conference (M = 4.20, SD = .86) and E-pal project (M = 4.20, SD = 1.01) improved students cultural
understanding, but the E-pal project, with its close individual interaction, provided the participants the
most opportunities to explain their own culture (M video = 3.93, SD = .70; M e-pal = 4.20, SD = .86).
Table 4. Students Learning Benefits from TEILI
Folk tales/storytelling
Improving language
ability (vocabulary and
writing skills)
I forgot a lot of English vocabulary that I had memorized before, but now I know these
words again because I have to use them in my writing. I learned more vocabulary. (S4)
I learned how to respond to messages and interact with foreign students on the forum.
At first, I had to think about the feedback content for a long time. Now, I can reply
faster. (S6)
During the project, I felt that I was making progress in writing. Now, I can write a
longer forum message than before, without just using a lot of emoticons. (S7)
Improving technology
ability
I learned a new way, iEARN, to communicate with foreign students through the
Internet. I know more about the world. (S8)
During the writing process, ...I also learned how to use online dictionaries in learning
English. (S3)
I learned more technology tools, and now I can type faster in English. (S9)
Developing
collaboration skills
I learned how to work collaboratively with others. (S11, S12)
Understanding
different cultures
I learned more about our own folk tales and also learned about other countries folk
tales. (S3)
Video conference
Understanding
different cultures
We shared our stories, food, and festivals; they also shared their culture...I learned a lot
and understood that others lives are not the same. (S2)
I touched a very different culture. I saw that the clothes they wore were so different
from ours. Every country has its own unique culture. (S5)
I knew about Pakistani students school life. Their school life is similar to ours. They
wear school uniforms, and they also have PE classes. (S8)
E-pal project
Understanding
different cultures
Their school schedules are very different from ours. Their school starts later, but they
go home earlier than us. (S8)
American students like sports very much. They play many outdoor sports. They also
have field trips (over 3 or 4 days). They have to do reports after the trips. (S7)
They dont have to wear uniforms, and they dont always have the same subjects. For
example, some choose to study Spanish, and some study French. We all have the same
subjects and only study English. They are freer and have more choices. (S10)
The Whole Program

Improving language
ability
My largest gain is that I have learned a lot of vocabulary. To write articles and emails
correctly, I looked up words and made sure their usages, so I learned a lot of new
words. (S2)
67
Through the communicative process, I understand my learning problems. The projects

improve our English ability and let us learn new vocabulary. (S9)
Improving technology
ability
I have better technology skills than before. I didnt know we can use these tools to learn
English before. The Internet has a lot of learning resources and I learn knowledge
beyond textbooks, (S11)
Understanding
different cultures
I understand different countries and cultures deeply. I also understand the differences
between us and foreign students. (S6)
The students agreed that TEILI created opportunities for learning about cultures, including their own and
their interlocutors. Based on Byrams model, we found that, through TEILI, the students developed their
ICC skills in four dimensions (see Table 5): knowledge of their own and their interlocutors cultures,
open attitudes towards intercultural exchanges, skills for interacting with interlocutors, and a critical
cultural awareness necessary to evaluating cultures.
Due to the interlocutors diverse cultural backgrounds, the students reflections indicated that their
knowledge about cultures was largely acquired through the process of interaction, which is one of the
main advantages of the TEILI model. The students introduced their own culture and learned about life in
Dubai, Pakistan, and the USA through their communications. They developed cultural awareness of both
their home culture and the cultures of their interlocutors.
Students reported that most of their critical cultural awareness was developed in the E-pal project. Unlike
the group interactions, the students experienced an intensive compare/contrast process in the individual
interactions of the E-pal project. They learned facts about American school life, gave responses to their
interlocutors, and shared their own experiences of attending school in Taiwan. In every email exchange,
the students familiarized themselves with the use of comparison and contrast to view cultures critically.
Table 5. Students Development of Intercultural Communication
Dimensions
Knowledge:
Students descriptions
of ones own culture
and the interlocutors
culture
Intercultural
attitudes:
Curiosity and openness to

learn about ones own
culture and the
interlocutors culture
Discovery and
interaction skills:
Critical cultural
awareness:
The ability to acquire

new knowledge of a
culture through
synchronous interaction
The ability to evaluate a
culture
For me, these learning experiences broaden

my vision! I learn foreign cultures, customs,
and traditions. (S6)
I not only learn other cultures but know more
about my own culture. (S7)
I hope to have more opportunities to
communicate with foreign students; then, I
can practice intercultural communicative
skills and know about the world. (S7)
I hope we can keep working with other
countries on new projects. (S5)
I learn different cultures by using English and
Internet tools in real communication. (S6)
I learn how to exchange ideas with foreign
students. (S10)
They dont have to wear school uniforms, and
they have a monthly dance. Their school life
seems more colorful, with fewer restrictions
than ours. (S10)
They also have exams like us, but only at the
68
beginning and the end of the semester...It

seems that we pay more attention to studying.
They have more time to play sports, but we
dont. American students seem freer, with less
pressure. (S13)
Students self-perceived challenges with TEILI
Though students showed positive attitudes towards the new teaching model, several of them still had
difficulty interacting. Table 6 reveals that low vocabulary knowledge was the major barrier for students in
their projects. Due to their limited vocabulary, the students did not know which words to use to express
their ideas well. Such challenges may be unavoidable in the Taiwanese EFL context. These problems also
reflect the value of the course: building language ability through realistic communication. Students are
able to enlarge their vocabularies as they experience cross-cultural communication challenges.
To solve their vocabulary deficiencies, students usually used online tools to look up words or directly
asked the instructor for help. However, some relied on online translation tools to translate entire
messages. This was convenient, but improper. These translation tools may work well for words or simple
sentences, but they always almost translate complex descriptions incorrectly. Using improper tools may
create other learning problems and cause confusion and misunderstanding. The result indicates that it is
necessary to teach students how to use technology tools correctly and in ways that will benefit their
language learning.
Accents were another challenge. Because American English was taught in their school, the Taiwanese
students were only familiar with American English pronunciation. With few opportunities to listen to
other accents, it was not easy for students to immediately become involved in real language situations
during the videoconferences. For these students, an unfamiliar accent sounded like a different language.
However, this provided an opportunity for students to realize the variety of authenticity within the
English-speaking community. One of the students commented,
I understand our accent is not always correct. There are different accents in the world. We
should try to understand and accept others accents. (S10)
From the following students feedback, we can infer that if students are exposed to this authentic oral
interaction for a longer period of time, they may accept the accent more and understand it better:
At first, I totally couldnt understand what they said, but as we spoke to the second school, it
became better. The accent was not so strange to me. I could understand more. (S1)
Table 6. Students Learning Challenges with TEILI

Folk tales/storytelling
Lack of language skills
(vocabulary)
Problems of withingroup cooperation
When I read Dubai information on the English website, there were a lot of words I
didnt know. I tried to use online dictionaries or the Google translator, but some
English words were still difficult to understand. (S4)
My writing was a big problem. I need more vocabulary and to learn to write good
sentences. (S6)
I had difficulty finding the right words to clearly express the meaning I wanted. (S12)
Sometimes I thought the story we wrote was very strange. Though the group worked
together, members usually had different ideas. We didnt know which one was better,
so it usually took us a long time to discuss the story plot. (S6)
Video conference
69

(accent, vocabulary,
grammar)
Language anxiety
Because of the accents, I had to listen to their words very carefully. However, I still
couldnt understand all their words. Sometimes they spoke fast; then it was even more
difficult to understand what they meant. (S8)
Because of the accents, I couldnt hear clearly. I often misheard what they said or
misunderstood the meaning. It seemed we spoke different languages. (S9)
I feel that writing the script is the most difficult part. I have problems in vocabulary,
grammar, etc. I usually think in Chinese, and then have trouble in writing it in
English. (S10)
As I wrote the script, my problem was that I didnt have enough vocabulary to
express the meaning. I depended on the Internet dictionary to look up the words. (S7)
Though I had prepared in advance, I was very nervous and forgot what to say. (S11)
I was nervous during the VC. When I was nervous, I started to stutter and didnt
know what to say. (S14)
E-pal project
(vocabulary, grammar)
There were no serious problems; only some problems with vocabulary or grammar.
(S5)
There were no big problems or difficulties. My problem is with vocabulary. I should
learn more English words. (S10)
The students feedback reveals the disadvantages of Taiwans English language learning environment.
Typically, English is taught in decontextualized forms; students learn the correct forms of the language
and are able to use English in a structured situation. However, the authentic context is unstructured, ill
defined, and interactive. If students cannot transfer their knowledge into functional use, they will not be
able to use the language actively outside the class.
Teaching reflections on TEILI
Though research, including this action study, has verified the positive effects of TEILI, the
implementation of TEILI in classrooms is a challenge for teachers. Unlike traditional, static, classroombased instruction, TEILI is a dynamic process based on interactions among participants. The differences
in educational systems and participants backgrounds create many uncertain features in the interactive
process. The teacher, as a project manager, must connect all of the elements of the TEILI program and
consider not only instructional tasks but also project arrangements. The teacher must also resolve
unexpected problems in the interactive process. It is important for teachers of partner classes to work
closely to create suitable and realistic project plans in advance. Then, during the project process, they
must remain in close communication with each other to ensure that the whole TEILI program advances
smoothly and successfully.
Since instructors must exchange ideas frequently, TEILI builds authentic intercultural communication not
only for learners but also for instructors in the EFL community. Teaching is learning; collaborating with
other foreign educators will enable instructors to use the language that they teach, establish pragmatic
competence, and gain professional development from real experience. This course instructor benefited
greatly from interacting with the partner teachers, and especially from their instructional designs for the
project activities. The Dubai teacher shared his method for inspiring his students to discuss different types
of folk tales before writing; the American teacher worked with his colleagues to integrate issues of
globalization into the email exchanges. They demonstrated creativity in the teaching designs and
prompted the course instructor to think about teaching and learning from a different perspective.
Though it takes much time and effort to develop a TEILI course, the students positive feedback during
this study was encouraging. The students always rushed into the classroom, eager to learn and
communicate with their foreign partners. I did not have to struggle with classroom discipline or urge the
70
students to work hard because their foreign partners were the best stimuli for motivating language
learning. As they heard the Pakistani students fluency in English, they were shocked, and the students
commented that the Pakistani students were so active, and their English was so good. It seemed English
is easy for them. They could say what they want to say. I should work hard on my English (S9). Though
I couldnt understand all their English, their English ability shocked me. If I have another opportunity, I
will avoid the problems and improve my English (S13). The comments reveal that students can reflect on
their own language learning through TEILI and find the motivation to become active learners. The
instructors challenge is to provide the context, direct the process of learning, and foster students abilities
to understand the changing world around them.
DISCUSSION
The participants responses to TEILI and its instructional design validated the pedagogical benefit of
extending the learning context outside the classroom to provide authentic and meaningful language
learning. The result agrees with the current research on technology-enhanced learning (Warschauer, 1997;
Kern, Ware, & Warschauer, 2008). The findings are discussed in terms of what we can learn about
language learning and instructional design from TEILI.
Language learning
The analysis of the learning process showed that a lack of FL skill was the students main barrier during
the intercultural projects (see Table 6). In the process of intercultural interaction, the participants were
aware of their language difficulties. The students described the obstacles presented by vocabulary,
grammar, writing and accents. Leow (2000; 2001) found that awareness plays a critical role in second
language learning. Gilakjani (2011) also concluded from previous studies that conscious awareness of
language is necessary for second language acquisition. As Gilakjani (2011) comments, Both
consciousness and language are inextricably connected like two sides of a coin. TEILI, which recreates
authentic situations, provides opportunities for social interactions that evoke learners conscious
awareness of the target language system. This helps learners to use language correctly and appropriately
in later tasks. The findings confirmed that the students perceived their language skills to be improved
through the project tasks (see Table 4). From the instructors perspective, the students language
challenges brought about their learning benefits.
The first two projects of this action study focused on the same topic: folk tales. Multimodal projects
integrating multiple semiotic modes, such as written words and imagery, were expected to enable students
to transform the mode of communication and expression from text into speech, reducing the students
anxieties when speaking in English. However, the teaching result did not entirely meet the expectation.
The students experiences revealed that the multimodal design lacked transformation between tasks and
had no active effect on lowering anxiety (see Table 6: Language anxiety). According to Krashens (1982)
affective filter hypothesis, second language learners with high motivation, self-confidence, positive
attitudes, and low levels of anxiety are better equipped for successful language learning. For the students
with high anxiety, the affective filter became a barrier to language performance. To reduce the negative
affective factor during videoconferencing, it is necessary to revise the learning tasks to allow students to
become more familiar with synchronous communication.
In his research on multimodality, Nelson (2006) examined the synesthetic functions of transformation and
transduction through the process of multimodal communication. Other research has shown that language
skills can be transformed from text, as in online chatting, to second language speech (Blake, 2005; Payne
& Ross, 2002; Sanders, 2006). The resemblance between an online task and a real-time conversation is a
vital factor for successful transformation across semiotic modes. A re-examination of the learning tasks in
the first two projects may reveal that the similarity across tasks may be not high enough to facilitate
learning transformation, especially given the differences between asynchronous and synchronous tasks.
Based on the positive effects found in multimodality research, conducting an online real-time chat
71
(textual exchange) before videoconferencing (audio and video exchange) may increase students selfconfidence in synchronous conversations.
Instructional design and ICC
Thorne (2005) once described Internet-mediated intercultural FLE as a complex Venn diagram that
overlapped with other areas of research and with numerous models. In addition, each model has varied
instructional designs. Educators must consider which model or instructional method will be best suited for
their context. The research suggests that students levels of cognitive development could be one of the
most important considerations when creating an effective TEILI instructional design.
Richards (2010) launched a study on technology-enhanced intercultural collaboration between elementary
school students in the U.S. and Jordan. The 16-week project focused on guided story writing, and ePals
and Skype were the communicative tools. The researcher, who was also the project facilitator, worked
closely with two instructors to design the project plan and assign tasks. This tended to be a teachermanaged instructional design. Our action research chose the same tools (ePals and Skype) and similar
themes (sharing stories and information about daily life), but we utilized a teacher-and-student-managed
instructional model in a junior high school context. The instructor was a facilitator; the students engaged
in discussion and had great freedom in deciding how to interact with their foreign partners. Based on
theories of cognitive development, the tasks were designed to require different cognitive skills and
proceeded in an orderly developmental sequence, gradually transforming from sensory-motor actions to
representations and then to abstractions (Fischer, 1980). At different ages, students think in different
ways. Generally, middle school students are becoming increasingly capable of handling more
sophisticated and abstract learning tasks. We therefore required our students to engage in complicated
tasks autonomous learning and critical thinking tasks that would challenge them.
Telecollaboration conducted at the university level tends to adopt a more student-managed instructional
model. For instance, Liaw (2006) created a web-based system with self-help resources in the form of
online reference tools to foster reading discussions between Taiwanese university students and their
American partners. The participants were seen as independent, self-directed learners.
Above all, telecollaboration cases insist on the effectiveness of international class-to-class partnership
within institutionalized settings (Thorne, 2005, p. 4). With proper designs that consider students levels
of cognitive development, TEILI can be put into practice across a variety of education levels.
In addition to language learning, another core aim of TEILI is to develop intercultural communication
awareness. All three of the above cases revealed that intercultural collaboration may enhance cultural
understanding and cultural awareness. Due to differences in the project designs and the ages of
participants, the aspects of the students intercultural development were not the same. However, in all
three cases, students perceived that they could gain knowledge about their own and other cultures. Liaw
(2006) showed that students developed the ability to change their perspective of and knowledge about the
intercultural communication process, which was not discovered in Richards (2010) nor in our research. It
is possible that projects with more discussion tasks could encourage participants to interact with each
other more deeply and facilitate high-level ICC. It is also possible that university students have higher
order cognitive abilities and better development of ICC.
CONCLUSION
The action research revealed that the TEILI course provided authentic opportunities for students to
develop their language skills and ICC. The students acknowledged that they improved their knowledge of
the varieties of English language use and culture, improved their vocabulary, writing, and technological
abilities, and learned collaboration. TEILI participation revealed the students inadequate expressive skills
and the limits of traditional language instruction. Online educational communities provide diverse
collaboration projects for classes around the world. The study found that, with proper planning and
72
project management, learners could experience varied intercultural projects without difficulty. TEILI
promoted language-learning motivation and helped learners develop active attitudes towards language
learning and intercultural communication.
TEILI allows students use language in school that is relevant to the way they will use language outside
school. This action study implemented only one type of TEILI curriculum. Future studies should
investigate the results of integrating other types of projects into such a course. They should consider
whether it is better to work on a project over a long period of time or to try various short-term projects,
what type of TEILI curriculum design will benefit students most, and how students learning autonomies
and cultural awareness can be systematically developed through TEILI. As this action research has
shown, authentic language learning environments can be created through technology use. Further research
on curriculum design and the impact of TEILI on language teaching and learning will allow the model to
be evaluated from different perspectives.
ACKNOWLEDGEMENTS
This research was partially supported by the National Science Council of Taiwan, grant number
(NSC100-2628-H-110-007-MY3) and The Aim for the Top University Plan of the National Sun Yatsen University and Ministry of Education, Taiwan, R.O.C.
ABOUT THE AUTHORS

Jen Jun Chen is currently a Ph.D. student of the Graduate Institute of Education, National Sun Yat-sen
University, and is also a member of the country coordination team of iEARN Taiwan as well as an
English teacher at Kaohsiung Municipal Jhengsing Junior High School. She received an M.S. in
Education from National Sun Yat-sen University in 2004.
Shu Ching Yang is a professor at Graduate Institute of Education, National Sun Yat-sen University,
Taiwan. She received her PhD in Instructional Systems Technology at Indiana University of
Bloomington. Her current research focuses on the learning processes associated with various kinds of
interactive technologies. She has additional articles published in Computers and Education, Computers in
Human Behavior, etc.
E-Mail: shyang@mail.nsysu.edu.tw
REFERENCES
ACTFL. (1996). Standards for foreign language learning: Preparing for the 21st century. Lawrence, KS:
Allen.
Blake, R. J. (2005). Bimodal CMC: The glue of language learning at a distance. CALICO Journal, 22(3),
497512.
Byram, M. (1997). Teaching and assessing intercultural communicative competence. Clevedon, UK:
Multilingual Matters.
Byram, M. (2012). Language awareness and (critical) cultural awareness relationships, comparisons and
contrasts. Language Awareness, 21(12), 513.
Byram, M., Gribkova, B., & Starkey, H. (2002). Developing the intercultural dimension in language
teaching. A practical introduction for teachers. Strasbourg, France: Council of Europe Publishing,
73
Language Policy Division.

Chen, Y. C. (2007). An analysis study on the knowledge contents of cultural globalization in Junior
High School English textbooks. (Unpublished masters thesis). National Taipei University of Education,
Taipei, Taiwan.
Cunningham, U. M., Fgersten, K. B., & Holmsten, E. (2010). "Can you hear me, Hanoi?" Compensatory
mechanisms employed in synchronous net-based English language learning. The International Review of
Research in Open and Distance Learning, 11(1), 161177.
Cziko, G. A. (2004). Electronic tandem language learning (eTandem): A third approach to second
language learning for the 21st century. CALICO Journal, 22(1), 2540.
Deardorff, D. K. (2006). Identification and assessment of intercultural competence as a student outcome
of internationalization. Journal of Studies in International Education, 10(3), 241266.
Erling, E. J. (2005). The many names of English. English Today, 21(1), 4044.
Fischer, K. W. (1980). A theory of cognitive development: The control and construction of hierarchies of
skills. Psychological Review, 87(6), 477531.
Gilakjani, A. P., & Ahmadi, S. M. (2011). Role of consciousness in second language acquisition. Theory
and Practice in Language Studies, 1(5), 435442.
Greenfield, R. (2003). Collaborative e-mail exchange for teaching secondary ESL: A case study in Hong
Kong. Language Learning and Technology, 7(1), 4670. Retrieved from
http://llt.msu.edu/vol7num1/greenfield/default.html
Jenkins, J. (2006). Current perspectives on teaching world Englishes and English as a lingua franca.
TESOL Quarterly, 40(1), 157181.
Kern, R., Ware, P., & Warschauer, M. (2008). Network-based language teaching. Encyclopedia of
Language and Education, 4, 281292.
Kilickaya, F. (2009). World Englishes, English as an international language and applied linguistics.
English Language Teaching, 2(3), 3538.
Kilimci, S. (2010). Integration of the internet into a language curriculum in a multicultural society.
TOJET, 9(1), 107113.
Kramsch, C. (1998). Language and culture. Oxford, UK: Oxford University Press.
Krashen, S. (1982). Principles and practice in second language acquisition. Retrieved from
http://www.sdkrashen.com/content/books/principles_and_practice.pdf
Lee, L. (2007). Fostering second language oral communication through constructivist interaction in
desktop videoconferencing. Foreign Language Annals, 40(4), 635649.
Leow, R. P. (2000). A study of the role of awareness in foreign language behavior. Studies in Second
Language Acquisition, 22(4), 557584.
Leow, R. P. (2001). Attention, awareness, and foreign language behavior. Language Learning, 51, 113
155.
Liaw, M. (2006). E-learning and the development of intercultural competence. Language Learning &
Technology, 10(3), 4964. Retrieved from http://llt.msu.edu/vol10num3/liaw/default.html
Liu, G-Z. (2005). The trend and challenge for teaching EFL at Taiwanese universities. RELC Journal,
36(2), 211221.
McKay, S. L. (2003). Toward an appropriate EIL pedagogy: Re examining common ELT assumptions.
74
International Journal of Applied Linguistics, 13(1), 122.

McKay, S., & Bokhorst-Heng, W. D. (2008). International English in its sociolinguistic contexts:
Towards a socially sensitive EIL pedagogy. New York, NY: Routledge.
Nelson, M. E. (2006). Mode, meaning, and synaesthesia in multimedia L2 writing. Language Learning &
Technology, 10(2), 5676. Retrieved from http://llt.msu.edu/vol10num2/nelson/default.html
ODowd, R. (2007). Evaluating the outcomes of online intercultural exchange. ELT Journal, 61(2), 144
152.
Payne, J. S., & Ross, B. M. (2005). Synchronous CMC, working memory, and L2 oral proficiency
http://llt.msu.edu/vol9num3/payne/default.html
Peters, L. (2009). Global education: Using technology to bring the world to your students. Washington,
D.C.: International Society for Technology in Education.
Richards, B. E. (2010). An exploration of what happens when second and third graders in two countries
interact through a technology enhanced multicultural collaboration. Unpublished doctoral dissertation,
University of South Carolina, Columbia. Retrieved from http://gradworks.umi.com/34/34/3434964.html
Sanders, R. (2006). A comparison of chat room productivity: In-class versus out-of-class. CALICO
Journal, 24(1), 5976.
Shomoossi, N., & Ketabi, S. (2008). Authenticity within the EIL Paradigm. Iranian Journal of Language
Studies, 2(2), 173185.
Smith, E. A. (2000). Making culture and language real in a rural setting: The technology connection.
Inquiry, 5(1), 3741.
Spitzberg, B. H. (2000). A model of intercultural communication competence. Intercultural
Communication: A Reader, 9, 375387.
Spitzberg, B. H., & Changnon, G. (2009). Conceptualizing intercultural competence. In D. K. Deardorff
(Ed.), The Sage handbook of intercultural competence (pp. 252). Thousand Oaks, CA: Sage.
Stier, J. (2006). Internationalisation, intercultural communication and intercultural competence. Journal
of Intercultural Communication, 11(1), 112.
Su, Y. C. (2008). Promoting cross-cultural awareness and understanding: incorporating ethnographic
interviews in college EFL classes in Taiwan. Educational Studies, 34(4), 377398.
Thorne, S. L. (2005). Internet-mediated intercultural foreign language education: Approaches, pedagogy,
and research. CALPER Working Paper Series No. 6. The Pennsylvania State University, Center for
Advanced Language Proficiency Education and Research.
Warschauer, M. (1997). Computer mediated collaborative learning: Theory and practice. Modern
Language Journal, 81(4), 470481.
Wu, W. V., & Marek, M. (2010). Making English a habit: Increasing confidence, motivation, and
ability of EFL students through cross-cultural, computer-assisted interaction. TOJET, 9(4), 101112.
Wu, W. V., Yen, L. L., & Marek, M. (2011). Using online EFL interaction to increase confidence,
motivation, and ability. Educational Technology & Society, 14(3), 118129.
75

http://llt.msu.edu/issues/february2014/abuseileekabualshar.pdf

pp. 76-95
USING PEER COMPUTER-MEDIATED CORRECTIVE FEEDBACK TO

SUPPORT EFL LEARNERS WRITING
Ali AbuSeileek, Al al-Bayt University
Awatif Abualshar, Al al-Bayt University
This study investigated the effect of using computer-mediated corrective feedback on EFL
learners' performance in writing. Sixty-four intermediate-level learners were randomly
assigned to either a no-feedback control condition or a corrective feedback, including three
treatment conditions. The first is the track changes feature of Microsoft Word 2010
which double strikes through deletions and marks insertions in a different color. The
second is recast feedback while the third is metalinguistic feedback. Over the course of
eight weeks, each student was required to provide corrective feedback on drafts written by
another group member and to discuss it with the group members. The study yielded three
major findings. First, students who received computer-mediated corrective feedback while
writing achieved better results in their overall test scores than students in the control
condition who did not receive feedback. Second, there was a significant effect for the track
changes made in the corrective-feedback type when compared with that made in the recast
and metalinguistic feedback types. Students in the recast treatment condition also obtained
higher significant mean scores than those who received metalinguistic corrective feedback.
The study concluded with suggestions for further research and pedagogical implications.
Keywords: Corrective feedback, foreign language writing, computer-mediated
communication, peer review, error type.
APA Citation:. AbuSeileek, A., & Abualshar, A. (2014) Using peer computer-mediated
corrective feedback to support EFL learners writing. Language Learning & Technology
18(1), 7695. Retrieved from
http://llt.msu.edu/issues/february2014/abuseileekabualshar.pdf
Received: May 1, 2012; Accepted: March 29, 2013; Published: February 1, 2014
Copyright: Ali AbuSeileek, Awatif Abualshar
INTRODUCTION
Corrective feedback is one of the major tools used for enhancing English language learning and teaching
by providing students with feedback to correct their errors. It may support learners with explicit or
implicit feedback that may be useful in developing writing. Although corrective feedback has been used
for a long time in traditional classrooms, computer-mediated corrective feedback has been introduced
only recently. The user may have access to different types of computer-based facilities that may help in
providing corrective feedback for students regarding their errors and mistakes. According to Yeha and
Lob (2009), corrective feedback or error correction provided via written computer-mediated
communication could play an important role in developing learners metalinguistic awareness, especially
marking up text with colored annotations and focusing the learners attention on limited information. This
makes corrective feedback an efficient way to draw learners attention to the error and the feedback about
it in the written text. As the use of computer-mediated corrective-feedback method has become more
common in writing classes, different studies have looked for more innovative ways to aid learners in
developing their writing abilities (Hyland & Hyland, 2006), and called for a test of the effectiveness of
new technologies on teaching learners how to write (Ware & Warschauer, 2006). Moreover, as "the
computer is a pedagogical tool that needs to be carefully used, evaluated, and explored by pedagogues
and researchers alike" (Caws, 2006, p. 20), there is a need to search for more helpful computer-mediated
76
Ali AbuSeileek and Awatif Abualshar
Peer CMC Feedback to Support EFL Learners' Writing
corrective-feedback methods and techniques that may support students when receiving peer-corrective
feedback in a manner that may aid them more in the development of their writing performance. Despite
the fact that some studies were conducted on the effectiveness of computer-mediated corrective feedback
on learners' writing performance, the present study is based on introducing new computer-mediated
corrective-feedback types for learning and teaching writing, including track changes and its effect on
developing learners' performance in seven writing aspects: content, structural organization (text level),
structural organization (sentence level), grammatical accuracy, lexical appropriateness, punctuation, and
spelling.
Corrective Feedback and Language Learning
Some studies focused on providing feedback only on grammar (Bitchener, Young, & Cameron, 2005 &
Ferris, 2006), content (Ferris, 2004 & Hendrickson, 1978), or on both (Arnold, Ducate & Kost, 2009; De
Jong, & Kuiken, 2012; ODonnell, 2007; Sheppard, 1992; Van Beuningen, 2010; Vyatkina, 2011).
Though the term "corrective feedback is generally used for errors on form not content, it may mean
different things for the readers. Corrective feedback in this study refers to both feedback on linguistic
forms and content (e.g., rhetorical aspects of writing) in the computer-based written form. Corrective
feedback or error correction may aid learners writing development and can act as a functional method for
language learning. Van Beuningen (2010) reports that it fosters language learning and develops accuracy
as it offers learners opportunities to notice the gaps in their linguistic systems, testing interlanguage
hypotheses, and engaging in metalinguistic reflection. The accuracy of students who received written
corrective feedback in the immediate post-test outperformed those in the control group, and this level of
performance was retained two months later (Bitchener, 2008).
Many studies, particularly by Truscott (1996, 2009), raised objections to corrective feedback use and
pointed out that corrective feedback is unhelpful or even harmful for students' writing development
(Truscott & Hsu, 2008). In conclusion, the usefulness of written corrective feedback is a topic of
considerable debate (Ferris, 2004, 2010, 2012; Truscott, 2007, 2009). According to Gunette (2007), there
has been considerable controversy among theorists and researchers about the usefulness of the process of
corrective feedback in language learning and teaching. He added,
The debate continues between those who believe in giving corrective feedback to students
to improve their written accuracy and those who do not. Indeed, the results of the many
experimental studies on written corrective feedback carried out over the last 20 years have
been so contradictory that second language teachers looking to support their pedagogical
choice to correct, or not to correct, the grammar of their students' written production are
left in the midst of controversy. (p. 40)
For the researchers who believe in the effect of corrective feedback on students' writing, different types of
corrective feedback can be used to aid students' writing development including explicit error correction,
metalinguistic feedback, elicitations, repetitions, recasts, translations, clarification requests (Lyster &
Ranta, 1997), and reformulation of correcting error types (Santos, Lpez-Serrano, & Manchn, 2010).
Other studies (Ellis, 2004; DeKeyser, 1998; Sauro, 2009) referred to two types of corrective feedback,
including explicit and implicit or conscious knowledge about error vs. procedural knowledge. Another
classification is based on conceptualizing the feedback types as indirect or direct feedback (Ferris, 2003,
2006; Truscott & Hsu, 2008) where "direct CF [corrective feedback] consists of an indication of the error
and the corresponding correct linguistic form, [while] indirect CF only indicates that an error has been
made" (Van Beuningen, 2010, p. 12). There are also other subcategorizations of these major types. For
instance, indirect corrective feedback is divided into underlining and codes (Chandler, 2003; Ferris &
Roberts, 2001), while the effect of direct feedback is examined as direct corrective feedback with or
without written metalinguistic explanations (Bitchener, 2008; Bitchener & Knoch, 2009).
In this study, three types of corrective feedback are used, namely track changes, recast, and metalinguistic.
77
Table 1 illustrates these types. The examples are based on the sentence: They speaks English fluently. The
definitions for the terms recast and metalinguistic feedback and the categorization of the characteristics of
these corrective-feedback types are based on the model introduced by Lyster and Ranta (1997) and Sauro
(2009). However, the definitions are operationalized and adapted to suit this study.
Table 1. Characteristics of Corrective-Feedback Types in Microsoft Word 2010
Feedback type
Definition
Location
in text
Example
Nature of
error
indicated
Target-like
reformulation
provided
Elicited
output
Track changes
It is a computer-mediated
method for providing
corrective feedback that can
be used in an implicit or
explicit manner. It is based
on reformulation of the error
where the program strikes
through deletions and marks
insertions in a different
color.
Inline
They
speaks
speak
English
fluently.
Provided
indirectly
Yes
Error is
identified
and
reformulated
Recast
method used for providing
feedback. The error is
always reformulated without
providing any metalinguistic
information about it.
Marginal
comment
displayed
inline
They
speak
English
fluently.
Yes
Reformulation
provided
Repetition of
the error
in the
correct
form
Metalinguistic
feedback
method based on providing
metalinguistic information
or comment about the error
without reformulating it.
Marginal
comment
displayed
inline
Subjectverb
agreement
Yes
Provided
indirectly
Reformulation
of error
On the other hand, the location in text characteristics of feedback types represent an attempt made in this
study to specify where and how the note is displayed in the text. Figure 1 also provides an illustrative
example about computer-mediated corrective-feedback types.
Figure 1. A screenshot for computer-mediated corrective-feedback types. Here, marginal comments are
used for clarification to the reader, but they were displayed in line for students.
78
Peer written Corrective Feedback

The peer-written corrective feedback is based on the Vygotsky's (1978) sociocultural theory. That is, the
computer-mediated environment helps learners to provide corrective feedback about writing errors where
one can achieve more when they interact with peers than when they work alone. Wells (1999) also
demonstrated that the cooperative context helps the learners to receive help from their peers in the
cooperative activity. It is a learner-centered approach based on learners' interactions to correct their errors.
The learning process is considered as collaborative and cooperative (Oskoz & Elola, 2011). Therefore,
peer-written corrective feedback is based on the communicative approach which is different from the
traditional method that focuses on teaching isolated linguistic features and grammatical rules. It aims at
developing learners' ability to use language in meaningful communication where they help each other to
develop their linguistic competence while interacting. Inspired by the theory of communicative
competence introduced by Canale and Swain (1980), this study focused on creating opportunities for
asynchronous peer written interaction and meaningful communication. It is based on providing peer
written corrective feedback where each student reads essays written by other classmates and provides
written corrective feedback about their errors.
The major goal of peer-written corrective feedback is to foster an atmosphere of reciprocal learning
between learners based on the communicative approach to language teaching (Savignon, 2002). Ho and
Savignon (2007) also reported that numerous studies have explored aspects of peer-review activities,
including learners reactions and negotiation patterns, the effects of peer response on learners writing,
and learners attitudes and affective benefits. Studies conducted to identify the effect of peer-written
corrective feedback on learners' writing abilities (Elola & Oskoz, 2010; Ho & Savignon, 2007; Ware &
ODowd, 2008) reported that it is quite helpful for learners to focus more on structure and organization,
as well as content because they work collectively to address different types of errors in a cooperative
environment. Finally, this study focused on the effect of asynchronous, computer-mediated, peer-writing
corrective feedback. "One main benefit of CMC in asynchronous writing environments is that it promotes
collaboration amongst students and reduces inhibition towards communication" (Caws, 2006, p. 20).
Other studies made comparisons between presenting computer-mediated corrective feedback
cooperatively and individually. For example, Elola and Oskoz (2010) made a comparison between
learners peer-writing performance and individual writing performance using wikis and chats. Findings
showed that there were no statistically significant differences in terms of complexity, accuracy, and
fluency when comparing the individual and collaborative assignments. Finally, another study (Arnold,
Ducate, & Kost, 2012) compared students' revisions only (cooperation) and their revision for the essays as
a whole (collaboration). The results indicated that learners used both cooperative and collaborative
strategies to revise for form. However, they were more cooperative when they made content revision.
Finally, peer-revision computer-based collaborative writing created an environment which helped learners
to state their views and to reorganize their essays (Oskoz & Elola, 2011), to contribute to autonomy
through providing opportunities for practicing language (Kessler, 2009), and to develop their writing
skills (Lee, 2010).
Error Type
There are two major different error types that address both content and form, including focused and
unfocused (Ellis, Sheen, Murakami, & Takashima, 2008), treatable or rule governed (e.g., article use) and
untreatable (e.g. lexical use) (Ferris, 2002), or targeted and untargeted (Storch, 2010) error types.
According to Van Beuningen (2010, p. 11), focused corrective feedback "targets a (number of) specific
linguistic feature(s) only" while unfocused corrective feedback "involves correction of all errors in a
learners text, irrespective of their error category." Despite the fact that some studies (Sheppard, 1992)
focused on classifying error types into global vs. local, other studies (Arnold, Ducate & Kost, 2009;
ODonnell, 2007; Vyatkina, 2011) provided more detailed types, including content, structural
79
organization (text level), structural organization (sentence level), grammatical accuracy, lexical
appropriateness, punctuation, and spelling. Moreover, research has also explored the effect of corrective
feedback on separate error types. Ferris (2006), for example, discerned between five major error types
related to verb, noun, article, lexicon, and sentence, while Bitchener, Young, and Cameron (2005)
distinguished between three error types (past simple tense, articles, and prepositions).
Research about the effectiveness of corrective feedback proposes focusing on global error types rather
than local errors since they impair communication (Hendrickson, 1978). However, other studies
recommended focusing on both content and form, and different types of feedback should be used to
address both content and form (Ferris, 2003, 2006; Hyland & Hyland, 2006; Lee, 2010; Sheppard, 1992;
Vyatkina, 2011). This may help learners focus on both local error correction and global revision related to
different writing aspects, including content, organization, and form. Thus, based on these studies, the
evidence in support of writing corrective feedback on a limited number of error types is not very strong.
Moreover, research did not generally support conducting studies based on a limited number of errors. For
example, Ferris (2010) noted that the purpose of error correction is to improve writing accuracy in general
and not just to use one grammatical structure, though exploring the effect of focused error types might not
be functional in improving learners' performance in writing from a practical perspective. Similarly, Storch
(2010) demonstrated among other things that it is questionable whether we can draw generalizations
about the efficacy of writing corrective feedback based on the evidence from such a limited range of
structures. She added, "Studies which provide feedback on one type of error and only on one piece of
writing and in controlled environments are unlikely to be relevant to language teachers" (p. 43). Other
studies (Storch, 2010; Van Beuningen, 2010;) recommended provision of feedback within a realistic
writing context to enable valid assessment of the role of corrective feedback in language-learners
written-accuracy development. That is, learners provide corrective feedback on different error types as
they learn how to write.
Studies (DeKeyser, 2007; Ferris, 2004) suggested that the administration of extensive and sustained
meaningful exposure and practice is effective in language learning and recommended the incorporation of
feedback on content even at earlier stages of language instruction. This study is concerned with this issue.
It provides a writing context in which students' writing performance was investigated over eight weeks to
enable a valid assessment of the role of corrective feedback in language-learners written-accuracy
development and in which the writing aspect was mainly developed by computer-mediated corrective
feedback. The marking scale used by Vyatkina (2011, p. 73) related to the seven writing aspects was
adopted. Appendix A contains this studys operational definitions for each of the seven writing aspects,
examples about them, metalinguistic feedback, and reformulation of the error.
Computer-mediated Corrective Feedback
Though many studies have face-to-face comparison for certain types of corrective feedback over others
(Russell & Spada, 2006), there is a limited number of outcome-based studies on peer-writing corrective
feedback in computer-mediated communication (CMC) tools like using wikis and chats (Elola & Oskoz,
2010), and Word (AbuSeileek, 2006). Some studies (Caws, 2006; Duff & Li, 2009; Ho & Savignon,
2007,) have focused on investigating the effect of computer-mediated corrective-feedback method using
WebCT, email and Word on participants' attitudes, and reported that it was quite helpful for the
development of students' writing. Lee (2005) also demonstrated that web-based learning reinforced
learners' communication skills and writing, and actively involved them in being responsible for their
learning. Other studies (Ho & Savignon, 2007; Savignon & Roithmeier, 2004) found that computermediated peer review based on online-text discussion offers flexibility for learners as it reduces
psychological pressure on learners who do not like to give feedback in face-to-face situations because it
allows them to offer remote feedback asynchronously at the learners' convenience.
Furthermore, other researchers compared between computer-mediated corrective feedback and traditional
80
feedback. For example, Yeha and Lob (2009) developed an online corrective feedback and error analysis
system called Online Annotator for EFL Writing. The system consisted of five facilities: Document
Maker, Annotation Editor, Composer, Error Analyzer, and Viewer. The findings indicated that students'
corrective feedback revealed significantly better performance in the experimental group on recognizing
writing errors. Moreover, Liu and Sadler (2003) also explored whether differences in both modes of
commenting (Microsoft Word) and interaction (MOO) were more effective than traditional instruction in
writing. The findings show that the overall number of comments, the percentage of revision-oriented
comments, and consequently the overall number of revisions made by the technology-enhanced group
were larger than those made by the traditional group. However, the study suggested using Microsoft Word
2010 for editing in an electronic peer review mode, combined with face-to-face interaction in the
traditional peer-review mode, for effective peer-review activities in L2 writing classrooms. Moreover,
wikis were found to be an effective educational tool that encouraged the revision behavior and
collaborative writing skills, and collaborative electronic writing may enhance more revisions (Arnold,
Ducate, & Kost, 2009).
In conclusion, the number of computer-mediated studies examining the effect of corrective-feedback
types was too limited to argue for the advantage of this method compared with the traditional method or
type of corrective feedback, though many studies showed that CMC was very helpful for developing
learners' performance in writing (AbuSeileek, 2012; Sotillo, 2000; Ware & Warschauer, 2006). The
present study aims to introduce a new corrective-feedback type of track changes that may be different
from the other types used in previous studies. Using this method, the learner can strike through deletions
and mark new insertions in the text in a different color where the error is identified indirectly and
reformulated. This makes using track changes a combination of both implicit and explicit correctivefeedback types. Furthermore, the study used different feedback types to address both content and form,
including focused and unfocused error types. It sought to determine the effect of corrective-feedback type
on EFL learners' performance in different writing aspects, including content, structural organization (text
level), structural organization (sentence level), grammatical accuracy, lexical appropriateness,
punctuation, and spelling. It seemed that no studies have investigated its effect on developing learners'
proficiency in different writing aspects, including correcting local and global errors over a long period
using the track changes method. Consequently, this was a major goal of this study.
THIS STUDY
1. This study investigated the effect of using peer-generated computer-mediated corrective
feedback on EFL learners' performance in writing. It also explored how three types of
computer-mediated corrective feedback (track changes, metalinguistic, and recast) support EFL
writing. More specifically, it sought to answer the following three research questions:
2. Do learners who receive computer-mediated corrective feedback from their peers perform
significantly better when writing post-tests than those who do not receive feedback?
3. Which type of computer-mediated corrective feedback (track changes, recast, or metalinguistic)
is more effective for EFL learners' writing performance?
4. Which writing aspect (content, structural organization (text level), structural organization
(sentence level), grammatical accuracy, lexical appropriateness, punctuation, and spelling) is
mainly developed by computer-mediated corrective feedback (track changes, recast, or
metalinguistic)?
Participants and Design
Sixty-four undergraduate participants were enrolled in a writing course in a university. The course aimed
at improving students written English communication. This included improving learners' ability to
generate, organize, compare, contrast, and develop ideas supported by examples or evidence. Students in
81
this class were freshmen in the Department of English. A survey was done for their age, gender, and
language capabilities. Their age range was 18.319.1 years. The sample consisted of 48 females and 16
males. However, the four study groups were balanced in their composition with twelve females and four
males assigned to each group.
A score range between 68 and 75/100 in the universitys English Placement Test is considered as an
intermediate level and is roughly equal to level 5 on the IELTS, FCE, and CCSE, level 2 on the
Cambridge exams, AP according to the ARELS exams, and OP according to the Oxford exams. Based on
their scores, the 64 participants were placed in the intermediate linguistic level. They were randomly
assigned into four groups (one control and three experimental) of sixteen students each. The MANOVA
which was run on pre-test scores found no statistically significant difference between the computermediated groups (N = 48; mean = 24.19; standard deviation = 1.93) and the control group (N = 16; mean
= 23.63; standard deviation = 1.79); F(1,63) = 1.06, p =.31. The MANOVA test also showed that there
was no significant effect for computer-mediated corrective-feedback type, F(2,47) = 1.06, p =.31, on the
writing pre-test for all groups: track changes (N = 16; mean = 24.19; standard deviation = 2.26), recast (N
= 16; mean = 24.38; standard deviation = 1.82), and metalinguistic (N = 16; mean = 24.00; standard
deviation = 1.79). This indicates that all four groups began with similar levels of writing ability.
The same instructor taught all of the participants in this study. Moreover, all students studied the same
material and wrote eight essays over the same period: eight sessions over eight weeks. Students were
informed that they were participating in a study that was being conducted solely for academic purposes.
They were also informed that their participation in the study would not affect their evaluation in the
course. To reduce the possibility of students' sharing what they were doing in each group, they had access
to the materials and used the computer-mediated techniques only during the lessons. Due to the absence
of some students, the instructor was forced to postpone two lessons (one for the track changes group and
the other for the metalinguistic group) in order to enable all students in different groups to participate in
the discussions.
Test
For the purposes of this study, a test was designed by the researchers to gauge the effect of the corrective
feedback-type on EFL learners' writing abilities. It aimed at measuring students' ability to write correctly
in English, including the ability to generate, organize, compare, contrast, and develop ideas, and to
support those ideas with examples or evidence. Two raters, a native speaker and an EFL instructor, who
were applied linguists, validated the test with regard to the clarity of instructions and the suitability of the
questions to the goals and content of the course. They presented several suggestions such as changing the
marking scale. Their suggestions and comments were taken into consideration. The test consisted of two
main questions.
The first question required students to write an essay of 150200 words within a 30-minute time limit.
For this question, the marking scale included the following seven writing aspects: content, structural
organization (text level), structural organization (sentence-level), grammatical accuracy, lexical
appropriateness, punctuation, and spelling. Each writing aspect received ten points, with a maximum
question score of 70 (Table 2).
Table 2. Writing Scoring Rubrics
Question
Writing aspect
1. Write an essay
Content
10
Structural organization (text level)
10
Structural organization (sentence level)
10
Number of errors
Points
82
2. Underline and
correct errors
Grammatical accuracy
10
Punctuation
10
Lexical appropriateness
10
Spelling
10
Total
70
Content
10
10
10
10
Punctuation
10
10
Spelling
10
35
70
Total
The second question required each student to search and correct, also in 30 minutes, a writing sample that
had 35 errors covering the seven writing aspects. Each writing aspect received ten points, two points for
each error, with a maximum question score of 70 (Table 2). Each student was required to underline the
error and identify the type. The test re-test technique was used to establish the reliability of the test. The
test was administered twice to a pilot study of twenty students within a two-week period between the two
tests. The reliability coefficient of the test was .83, which is statistically acceptable for the purposes of the
study. Before the test, students were briefed about its components and the criteria to be used in assessing
their work. Each essay was scored by the same two raters who jointly assessed five of the essays and
discussed points of disagreement until a consensus was obtained. Each essay was then rated
independently by the two raters. Neither of them knew the rating assigned by the other. The inter-rater
reliability between them was .86, which was statistically acceptable for the purposes of this study. It
should be noted that the research questions in this study were addressed through the writing pre- and posttests. However, students essays were analyzed only to address the extent to which corrective feedback
was delivered for each writing aspect and corrective-feedback type.
Description of Procedures and Instructional Treatment
Students were surveyed about personal information. They were then assigned into four treatment groups
of 16 students each. Each of them was also randomly assigned into four small cooperative groups of four
students each. All the writing tasks, including providing the corrective feedback, were conducted in the
computer laboratory using Microsoft Word 2010 under the supervision of the instructor. Each student was
set in front of a computer. They had to write an essay of 150200 words about a specific topic. The topics
included describing a person, describing a place, the similarities/differences between two things, hobbies,
how to spend spare time, telling a story, a journey, and the Internet. Each student then submitted their
work to the instructor. To reduce students' anxiety about making mistakes and errors, they were instructed
not to write their names on the drafts; numbers were used instead. Each student in the small cooperative
groups wrote one essay in each session. The teacher then assigned each student in the group one essay to
be corrected. Each of them worked individually to provide corrective feedback related to the seven
writing aspects in the draft written by another student from the group. Each group then had face-to-face
discussion about the corrective feedback provided by each of its members about the errors in the draft
83
he/she had checked. This step was adopted to help them trust the corrective feedback from their peers
who were still developing their own writing proficiency (Paulus, 1999). Students did not directly receive
teacher corrective feedback about their errors. However, each group then checked the corrective feedback
they provided about the errors in the drafts with their instructor as an additional step to instill confidence
in the feedback. Finally, the essays with corrective feedback were brought back to the instructor who sent
each student his/her essay, a procedure that ensured confidentiality. Then each student worked alone to
revise and rewrite his/her draft, considering the corrective feedback he/she had received.
The instructor familiarized the participants in all groups with the target writing aspects. He also trained
them to provide different types of corrective feedback as well as strategies to avoid when providing their
own corrective feedback about the writing aspects. Four instructional treatments were included in the
present study: track changes, recast, metalinguistic, and no feedback. In the first treatment, students
commented on their peers writing by using the track changes feature of Microsoft Word 2010. From the
Review menu, the student used the option Track Changes. It allowed the learner to double-strike through
deletions and to insert changes and mark them in red without providing metalinguistic information. In the
second treatment, recast feedback, the student used the option New Comment from the Review menu.
Comments were displayed inline through the option "Show All Revisions Inline" which can be reached
through Show Markup and then Balloons in the Review menu in Microsoft Word 2010. Students had to
reformulate learners' errors related to the seven writing aspects without providing any metalinguistic
information about the error. The third group used the same comment format; however, each student in this
group had to provide metalinguistic information related to the seven writing aspects without
reformulating the error. Finally, the fourth treatment was the control group in which no corrective
feedback was provided by peers or the instructor. Each student worked alone on the writing task.
However, students in this group received feedback once the study was finished because it was a long time
for them not to receive feedback. All participants were post-tested, and their writing mean scores were
analyzed per the study variables. Table 3 shows the extent to which corrective feedback was delivered for
each writing aspect and corrective-feedback type. Based on the table, the amount of corrective feedback
each group received was comparable as there were no statistical differences in the mean error between the
corrective-feedback types as well as the writing aspects for each group (F(2,47) = 0.07); p = .13.
Table 3. Total Errors and Mean Errors for Corrective-feedback Types and Writing Aspects
Writing aspect
Corrective feedback type

Track changes
Recast
Total
errors
Total
errors
Mean
error
Total
Explicit
Mean
error
Total
errors
Mean
error
Total
errors
Mean
error
Content
68
4.25
73
4.56
72
4.5
213
13.31
58
3.63
49
3.06
62
3.88
169
10.57
152
9.50
147
9.19
157
9.81
456
28.50
186
11.63
195
12.19
190
11.88
571
35.70
Punctuation
167
10.44
154
9.63
168
10.50
489
30.57
81
5.06
86
5.38
79
4.94
246
15.38
Spelling
164
10.25
174
10.88
170
10.63
508
31.76
Total
876
54.76
878
54.89
898
56.14
2652
165.79
Statistical Analysis
Means and standard deviations were calculated for either a no-feedback control condition or a correctivefeedback treatment including one of the three types: track changes, metalinguistic, or recast. To find the
84
presence or absence of any effect of computer-mediated corrective feedback on EFL learners'

performance in writing, and explore how the three types of computer-mediated corrective feedback
supported EFL writing, a MANOVA test was administered to the overall scores of students who
participated in this study. Scheffe post-hoc tests were also conducted to compare any differences among
the different groups in corrective-feedback types. Descriptive statistics, including mean scores and
standard deviations, and the MANOVA test, relating to the seven writng aspects in the third question of
the study, were calculated. Post-hoc Scheffe multiple comparisons were calculated to find the effect of
corrective-feedback type on the seven writing aspects. The corrective-feedback types were the
independent variables, and the students mean scores on the writing post-test and writing aspects were
the dependent variables.
RESULTS
Corrective Feedback Types
Question number one sought to determine whether learners who received computer-mediated corrective
feedback from their peers while writing performed significantly better on the writing post-test than those
who did not receive feedback. Table 4 lists the means and standard deviations of the students' scores on
the post-test comparison of those means (F values). The data in Table 4 show that the overall mean of
students' scores in the experimental condition was significantly higher than that of the students' scores in
the control group on the writing post-test (F(1,63) = 27.65, p = 0.00). In other words, the MANOVA test
showed that the computer-mediated groups had significantly higher mean scores than the control group
who did not receive corrective feedback from their peers about their writing errors. Moreover, the
standard deviation is double in the group that received feedback, indicating that students in this group did
not receive homogeneous mean scores on the writing post-test. As the instructor worked to ensure that all
students had the same opportunity to participate in the writing sessions, except getting or not getting peer
corrective feedback, the main significant effect for corrective-feedback type in this study may be
attributed to receiving and giving computer-mediated corrective feedback by the sample while learning to
write (Table 4).
Table 4. MANOVA of Students Scores on the Post-test for Corrective-feedback Types
Corrective feedback
Mean
Standard
deviation
Computer-mediated feedback
48
49.04
4.37
27.65*
Control
16
41.44
1.93
Note: * p < 0.05.
Computer-mediated Corrective-feedback Types

Question number two required students to indicate which type of computer-mediated corrective feedback
(track changes, recast, or metalinguistic) was more effective for EFL learners' performance on the writing
post-test. Table 5 presents the results of a MANOVA of students post-test scores for the three computermediated corrective-feedback types.
According to the table, there was a significant effect for the computer-mediated corrective-feedback type
on the students' scores on the writing post-test (F(2,47) = 127.10, p = 0.00). Post-hoc comparison
(Scheffe test) shows that the track-changes treatment group significantly outperformed the other groups.
The recast group also significantly outperformed the metalinguistic group (p = 0.01) (Table 6). These
findings show the following order of the effect of computer-mediated corrective-feedback type on EFL
learners' writing performance on the post-test from strongest to weakest: track changes > recast >
metalinguistic. As the three treatment groups had the same treatment conditions and opportunities for
85
participation and commenting, except the different computer-mediated corrective-feedback type, it can be
concluded that the main significant effect here could be attributed to the feedback type. That is, the track
changes feedback type tends to have the most significant positive effect on EFL students writing,
followed by recast feedback, and then metalinguistic feedback type.
Table 5. MANOVA of Students Scores for the
Computer-mediated Corrective-feedback Type on
the Writing Post-test
Feedback
type
Mean
Standard
deviation
Track
changes
16
54.00
2.16
Recast
16
48.89
1.63
Metalinguistic
16
44.26
1.29
Total
48
49.04
4.37
Table 6. Results of Scheffe Test for

Computer-mediated Corrective-feedback
Type on the Writing Post-test
Feedback
type
Track
changes
127.10*
Recast
Metalinguistic
MD
Sig.
MD
Sig.
4.19*
0.00
7.94*
0.00
3.57*
0.00
Recast
Note: * p < .05.
Note: * p < 0.05
Effect of Corrective-feedback Types on Writing Aspects

The third question sought to determine which writing aspects the computer-mediated corrective feedback
(track changes, recast, or metalinguistic) developed the most. Table 7 shows the means and standard
deviations for the writing aspect by computer-mediated corrective-feedback type on the writing post-test.
Table 7. MANOVA of Students Scores for Computer-Mediated Corrective-Feedback Type on the PostTest in Writing Aspects
Writing aspect**
Corrective feedback type
F*
Track changes
Recast
Metalinguistic
SD
SD
SD
Content
8.06
0.93
7.00
0.73
6.25
0.58
23.02*
8.25
0.77
7.38
0.89
6.19
0.75
26.44*
8.25
0.86
7.06
0.68
6.13
0.62
34.47*
8.38
0.62
7.31
0.95
6.31
1.01
22.12*
Punctuation
8.38
0.62
7.50
1.10
6.38
0.62
24.53*
6.06
1.24
6.50
0.52
6.56
0.51
1.73
Spelling
6.63
0.50
6.13
0.81
6.44
0.63
2.36
Total
54.00
2.16
48.89
1.63
44.26
1.29
127.10*
Note: * p < 0.05; ** The maximum score for each writing aspect is 10.
According to the table, the MANOVA test revealed that students in the track-changes group obtained
more mean scores in all writing aspects except two, lexical appropriateness and spelling, at the p < .05
level. A post-hoc Scheffe test also indicated that there was a significant effect for writing aspect between
the track-changes group and the other two groups (recast and metalinguistic) in favor of the track-changes
group in all writing aspects, except lexical appropriateness and spelling (Table 8). This seems to indicate
that the track-changes feedback was more effective for learning the five writing aspects than the other
feedback types (recast and metalinguistic) on the writing post-test in the present study. However, the
Scheffe Test shows that there were no significant differences between corrective-feedback types (track
86
changes, recast, and metalinguistic) in lexical appropriateness and spelling (p = 0.01), indicating that
students in the three computer-mediated corrective-feedback groups in this study could develop the two
writing aspects (lexical appropriateness and spelling) in the same way in the writing post-test. The posthoc comparison (Scheffe Test) also shows that the recast group significantly outperformed the
metalinguistic group in all writing aspects except lexical appropriateness and spelling at the p = .01
(Table 8). This appears to suggest that the recast corrective-feedback type is more helpful for the
participants in learning the writing aspects than the metalinguistic corrective-feedback type. As the
instructor worked to ensure that participants in all treatment conditions had similar opportunities to locate
errors and to correct them except the different corrective-feedback types, it can be concluded that the
main significant effect related to the writing aspects could be attributed to the corrective-feedback type
from track changes.
Table 8. Results of Scheffe Test for Writing Aspect by Computer-mediated Corrective-feedback Type on
the Writing Post-test
Writing aspect
Recast
MD*
Content
1.06*
0.88*
Structural organization (sentence

level)
1.19*
1.06*
Punctuation
Spelling
.88*
-0.44
0.50
Metalinguistic
Sig.
0.00
0.01
0.00
0.01
0.01
0.34
0.11
MD*
Sig.
1.81*
0.00
0.75*
0.03
2.06*
0.00
1.19*
0.00
2.13*
0.00
0.94*
0.00
2.06*
0.00
0.94*
0.00
1.13*
0.00
1.13*
0.00
-0.50
0.24
-0.06
0.98
0.19
0.72
-0.31
0.41
Note: * The mean difference is significant at the 0.05 level.
DISCUSSION
Corrective Feedback Types
The findings of this study affirm that learners who received corrective feedback delivered via computer
about error types while writing essays performed significantly better than those who did not receive
corrective feedback. Providing computer-mediated corrective feedback by peers seemed to have enhanced
students' writing performance. This finding may be attributed to two reasons. The first is that the sample
of the study may need more help than other groups of learners, such as advanced learners or native
speakers, as it consisted of intermediate level EFL learners. Second, students in the treatment conditions
used the computer to provide corrective feedback about errors in their classmates' essays. Most likely, the
computer might support them to improve their writing performance. This finding has been supported by
other studies conducted to identify the effect of computer-mediated corrective feedback on both form and
content (Arnold, Ducate, & Kost, 2009; Liu & Sadler, 2003; Yeha & Lob, 2009). They found that the
corrective feedback delivered via computer was quite useful for the development of students' writing
87
abilities on post-tests. Similarly, Savignon and Roithmeier (2004) and Ho and Savignon (2007) found that
computer-mediated written corrective feedback has the ability to foster language learning and to help
learners in finding errors and correcting them.
The present study also lasted over eight weeks and included different writing aspects, focusing on both
content and form. Studies that explored the effect of corrective feedback on students' writing development
(Ferris, 2003, 2006; Hyland & Hyland, 2006; Lee, 2010) supported the findings of this study. They
recommended focusing on both the local and the global, or organization, grammar, and mechanics on the
one hand and content on the other. Vyatkina (2011) was also in agreement with this finding. She reported
that the majority of the participants felt that providing foreign language programs with corrective
feedback on different writing aspects, including both content and appropriateness and grammatical
accuracy, was quite useful; using wikis and chats in collaborative writing also allowed learners to
concentrate more on different writing components (Elola & Oskoz, 2010). Oskoz and Elola (2011) also
reported that computer-mediated communication facilities helped learners in refining the organization of
their essays, thus becoming better writers.
Computer-mediated Corrective-feedback Types
This study also investigated the effect of computer-mediated corrective-feedback type (track changes,
recast, and metalinguistic) on EFL learners' performance in writing. The findings of this study showed
that students who received recast feedback significantly outperformed those who received metalinguistic
feedback. Moreover, the results of this study revealed that students who received track-changes
corrective-feedback type obtained the highest mean scores compared with the other groups, indicating
that it is the most useful computer-mediated corrective-feedback type for developing learners' writing
performance on the post-test. This finding was supported by the research carried out to identify the effect
of track changes on the development of students' writing abilities. Ho and Savignon (2007), for instance,
concluded that most participants think that the track-changes technique is very convenient for providing
feedback about writing and facilitates the editing process. Caws (2006) also found that students felt that
using track changes in the written evaluations helped them to identify and analyze their errors.
This finding may be attributed to several reasons. First, track-changes corrective feedback has certain
advantages as it reformulates the ill-formed text, sentence, phrase, or word through double-striking
deletions without providing metalinguistic information about the incorrect form. It also marks insertions
in a red color, which reformulates the error and provides the correct form of the problematic
word/phrase/sentence. According to Ho and Savignon (2007), the major function of track changes is to
record any change in a written text, including notes, questions, insertions, and deletions. This may attract
the user's attention to the error. Furthermore, track changes is actually different than either the recast or
metalinguistic corrective-feedback types. It is different from the metalinguistic feedback which provides
the learner with metalinguistic information or comments about the error explaining the nature of the error
indicated and providing a reformulation indirectly. Track changes does not provide metalinguistic
feedback about the error; however, it allows for a direct reformulation of the error. Track changes also
differs from recast feedback in that the error is always repeated in the correct form and in the
reformulation. However, track changes allows for error identification and provides target-like
reformulation. The original ill-form produced by the learner was preserved so that he/she could make a
cognitive comparison and notice the difference between the error and the suggested correct form.
However, the same could not be said for recast corrective-feedback type because the learner's original
output (error) was deleted and no longer available. This made it not possible for him/her to make a
cognitive comparison. All of this makes track changes a unique corrective-feedback type; it has
distinctive characteristics making it different from other corrective-feedback types.
Finally, this study concludes that using the track-changes corrective-feedback type may narrow the gap
between explicit and implicit feedback. Many researchers think that there is a connection between
88
implicit and explicit knowledge bases (DeKeyser, 1998; Hulstijn, 1995), while others (Doughty &
Williams, 1998; Long & Robinson, 1998) see implicit and explicit knowledge as being separated and
adhere to an intermediate position between these two types. Track-changes corrective feedback can be
used in both manners: implicitly and explicitly. That is, there is no direct or metalinguistic feedback
showing that an overt error has been committed, so it is implicit. However, the error is identified
indirectly and reformulated, so it is explicit. Therefore, it is a learning/teaching method that is different
from both explicit and implicit feedback, having unique characteristics and advantages. This may explain
the high significant mean scores obtained by students who received feedback using this method. Thus, the
use of this method, which is based on technological innovations, may have prospective opportunities for
language learning in general and in the area of providing corrective feedback about errors while writing in
a more specific way.
Effect of the Corrective-feedback Types on the Writing Aspects
Students in the track-changes group significantly outperformed participants in other conditions in most
writing aspects related to both form and content on the writing post-test. Some studies supported this
finding. For example, Vyatkina (2011) found that most respondents provided feedback to intermediatelevel learners on certain writing aspects, including content, lexical appropriateness, grammatical accuracy,
organization, spelling, and punctuation. Students might find an opportunity in the corrective feedback
they received from other peers to find their errors and correct them. This finding also entails that
corrective feedback should cover comprehensive writing aspects, including content, organization, and
form. Restricting corrective feedback to include error correction related to treatable or focused errors may
be less helpful in aiding the development of learners' writing abilities. In this way, corrective feedback is
not provided in a real writing context, which focuses on developing all writing aspects concurrently. Van
Beuningen (2010) and Storch (2010) demonstrated that students might find it confusing when they
observe that some of their errors have been corrected while others have not been.
The findings of this study also showed that there was a significant effect in all writing aspects on the posttest except for two: lexical appropriateness and spelling. Students also obtained the lowest means in these
aspects. This finding was not expected. Despite the fact that students in the three treatment conditions
(track changes, recast, and metalinguistic) had the same opportunities to study the seven writing aspects
in the same way during the writing course, they obtained lower mean scores in these two writing aspects.
However, this finding may be attributed to the nature of errors related to these writing aspects that
students had to find and correct. Most likely, these error types were not focused. That is, students learned
to use certain lexical items, but this did not ensure that they learned to use other items because they were
different and had different lexical usages. Similarly, spelling errors were generally unfocused
(untreatable). Participants might learn the spelling of a number of words. However, this does not
necessarily show that they have learned the spelling of other new words compared to other learning
focused (treatable) grammatical aspects, such as the definite or indefinite article. The findings in the
tables indicated that there was actually an improvement in all students' mean scores on the writing posttest in lexical appropriateness and spelling. However, this does not show an established level of
significant effect among the three groups for these writing aspects.
CONCLUSIONS, LIMITATIONS, AND IMPLICATIONS
This studys unique contribution is to look at asynchronous peer-generated computer-mediated corrective
feedback and represents a bridge between computer-assisted language learning (CALL) or computermediated teaching methods and work being carried out on corrective feedback in language writing. It
yielded several major findings. First, students who received computer-mediated corrective feedback while
writing achieved better results in their overall test scores than the control subjects who did not receive
corrective feedback. Second, there was a significant effect for the track-changes feedback type when
compared with the recast feedback and metalinguistic feedback types. Third, students in the track-changes
89
group significantly outperformed those in the recast and metalinguistic group in most writing aspects. As
this may be the first study which has investigated the effect of track changes as a corrective-feedback type
while teaching writing, it is still early to claim that it is the best corrective-feedback type to be used by
learners, as the issue of how the different types of corrective feedback contribute to language learning
has been and still remains one of the most controversial issues in language pedagogy (Ellis, 2005, p.
214). Therefore, future research may be conducted using track changes to verify or refute the findings of
this study.
Even though providing feedback is still a controversial issue, corrective feedback is commonly used in the
classroom. Therefore, it is essential to continue investigating whether or not technology has further
implications for the creation of more efficient feedback because the increased use of technology for
feedback purposes has been less explored. Track changes is a computer-mediated corrective-feedback
type which has its distinguished features and can be described as both partially implicit and explicit, and
somewhat metalinguistic and recast in usage. It is a unique method as it is based on error identification
and reformulation in which the nature of error is provided indirectly without providing an overt indicator
about the error. Therefore, there is a need for conducting more studies related to using different computermediated corrective-feedback techniques and which also facilitates the integration of the track-changes
feedback type in the teaching and learning of writing, including different modes of commenting and
tracking such as Reviewing Panel, Balloons, and Show Markup, and Track Changes Options. It is also
essential to continue investigating whether or not technology is providing benefits when compared to the
absence of corrective feedback. As the current study has revealed that the presence of certain techniques
delivered via computer are more useful in supporting learners' writing performance than the absence of
corrective feedback, researchers and pedagogues may think of conducting more studies using other
computer-mediated corrective-feedback methods and techniques.
Another implication is related to studies about the effectiveness of corrective feedback using paper-andpencil. Researchers may think of conducting these studies differently. The need arises to compare
between the effectiveness of track changes and paper-and-pencil corrective feedback to find whether it
may aid learners in getting more efficient corrective feedback. Researchers may also think of using
computer-mediated corrective-feedback methods that focus on both content and form over several
sessions. The findings of this study may raise a question about the validity of the methodology and
findings of studies that have tested the effectiveness of corrective feedback about focused-error types, or
did not provide corrective feedback on unfocused-error types (see, for more information, DeKeyser, 2007;
Ferris, 2004, 2010; Storch, 2010; Van Beuningen, 2010). Such studies might not be conducted in a real
classroom context where students did not learn how to write; they just examined the effect of corrective
feedback on certain error types and ignored others. According to Van Beuningen (2010), conflicting
findings on the effect of corrective feedback on developing-learners' writing abilities could be due to
methodological issues and study design. Furthermore, in light of the push towards more distance
education and the use of online peer and group work in writing courses, this study may present a practical
model that can be used as guidance for instruction on the use of peer feedback in language-learning
contexts. Finally, this study was based on giving and receiving peer-generated corrective feedback. A
future study may compare between the act of giving versus receiving feedback and which of these is
responsible for the learning differences observed.
The findings of this study should be interpreted with caution for the following reasons. Firstly, the results
are limited to using a specific word processor: Microsoft Word 2010. Further studies are needed using
other word processors, including their different modes of commenting and tracking changes. Secondly,
the study was conducted on a limited sample (sixty-four learners) over a limited period (eight weeks) on
certain feedback types (track changes, recast and metalinguistic) in a particular situation. Therefore, there
is a need for other studies to be conducted on a greater number of students over longer periods and tracing
error type over time. Thirdly, the analysis was restricted to seven major writing aspects. Other studies
90
may be devised to measure different error types, such as local vs. global, or focused vs. non-focused, and
their subcategories using different corrective feedback types in the computer-mediated correctivefeedback environment. Finally, both the experimental and control groups were taught by one of the
researchers, which does not make it an ideal situation because the involvement of the researcher in the
teaching could introduce bias.
APPENDIX. Vyatkina's (2011) Classification for Writing Aspects and their Operational Definitions
Writing aspect
Definition
Example
Content
It includes irrelevance
content, long sections,
unsuitable examples,
redundancy, missing
content, senseless ideas
(illogical information), and
unbalanced discussion.
English
language
includes four
basic skills:
reading,
writing, and
listening.
Structural
organization
(text level)
Ideas follow each other in

a logical, coherent order at
the text level to make
sense to the reader. Errors
include the wrong use of
transitions, main sentence
in the essay, main sentence
in each paragraph, and
consistency between them
and other sentences, and
correct paragraph
transition.
English is very
important to
study in
schools and
universities.
Since you
have to speak
English well
these days if
you want to
get ahead in
your study.
Structural
organization
(sentence
level)
Grammatical
accuracy
Ideas follow each other in

a logical order at the
sentence level to make
sense to the reader. Errors
include the wrong use of
transitions, and connection
between words and
phrases, and ideas at the
sentence level.
It includes incorrect word
form or word order.
Punctuation
This is restricted to the

wrong use of punctuation
marks.
Metalinguisti
c feedback
Missing
content
Reformulation
Wrong use of
transition
English is very
important to
study in schools
and universities.
Since Therefore,
you have to
speak English
well these days if
you want to get
ahead in your
study.
I think you
Wrong use of
have a nice but transition
very nice
future if you
have good
English.
I think you have

a nice but very
nice future if you
have good
English.
English are the

mother
language of
the world.
Finally I think
you have a
very nice
future if you
have good
English.
Subject-verb
agreement
English are is the

mother language
of the world.
Use a comma
after the
transition
Finally Finally, I
think you have a
nice but very
nice future if you
have good
English.
English language
includes four
basic skills:
reading, writing,
speaking, and
listening.
91
Lexical
appropriatenes
s
It refers to using
inappropriate use of lexical
items.
You must
speak English
well if you
want to get
ahead in your
business.
Use should
to express
advice
You must should

speak English
well if you want
to get ahead in
your business.
Spelling
It is related to using wrong

spelling of words.
English
Language is
very important
in our life
these days.
Capitalizatio
n
English
Language
language is very
important in our
life these days.
ABOUT THE AUTHORS

Dr. Ali Farhan AbuSeileek is an associate professor at Al al-Bayt University. He has published papers and
designed several CALL programs for EFL learners. His major research interest is CALL and its
application in EFL teaching and testing, machine translation, and CALL program development.
E-mail: alifarhan@aabu.edu.jo
Awatif Abu-al-Sha'r is an associate professor in TEFL, the Dean of the Faculty of Education at Al-AlBayt University, Jordan. Her main concern is in the fields of TEFL; e-Translation, Curricula and Quality
Insurance Standards, CALL, and Language Corpora.
E-mail: awatifabualshar@yahoo.com
REFERENCES
AbuSeileek, A. (2006). The use of word processor for teaching writing to EFL learners in King Saud
University. Journal of King Saud University, 19(2), 115.
AbuSeileek, A. (2012). The effect of computer assisted cooperative learning method and group size on
EFL learners' achievement in communication skills. Computers & Education, 58(1), 231239.
Arnold, N., Ducate, L., & Kost, C. (2009). Collaborative writing in wikis: Insights from culture projects
in intermediate German classes. In L. Lomicka & G. Lord (Eds.), The next generation: Social networking
and online collaboration in foreign language learning (pp. 115144). San Marcos, TX: CALICO.
Arnold, N., Ducate, L., & Kost, C. (2012). Collaboration or cooperation? Analyzing group dynamics and
revision processes in wikis. CALICO Journal, 29(3), 431448.
Bitchener, J. (2008). Evidence in support of written corrective feedback. Journal of Second Language
Writing, 17(2), 102118.
Bitchener, J., & Knoch, U. (2009). The relative effectiveness of different types of direct written corrective
feedback. System, 37(2), 322329.
Bitchener, J., Young, S., & Cameron, D. (2005). The effect of different types of corrective feedback on
ESL student writing. Journal of Second Language Writing 14(3), 227258.
Canale, M. & Swain, M. (1980). Theoretical bases of communicative approaches to second language
teaching and testing". Applied Linguistics, 1, 147.
Caws, C. (2006). Assessing group interactions online: Students perspectives. Journal of Learning Design,
1(3), 1928.
92
Chandler, J. (2003). The efficacy of various kinds of error feedback for improvement in the accuracy and
fluency of L2 student writing. Journal of Second Language Writing, 12, 26796.
DeKeyser, R. (1998). Beyond focus on form: Cognitive perspectives on learning and practicing second
language grammar. In C. Doughty & J. Williams (Eds.), Focus on form in classroom second language
acquisition (pp. 4263). Cambridge, UK: Cambridge University Press.
DeKeyser, R. (2007). Skill acquisition theory. In B. VanPatten & J. Wiliams (Eds.), Theories in second
language acquisition (pp. 97113). Mahwah, NJ: Lawrence Erlbaum.
Doughty, C., & Williams, J. (1998). Pedagogical choices in focus on form. In C. J. Doughty & J.
Williams (Eds.), Focus on form in classroom second language acquisition (pp. 197261). Cambridge,
UK: Cambridge University Press.
Duff, P., & Li, D. (2009). Issues in Mandarin language instruction: Theory, research, and practice. System,
32(3), 443456.
Ellis, R. (2004). The definition and measurement of L2 explicit knowledge. Language Learning 54(2),
227275.
Ellis, R. (2005). Principles of instructed language learning. System 33(2), 209224.
Ellis, R. Sheen, Y. Murakami, M., & Takashima, H. (2008). The effects of focused and unfocused written
corrective feedback in an English as a foreign language context. System, 36, 353371.
Elola, I., & Oskoz, A. (2010). Collaborative writing: fostering foreign language and writing conventions
Ferris, D. (2002). Treatment of error in second language student writing. Ann Arbor, MI: The University
of Michigan Press.
Ferris, D. (2003). Response to student writing: Implications for second language students. Mahwah, NJ:
Lawrence Erlbaum.
Ferris, D. (2004). The "grammar correction" debate in L2 writing: Where are we, and where do we go
from here? (and what do we do in the meantime?). Journal of Second Language Writing 13(1), 4962.
Ferris, D. (2006). Does error feedback help student writers? New evidence on the short- and longterm
effects of written error correction. In K. Hyland & F. Hyland (Eds.), Feedback in second language
writing: Contexts and issues (pp. 81104). Cambridge, UK: Cambridge University Press.
Ferris, D. (2010). Second language writing research and written corrective feedback in SLA. Studies in
Second Language Acquisition, 32(2), 181201.
Ferris, D. (2012). Written corrective feedback in second language acquisition and writing studies.
Language Teaching, 45(4), 446459.
Ferris, D., & Roberts, B. (2001). Error feedback in L2 writing classes: How explicit does it need to be?
Journal of Second Language Writing, 10, 161184.
Gunette, D. (2007). Is feedback pedagogically correct?: Research design issues in studies of feedback on
writing. Journal of Second Language Writing, 16(1), 4053.
Hendrickson, J. M. (1978). Error correction in foreign language teaching: Recent theory, research and
practice. Modern Language Journal, 62(8), 387398.
Ho, M., & Savignon, S. (2007). Face-to-face and computer-mediated peer review in EFL writing.
CALICO Journal, 24(2), 269290.
93
Hulstijn, J. H. (1995). Not all grammar rules are equal: giving grammar instruction its proper place in
foreign language teaching. In R. Schmidt (Ed.), Attention and awareness in foreign language learning (pp.
359386). Honolulu, HI: University of Hawaii.
Hyland, K., & Hyland, F. (2006). Contexts and issues in feedback on L2 writing: An introduction. In K.
Hyland & F. Hyland (Eds.), Feedback in second language writing: Contexts and issues (pp. 119).
Cambridge, UK: Cambridge University Press.
Learning & Technology, 13(1), 7995. Retrieved from http://llt.msu.edu/vol13num1/kessler.pdf
Lee, L. (2005). Using Web-based instruction to promote active learning: Learners' perspectives. CALICO
Journal, 23(1), 139156.
Lee, L. (2010). Exploring wiki-media collaborative writing: A Case Study in an Elementary Spanish
Course. CALICO Journal, 27(2), 260276.
Liu, J., & Sadler, R. (2003). The effects and affect of peer review in electronic versus traditional modes
on L2 writing. Journal of English for Academic Purposes, 2, 193227.
Long, M. H., & Robinson, P. (1998). Focus on form: Theory, research, and practice. In C. Doughty & J.
Williams (Eds.), Focus on form in second language acquisition (pp.1541). Cambridge, UK: Cambridge
University Press.
Lyster, R., & Ranta, L. (1997). Corrective feedback and learner uptake: Negotiation of form in
communicative classrooms. Studies in Second Language Acquisition, 19, 3766.
ODonnell, M. (2007). Policies and practices in foreign language writing at the college level: Survey
results and implications. Foreign Language Annals, 40, 650671.
Oskoz, A., & Elola, I. (2011). Meeting at the wiki: The new arena for collaborative writing in foreign
language Courses. In M. J. W. Lee & C. Mcloughlin (Eds.), Web 2.0-based E-learning: Applying social
informatics for tertiary teaching (pp. 209227). Hershey, PA: IGI Global
Paulus, T. M. (1999). The effect of peer and teacher feedback on student writing. Journal of Second
Language Writing, 8(3), 265289.
Russell, J., & Spada, N. (2006). The effectiveness of corrective feedback for the acquisition of L2
grammar: A meta-analysis of the research. In J. Norris & L. Ortega (Eds.), Synthesizing research on
language learning and teaching (pp. 133164). Amsterdam, Netherlands: John Benjamins Publishing.
Santos, M., Lpez-Serrano, S., & Manchn, R. (2010). The differential effect of two types of direct
written corrective feedback on noticing and uptake: Reformulation vs. error correction. International
Journal of English Studies, 10(1), 131154.
Savignon, S. J. (Ed.). (2002). Interpreting communicative language teaching. New Haven, CT: Yale
University Press.
Sauro, S. (2009). Computer-mediated corrective feedback and the development of l2 grammar. Language
Learning & Technology, 13(1), 96120. Retrieved from http://llt.msu.edu/vol13num1/sauro.pdf
Sheppard, K. (1992). Two feedback types: do they make a difference? RELC Journal, 23, 103l10.
Sotillo, S. (2000). Discourse functions and syntactic complexity in synchronous and asynchronous
communication. Language Learning & Technology, 4(1), 82119. Retrieved from
http://llt.msu.edu/vol4num1/sotillo/default.html
Storch. N. (2010). Critical feedback on written corrective feedback research. International Journal of
English Studies, 10(2), 2946.
94
Truscott, J. (1996). The case against grammar correction in L2 writing classes. Language Learning, 46,
327369.
Truscott, J. (2007). The effect of error correction on learners ability to write accurately. Journal of
Second Language Writing, 16(4), 255272.
Truscott, J. (2009). Arguments and appearances: A response to Chandler. Journal of Second Language
Writing, 19(1), 5960.
Truscott, J., & Hsu, A.Y.-P. (2008). Error correction, revision, and learning. Journal of Second Language
Writing, 17, 292305.
Van Beuningen, C. (2010). Corrective feedback in L2 writing: Theoretical perspectives, empirical
insights, and future directions. International Journal of English Studies, 10(2), 127.
Van Beuningen, C., De Jong, N., & Kuiken, F. (2012). Evidence on the effectiveness of comprehensive
error correction in second language writing. Language Learning, 62(1), 141.
Vyatkina, N. (2011). Writing instruction and policies for written corrective feedback in the basic language
sequence. L2 Journal, 3, 6392.
Vygotsky, L. (1978). Mind in society. Cambridge, MA: Harvard University Press.
Ware, P. D., & ODowd, R. (2008). Peer feedback on language form in tellecollaboration. Language
Learning & Technology, 12(1), 4363. Retrieved from
http://llt.msu.edu/vol12num1/wareodowd/default.html
Ware, P. D., & Warschauer, M. (2006). Electronic feedback and second language writing. In K. Hyland &
F. Hyland (Eds.), Feedback in second language writing: Contexts and issues (pp. 105122). Cambridge,
UK: Cambridge University Press.
Wells, G. (1999). Dialogic inquiry: Toward a sociocultural practice and theory of education.
Cambridge,UK: Cambridge University Press.
Yeha, Sh., & Lob, J. (2009). Using online annotations to support error correction and corrective feedback.
Computers & Education, 52(4), 882892.
95

http://llt.msu.edu/issues/february2014/yoonjo.pdf

pp. 96117
DIRECT AND INDIRECT ACCESS TO CORPORA: AN EXPLORATORY

CASE STUDY COMPARING STUDENTS ERROR CORRECTION AND
LEARNING STRATEGY USE IN L2 WRITING
Hyunsook Yoon, Hankuk University of Foreign Studies
Jung Won Jo, Hankuk University of Foreign Studies
Studies on students use of corpora in L2 writing have demonstrated the benefits of
corpora not only as a linguistic resource to improve their writing abilities but also as a
cognitive tool to develop their learning skills and strategies. Most of the corpus studies,
however, adopted either direct use or indirect use of corpora by students, without
comparing the effectiveness between the two applications. This case study seeks to
develop new lines of inquiry by comparing the effectiveness and learning strategy use in
corpus-based writing revision. Four Korean EFL students used introspective and
retrospective research instruments in an investigation of the effects of corpus use on error
correction, error correction patterns, and learning strategy use between the two
approaches. While we caution about drawing a conclusion from this small case study, the
needs-based approach to corpus use in L2 writing was found to be effective for
restructuring the learners errant knowledge about language use. The approach drove
students to actively adopt cognitive learning strategies by performing as language
detectives. Different effectiveness and learning strategy uses were also observed relative
to the corpus use contexts as well as according to student proficiency levels. We also
found pedagogical implications, which will be discussed, in relation to the two different
corpus applications.
APA Citation: Yoon, H., & Jo, J. W. (2014). Direct and indirect access to corpora: An
exploratory case study comparing students error correction and learning strategy use in
L2 writing. Language Learning & Technology 18(1), 96117. Retrieved from
http://llt.msu.edu/issues/february2014/yoonjo.pdf
Received: August 21, 2012; Accepted: March 29, 2013; Published: February 1, 2014
Copyright: Hyunsook Yoon & Jung Won Jo
INTRODUCTION
An ever-expanding body of work has demonstrated the benefits of corpus-based activities in second
language (L2) writing pedagogy (e.g. Charles, 2007; Gaskell & Cobb, 2004; Gilmore, 2009; Kennedy &
Miceli, 2010; Lee & Chen, 2009; Lee & Swales, 2006; OSullivan & Chambers, 2006; Sun, 2007; Yoon,
2008; Yoon & Hirvela, 2004). Those studies have generally shown the positive effects of corpus use on
the development of students linguistic and rhetorical aspects of L2 writing. Given that the linguistic
domain often leaves a major challenge even for advanced L2 writers (Lee & Chen, 2009; Yoon, 2008),
corpus-based learning can provide learners with a valuable resource to deal with chronic linguistic
problems. Many corpus studies in L2 writing have exploited corpora as resources to educe feedback on
learners writing (Gaskell & Cobb, 2004; Gilmore, 2009; OSullivan & Chambers, 2006). For example,
Gilmore (2009) showed how learners are able to integrate corpus observations into the redrafting stage of
writing to improve the naturalness of writing.
In addition to the benefits of corpus use as a linguistic tool, corpus-based studies have also observed the
contribution of corpus investigation as a cognitive tool to develop learners thinking skills in the learning
process (OSullivan, 2007; Sun, 2003). The analysis of students corpus examples is widely known as
data-driven learning (DDL) (Johns, 1991). DDL is said to not only raise students awareness of
96
Hyunsook Yoon and Jung Won Jo
Direct and Indirect Access to Corpora
conventional language patterns in L2 but can also develop strategies for learning the language and thus
encourage inductive learning. OSullivan (2007) argues that DDL inevitably involves the focus on the
learning process, which enhances learners mental activity, cognitive abilities and metalinguistic
knowledge. With the growing recognition of corpora as a learning tool, there has been increased attention
to the integration of corpus activities into the language classrooms as part of L2 instruction.
Corpora can be accessed in the language classrooms in two different ways: students direct use of
concordancing software, i.e. computer-based activities, and the presentation of teacher-prepared
concordance data in handouts, i.e. paper-based activities. Researchers have used these respective labels
for the two techniques: hands-on concordancing (Cobb, 1997) and corpus-printouts (Stevens, 1991), hard
and soft version (Gabrielatos, 2005), direct and indirect consultation of corpora (Chambers, 2007),
deductive and inductive DDL (Creswell, 2007), teacher-corpus interaction and learner-corpus interaction
(Rmer, 2008), and teacher-led concordance-based activities and learner-centered corpus-browsing
projects (Mukherjee, 2006). For the sake of convenience, this study will label the techniques direct
corpus use and indirect corpus use.
In direct corpus use, learners have direct access to concordancers to find language rules for themselves. If
the concordances do not offer enough clues, learners can get more texts by typing another key word or
clicking on an additional button. The sheer volume of corpus studies has examined the direct use
approach. Those studies demonstrated the potential of corpus use in a wide variety of implementations in
writing classes, including students corpus use to improve their knowledge about common usage patterns
of words and to increase confidence in L2 writing (Yoon, 2008; Yoon & Hirvela, 2004), enhancing
students awareness of lexico-grammatical patterning and rhetorical functions in EAP writing classes
(Charles, 2007), writing revisions based on the corpus information (Gaskell & Cobb, 2004; Gilmore,
2009), evaluating apprenticeship in corpus use (Kennedy & Miceli, 2001; 2010), examining students
evaluations of, and changes in, the use of lexical and grammatical features in writing (OSullivan &
Chambers, 2006), for proofreading activities centering on students learning processes and strategies
using a web-based concordancer (Sun, 2003), using a web-based scholarly writing template to enhance
students genre-specific language use (Sun, 2007), and using direct observation methodology, i.e.
computer tracking, to examine learners use of corpus data (Prez-Paredes, Snchez-Tornel, Calero, &
Jimnez, 2011).
On the other hand, voices have raised concerns about the difficulties students encounter in the direct use
of corpora. St. John (2001) commented that especially lower-level students are challenged by the daunting
amount of concordance examples, even for frequent words, which can easily become too numerous and
even meaningless. Charles (2007) stated that teacher control can help students manage the onerous
quantity of data and help them to gradually develop a better sense of corpus use without affecting the
original meaning of discovery learning. Boulton (2010a) argued that without losing the essential
characteristics of DDL, the use of published materials can help DDL to reach a wider audience of
teachers and learners (p.44). In this respect, where DDL is a new concept to new learners using
computers, corpus work in the familiar paper format can make the activities more accessible to harvest
long-term learning benefits. In other words, indirect corpus use is a compromise in an attempt to
reconcile the extraordinary (DDL) with the ordinary (published materials) (p.43).
In indirect use of corpora, the teacher has access to a concordancer and prints out examples from the
corpus. Here teachers can edit the concordances that may be too difficult for the learners. Then the
learners work with these edited concordances (Bernardini, 2004; Tribble, 1997; Tribble & Jones, 1990). A
relatively fewer number of studies have been reported on the implementation of indirect use of corpora in
the classroom. Stevens' (1991) study was the first attempt to explore concordance-based vocabulary
exercises as a viable alternative to the traditional gap-filler. He found that students showed better
performance on concordance-based exercises in contrast to those deployed with gap-filler exercises.
Tian's (2005) experimental study of 98 Taiwanese university students determined the distinctive
97
effectiveness of DDL relative to learning tasks and proficiency levels. The study adopted paper-based
corpus activities that were considered more feasible with the large class size. The results found that the
DDL group outperformed the control group in their learning of grammar and syntactic features of news
headlines, except for learning word usage, while there was no significant difference in learning outcomes
between students proficiency levels.
In more recent studies of indirect corpus use, Boulton (2008, 2009, 2010b) conducted several sequenced
studies to test the effects of corpus printed materials in language learning. He was mostly interested in the
effectiveness of paper-based corpus materials especially for teaching and learning phrasal verbs to lowerlevel students (2008), linking adverbials (2009), and 15 language items (2010b). He consistently found
that indirect corpus outputs were more effective than traditional references such as bilingual dictionaries
and grammar manuals. Boulton (2010a) argued that printed materials provided the conditions for
individual exploration later on with the accompanying benefits of greater autonomy, learner
centeredness, and life-long learning (p.44).
However, indirect use of corpora also has come under criticism for its limited access to the corpus data.
We do acknowledge Boultons (2010a) point that use of paper-based materials can be a transitional step
to train learners to become successful hands-on corpus users, but the teacher-edited limited sample of data
can restrict learners own discoveries so-called serendipitous learning (Bernardini, 2000).
Furthermore, it is difficult to ensure the representativeness of the sample in terms of frequency in the
editing process (Gabrielatos, 2005).
While direct and indirect corpus use have distinctive advantages and disadvantages as discussed above,
most of the studies have examined either approach exclusive of the other in L2 instruction. Earlier,
Chambers (2005) called for a comparative study on the benefit of direct consultation of corpora by
learners as opposed to consultation of concordances provided by teachers (p.121). Nevertheless, Boulton
(2010a) still noted that no studies to date directly compare the benefits of hands-on corpus consultation
with those of prepared materials (p.25). We need those comparative studies to understand pedagogical
effectiveness in writing development depending on the different use contexts. If either method can
differentially affect how L2 is taught and learned, it would have implications for course design and
material development. Following Chambers' (2005) and Boultons (2010a) suggestions, thus, this study
aims to demonstrate both methods respective influences on students error correction in a writing class.
It is tempting to succumb to a technological-determinist view that the most advanced technology is the
most effective tool for learning, i.e. direct corpus use. However, a critical pedagogical issue is how
students process the corpus information, relative to corpus use type, to deal with their linguistic problems.
For example, Zimmerman and Martinez-Pons (1988) found a high correlation between learners use of
self-directed learning strategy and learning achievements. That is, using the appropriate learning strategy
can lead to higher learning outcomes. Conversely, in order to obtain successful learning experiences in
corpus pedagogy, there should be an understanding of learning strategy, which can cultivate the
motivation of active learners. Consequently the important questions to examine in corpus investigation
are the strategies learners use to analyze the corpus data, whether there is a more effective strategy, and
whether there are differences in strategies relative to student proficiency levels and the two corpus use
settings.
Only a few studies have investigated learner strategies in corpus use and concordance analysis (Kennedy
& Miceli, 2001; 2010; Sripicharn, 2004; Sun, 2003). Kennedy and Miceli (2001) conducted a detailed
qualitative analysis of how students progressed towards becoming independent corpus users in Italian
writing instruction. They identified four steps in learners corpus execution: formulating the question,
devising a search strategy, observing the data and selecting examples, and drawing conclusions. The
study found that students had problems in all the steps mainly due to the lack of knowledge in L2. Sun
(2003) used a think-aloud protocol to analyze the learning process and strategies used by three English as
98
a foreign language (EFL) students when accessing the corpus data to proofread texts with grammar
mistakes. She identified four factors that influenced learners' investigations and strategies: learners prior
knowledge of the language, cognitive skills, concordancer skills, and the teacher's intervention. Sripicharn
(2004) conducted an interesting study that compared strategies used in concordance investigations
between learners and native-speakers. The results showed that while the learners mainly adopted datadriven strategies such as concordance-based generalization and hypothesis testing, the native speakers
depended more upon their intuitive knowledge, and they generalized beyond the concordance output.
While those innovative studies are very instructive, it is critical to garner more insights into the sorts of
strategies learners use to comprehend and process the information in corpus-based contexts. Learning
strategies are specific actions taken by the learner to make learning easier, faster, more enjoyable, more
self-directed, more effective, and more transferable to new situations (Oxford, 1990, p. 8). The need for
teaching the student learning strategy, i.e. to learn how to learn, has been front and center in L2
pedagogical research since Rubins (1975) seminal investigation of good language learners behaviors.
More generally, learners' approaches are contextual; they employ specific strategies to approach their
specific task and to achieve their particular purpose depending on the learning environment in which they
occur (OMalley & Chamot, 1990; Oxford, 1990; 1996).
The notion of enhancing discovery or learner autonomy in learning strategies has played a pivotal role in the
pedagogical investigations of corpus-based approaches to language learning. Numerous analyses have found that
concordances can contribute to the development of students cognitive skills and learning strategies
(Cheng, Warren, & Xun-feng, 2003; Kennedy & Miceli, 2001; OSullivan, 2007; Sun, 2003). Many
researchers have argued that corpus-based activities give learners more control over their language
learning process, which in turn promotes inductive learning and learner autonomy (Chambers, 2005;
OSullivan & Chambers, 2006; OSullivan, 2007). Given the premises shared in the literature of learning
strategy and corpus pedagogy, situating corpus-based activities within the theoretical framework of
learning strategy research is wholeheartedly legitimate and relevant. Nevertheless, very few attempts have
been made to integrate corpus-based research findings into the general classifications of learning
strategies used in L2 education literature such as those developed by OMalley and Chamot (1990) and
Oxford (1990). This is the critical missing link that should be established for corpus-based learning in the
whole framework of L2 pedagogy. Chambers (2005), who asked for a comparative study between direct
corpus use and indirect corpus use, also calls for further study on the learner strategies used in corpus
consultation and analysis, and the teachers role in providing guidance (p.121). Studies such as these are
essential for directing instructors to teach students how to exploit corpora more effectively and to devise
corpus-specific strategy training for teachers.
In short, we designed this study to examine a) the sorts of results that learners achieve in error correction
depending on the corpus use context, i.e. direct or indirect, and b) to investigate the sorts of learning
strategies students employ in either context.
The study is guided by three main research questions:
1. Can corpus-based writing revision improve the students' grammatical and lexical accuracy?
2. What are students' error correction patterns relative to direct and indirect corpus use, and is
there any difference in the effectiveness of such patterns?
3. What learning strategies do the learners employ in direct and indirect corpus use?
This study is original in two novel attempts to a) explore possible differences in students correction
behaviors in concordance analysis relative to corpus use settings, and b) incorporate corpus-based
activities into the general framework of L2 pedagogy by analyzing learners cognitive processes in
concordance investigations in relation to the language learning strategy literature. Given its exploratory
nature, the study adopted a case study methodology in order to obtain in-depth insights into the topic of
99
investigation.
METHODS
Participants
Four freshmen EFL students were chosen to participate in this study from a mid-sized university in Korea
in accordance with their different levels and degrees of interest in English writing. These students
participated with a self-professed interest to improve their English writing skills. Table 1 provides an
overview of the learners background information considered relevant to this study.1
Table 1. Overview of the Participants
Students
Age Gender
Pre-test:
grammar
Pre-test:
lexis
Level of English
Interest in English writing
Young
20
20/20
18/20
High
Intermediate
Joon
20
19/20
17/20
High
Very high
Hyun
20
15/20
12/20
Intermediate
Very low
Min
20
14/20
12/20
Intermediate
High
The participants were two females, Young and Min, and two males, Joon and Hyun, all aged 20. Their
language proficiencies were determined by administering a pre-test consisting of 20 questions each for
grammar and lexis respectively, followed by a personal interview. Based on the results, Young and Joon
were considered high-level, and Hyun and Min were intermediate-level students.
The interviews showed that the participants regarded English writing as fairly difficult relative to other
English skills. They began studying English at ages 10 or 11, and were educated by rote methods focusing
on grammar and reading. However, their interest in English writing varied greatly. Joon and Min had a
higher interest because they desired to study abroad, make foreign friends, and explore American culture.
On the other hand, Hyun emphasized no interest in English as did Young, who had relatively good
writing skills. Additionally, none of the four participants knew about a corpus, thus we can assume that
lack of prior exposure would produce no spurious effects.
Data Collection and Procedures
The main data set of the study includes a collection of students writings, and introspective and
retrospective reports from the students to establish their writing strategies. Table 2 demonstrates the
overview of the procedures for the data collection.
Table 2. Procedure of Data Collection
Pre-meeting
Experiment
Interview
Pre-test
Pre-writing
Indirect corpus use

Five week in-class writing
Think-aloud protocol
Students learning journal
Post-meeting
Direct corpus use
Five week in-class writing
Think-aloud protocol
Students learning journal
Interview
Post-writing
The experiment took place over 10 weeks, from March 28 to May 30, 2011. At the beginning, we oriented
the students about the purpose of the study and corpus-based learning, and then trained them about how to
utilize a corpus. We chose a free online corpus source, Lextutor, because of its easy accessibility and
100
operation. Lextutor is among the most well-known Web-based programs (www.lextutor.ca), which has a
built-in concordancer. A variety of corpora such as Brown, BNC, and other small-scale corpora can also
be accessed through this Web site. Students basic command of the software did not seem to be at issue
because Lextutor is quite simple to use.
The participants then took the grammar and vocabulary pre-test to determine their general English
proficiency. The pre-test required students to select 10 correct grammar items and to identify 10
grammatical errors. The lexical portion required students to fill in 15 blanks with appropriate words and
to supply five synonyms to a word list. They were also asked to write an essay on the topic, Why do you
have to learn English in 21st Century? on which no feedback was given. Finally we conducted an
individual interview to gather personal information and to corroborate the pre-test results.
During the experiment, the class met weekly for one and a half hours. Each class consisted of three tasks:
error correction, writing, and reflection. The in-class writings were returned with errors underlined, and
the participants corrected the errors by using concordance examples for 40 minutes. The students were
then asked to write 150-word essays within 30 minutes without consulting peers or dictionaries. The inclass writing topics were limited to opinion essays about general issues, such as Where do you prefer to
live in the country or in the city? and What is the most important quality of a good reader? Finishing
their essays, the students reflected by writing about the lesson in Korean in their learning journals. This
lasted 10 to 20 minutes.
After each class, the second author, who worked as a teacher in the experiment, evaluated the students
in-class writing with a native speaker teacher. We worked together to identify grammatical and lexical
errors in each students writing. Indirect use of corpora was implemented for the first five weeks in order
to familiarize with the students to the concordance and to raise their DDL awareness. During the lessons,
error-underlined writings were returned to the students with five to 15 pre-edited concordance examples
related to their errors. The errors and concordances were numbered and given as a one-page handout. The
concordances for grammatical errors provided answers for correct usage, while the handout for lexical
errors provided examples of two or three synonyms relative to the target word (Figure 1).
Figure 1. An example of concordance-based feedback of lexical errors
101
The last five lessons were designed to develop student self-discovery. The sessions met at the multimedia
classroom so each participant could access a computer to use the concordancer. Students were given their
error-underlined writings and access to Lextutor to search concordances to investigate their errors for a
solution.
The final procedure of the whole experiment required the participants to repeat the pre-writing essay on
the same topic. Ten weeks were considered enough time to avoid the sensitization effects of the topic.
These samples were evaluated to compare error rates in order to assess improvement in grammatical and
lexical knowledge.
Students subjective feedback was gathered by using introspective and retrospective types of instruments.
An introspective instrument is a think-aloud protocol in which participants verbally report their thoughts
during process-oriented tasks. The think-aloud protocol allowed the researcher to identify learning
strategies in using the corpus. A practice session demonstrated to students how to verbalize their thoughts
so they could easily perform think-aloud tasks. Other subjective-feedback instruments deployed were
retrospective reports: students learning journals and interviews. Those instruments have the advantage of
not interfering with the writing process, though the data may be distorted because they reflect the best
recollection of the subject (Roca de Larios, Murphy, & Marn, 2002). The participants also wrote
thoughts in learning journals, and interviews gave them the opportunity to express their evaluation of the
lesson, their perceptions of corpus use and any attitudinal changes. These reports allowed the researcher
to note affective aspects as well as learning strategies they used.
Data Analysis
Pre- and post-writings were analyzed to investigate the students overall improvement of grammar and
vocabulary use. Accuracy was evaluated by measuring the frequency of errors per 100 words (Chandler,
2003). A reduced error rate in the post-writing was regarded as an improvement.
The 10 in-class writings and the students introspective feedback were analyzed to understand individual
progress and error correction patterns. Any differences between the patterns and any effects of error
correction were sought relative to the first five lessons of indirect corpus use and the last five lessons of
direct corpus use. The students grammatical and lexical errors were coded by referring to the
classification used by O'Sullivan and Chambers (2006).
Different types of feedback resources were deployed respective to the first five and the last five sessions.
Voice recordings were used and analyzed for the first five lessons, because listening to their voices was
sufficient for identifying the participants in indirect corpus use as they were working with pre-edited
concordance examples. Video recordings were available for analysis of the last five sessions, so the
researcher was able to watch the students operating concordancers while concordantly listening to their
think-aloud reports. One camera was installed per computer to access each students computer-screen.
The voice and video recordings were transcribed, and every learning strategy was identified by referring
to the categories that were adopted from OMelley and Charmots (1990) and Oxfords (1990)
classifications.
FINDINGS
The findings are categorized in the order of the three research questions: a) learners' overall improvement
of grammatical and lexical knowledge, b) error correction patterns in indirect and direct use of corpora,
and c) use of learning strategies according to the two corpus use contexts.
Overall Improvement of Grammatical and Lexical Knowledge
After exposure to direct and indirect use of corpora, the students' overall grammatical and lexical
knowledge increased, as evidenced by pre-writing and post-writing evaluations of the same writing topic.
102
Table 3 provides the results on the error rates in pre- and post-writing.
Table 3. Number of Errors in Pre- and Post-writing
Grammatical errors
Pre-writing
Young
Preposition
Post-writing
Joon
Hyun
11(18.6)
5(8.5)
Noun agreement
Min
Total(%)
Young
Joon
3
Hyun
Min
Total(%)
11(28.9)
4(10.5)
Verb form/mood
12(20.3)
7(18.4)
Article
15(25.4)
6(15.8)
4(6.8)
8(21)
2(3.4)
6 (10.2)
4(6.8)
Pronoun
Use of negative
Elision
1
1
Conjunction
Total
Words in writing
Rate of grammatical
errors (%)
Lexical errors
11
21
18
59(100)
136
159
169
131
595
12.4
13.7
6.6
Informal usage
Total
Words in writing
Rate of lexical errors
(%)
9.9
Pre-writing
Young
Word choice/
inappropriate
vocabulary
6.9
1(2.6)
0(0)
12
11
37(100)
137
152
161
140
590
5.1
4.6
7.5
7.8
6.3
Post-writing
Joon
Hyun
Min
Total(%)
Young
Joon
21(95.4)
1
8
136
159
169
131
595
3.7
Hyun
3
Min
3
Total(%)
12(100)
1(4.6)
3
2.2
0(0)
1
4.1
22(100)
0(0)
2
137
152
161
140
1.5
2.6
1.8
12(100)
2.1
590
2
The rate of grammatical errors decreased from 9.9 percent in pre-writing to 6.3 percent in post-writing,
and lexical errors decreased from 3.7 percent to 2.0 percent, illustrating that the learners' overall
knowledge of grammar and lexis increased.
In pre-writing, there were individual differences in the number of grammatical errors, but all learners
were weak regarding uses of article, verb form/mood, and preposition. There also were some differences
in the frequency of lexical errors because Joon and Hyun tried to use new vocabulary and various lexical
constructions. Despite the errors, their initiative and motivation is apparent in their effort to try novel
constructions on their own. This is evidence of self-directed learning.
After the experiment, the learners corrected many errors in the pre-writing, especially the uses of article
and verb form/mood. Originally, the misuse of articles and verb form/mood accounted for more than 45
percent of the errors, but the students greatly improved by using concordance resources to make
corrections. In contrast, the rate of preposition and pronoun errors increased in the post-writing. In
interviews, the participants revealed they tried a variety of expressions using prepositions because the
more they wrote, the more they gained confidence. Again, as earlier argued, the commission of errors is a
positive sign of self-directed learning.
In summary, although learners made more mistakes for certain grammatical error types in the post-
103
writing, their grammatical and lexical knowledge increased overall. This comports to previous findings
that showed the positive role of corpus activities in error correction (Gaskell & Cobb, 2004; Gilmore,
2009; OSullivan & Chambers, 2006). We argue that students made more improvement as they made
more errors because they received immediate feedback from the corpus to correct errors. There is also a
cultural counterpoint given that East Asian pedagogies are based on rote learning methods following a
Confucian philosophical ethic. Students are expected to comply with the instructor who is considered the
master of the teaching material (Shi, 2006). However, corpus can be used to set students toward selfdirected strategies that this study shows are demonstrably effective.
Learners' Error Correction Patterns in Indirect and Direct Corpus Use
Our study begs the question about the effectiveness of indirect corpus use relative to direct use. This
section gets at the answer in regards to students error correction patterns relative to the two types of
corpus use. Table 4 provides correction patterns and rates for each student according to the two uses.
After 10 in-class-writings and corpus-based revisions, three major patterns were found. They are selfcorrection using concordance data (Pattern A), correction with teacher assistance (Pattern B), and no
correction (Pattern C).
Table 4. Correction Frequencies in Indirect and Direct Corpus Use
Indirect Use
Pattern A:
Self Correction
Young
Pattern B:
Correction
w/teacher
assistance
5/5
(100%)
Pattern C:
No
Correction
Pattern B:
Correction
w/teacher
assistance
4/4
(100%)
30/34
(88%)
(0%)
19/24
(79%)
5/5
(100%)
(0%)
25/39
10/26
(64%)
(38%)
14/14
16/16
(100%)
(100%)
(100%)
(100%)
36/59
12/25
(61%)
(48%)
23/23
13/13
(100%)
(100%)
5/5
6/6
(100%)
(100%)
34/66
7/17
(52%)
(41%)
32/32
10/10
(100%)
(100%)
(87%)
26/26
(100%)
125/198
(63%)
73/73
(100%)
(66%)
27/32
(84%)
48/92
(52%)
44/44
(100%)
Ga
35/40c
(88%)
Lb
13/18
(72%)
4/5
(80%)
0/1
G
L
32/38
16/28
(84%)
(57%)
6/6
8/12
(100%)
(67%)
0/4
Hyun
G
L
56/66
18/27
(85%)
(67%)
10/10
9/9/
Min
G
L
52/57
14/20
(91%)
(70%)
Total
175/201
61/93
Joon
Direct Use
Pattern A:
Self Correction
0/5
(0%)
Notes: * a: G Grammar, b: L Lexis, c: number of correction/number of errors
Overall, the rate of self-correction was higher in the indirect use than in the direct use. In the indirect use
of corpora, the learners corrected 87 percent of the grammatical errors and 66 percent of the lexical errors
by using concordance data on their own (Pattern A). The learners could not correct the rest of the errors
by themselves, so they asked for the teachers help. The teachers intervention helped the learners correct
most of the errors successfully (Pattern B). However, not all students followed the teachers advice. A
pattern emerged in which these learners rejected the suggested corrections and deliberately continued
their lexical errors (Pattern C).
On the other hand, the self-error correction rate was lower in the direct corpus use than in the indirect use
for most learners (Pattern A). When they could not find the relevant examples, students were able to
correct all their errors with the help of simple interventions by the teacher (Pattern B).
104
Reproduced in italic below are examples of students writing with errors underlined, as well as
introspective thoughts (translated from Korean to English) within quotation marks following think-aloud
protocol.
*butter, sugar and white powder: I didn't know the word for 'flour', so I just wrote white
powder. According to the examples, 'flour' is the correct word Im trying to say. (Min, third
composition, indirect corpus use)
*nonhuman lifestyle in Korea: In Lextutor, I can't find any examples of 'nonhuman'. I want to
say it is not human. Is it unhuman? inhuman? or not-human? Well, I will type containing
'human', then I can find negative prefix... Ah, I think [humane] is the right word Im trying to
say. I will type containing [humane] again. (Joon, seventh composition, direct corpus use)
Regardless of direct or indirect use, the students did not need to look over all the relevant examples
because the resources were understandable and compatible with their knowledge. In the case of indirect
use (1), Min successfully analyzed the error with translation though she did not know the word flour.
She was able to correct the error with only a few examples. In the case of direct use (2), Joon typed
human and detecting the relevant examples, he realized that he really wanted the word humane. He
returned to the start-up page to look up humane, after which he found the negative prefix to correct the
error. Students generally took much longer to correct errors in the direct use of corpora, but while
discovering examples, they seemed to naturally acquire new language information. The corpus
investigation enabled the learners to experience serendipitous learning (Bernardini, 2000) as well as
giving them opportunities to develop their cognitive skills (OSullivan, 2007).
Although direct corpus use can promote self-directed learning, students did need the teachers assistance
to find examples and correct errors. Below is a case of how Min described how she negotiated an error:
*In young people case, they are willing to stay: Is it a word choice error of young people, or
a grammatical error? Which word do I need to type in the concordancer? There is usually a
possessive form between in and case. Is that right? (Min, ninth composition, direct corpus
use)
As a result of the teachers scaffolding, Min found the right examples. The teacher typed in the clue case
preceded by in to the left side, so Min was able to analyze it correctly. She also found more examples
for in the case of than in + possessive + case, and thus she correctly changed it to in the case of young
people. While the teachers guidance was also helpful in the indirect use, a certain degree of teachers
intervention is also crucial in the direct use, which challenges a popular techno-deterministic assumption
that computers will replace teachers in the classroom. In contrast, we observed that computer use requires
continued teacher vigilance and attention for learners successful experiences with corpus activities
(Chambers, 2005; Sun, 2003).
Like Min, the other learners often asked for the teachers help, but Young, the highest level student, spent
much time exploring examples herself and did not need as much from the teacher. Young became very
skilled in scanning. She was in no hurry to find the answers, and she was able to winnow appropriate
examples among excessive data. This single case cannot be generalized, but it begs further exploration to
establish if direct corpus use appeals to learners who are patient, analytic, and fast enough to read and
scan the data.
Other interesting findings concern lower-level students successes in using both types of corpora. The
lowest level student, Min, got the highest rate of self-correction (91 percent) for grammatical errors in
indirect corpus use, while exhibiting the lowest correction rate (52 percent) in direct corpus use. Another
lower-level student, Hyun, showed a similar pattern: a higher rate of self-correction comparable with
higher-level students in indirect corpus use. We cannot generalize from these few cases, but it may be
105
worth exploring whether lower-level students benefit more from the teacher-edited corpus materials.
Although high-level students analyzed errors successfully with pre-selected examples in indirect corpus
use, we observed that they were not easily convinced if it conflicted with what they thought was a
standard grammatical rule. In these cases, a rule was overapplied to a special case that required a different
answer. The learners prior knowledge required the teachers assistance to make new generalizations to
teach new knowledge. An example is Youngs error and what she described:
English as the second language: According to the concordance data, it tells me 'as the' should be
corrected to 'as a', right? But while writing, after I had thought of selecting the usage "as a", I
came up with an idea that I have to put "the" to the left side of the ordinal number. Thats the
reason I wrote 'as the' instead of as a. I need your explanation rather than these examples.
(Young, first composition)
Young was not persuaded when the teacher gave her seven examples including an idiomatic expression,
as a. Young needed the teacher's protracted and insistent explanation why it was wrong. The teachers
persistence finally convinced Young to correct the error. This case reflects why the rate of Pattern A for
high-level students was lower in indirect corpus use. High-level students tended to defer to their prior
knowledge even when it conflicted with the concordance examples. In contrast, a lower-level student,
Min, satisfactorily corrected her errors based on the examples. This case is hard to generalize, but it is
worth further investigation about the effectiveness of indirect corpus materials for lower-level students
relative to higher-level students. This can be related to the previous finding that not all users adopted datadriven strategies in concordance-based investigation (Sripicharn, 2004). The present study found a similar
pattern in high-level learners as did Sripicharn, i.e. native speakers tended to rely upon their existing
knowledge.
Interestingly, high-level students only exhibited their resistance to corrections (Pattern C) for lexical
errors in the indirect corpus use. All learners corrected grammatical errors after receiving the concordance
feedback and the teachers help, but some learners wanted to leave lexical errors as they chose. Joon
especially was not afraid of creating words. He deliberately left 33 percent lexical errors thus resisting the
teachers explanation:
*not-well planned building arrangements: According to the data and your explanation, it is more
natural to change from not-well to 'poorly,' right? But I want to emphasize the negative
meaning. 'Poorly' doesn't fully cover my intention. It is not grammatically wrong, so I want to put
it this way. (Joon, fifth composition)
Although Joon analyzed the error successfully, he stuck to his original expression. He commented in the
post-interview:
I like creative expressions or using brilliant words. The concordance just has expressions many
people use, so there is no way to find creative uses of words. Different expressions are not wrong
usage, I think. (Joon, 5/31/2011)
Joon was upset that his novel expressions were found to be incorrect according to the concordance data.
In contrast, students who directly accessed the concordancer were agreeable to the usage as a result of
their own search through the data. Given the few numbers of participants, it is difficult to draw a firm
conclusion from this study. But in this particular case, the proficient students rather easily changed their
preconceived errant knowledge when they pursued the data through direct corpus access, implying the
effectiveness of direct corpus consultation for high-level students with conflicting errant knowledge.
106
Finally, although most participants were more successful in correcting errors for themselves in the
indirect use, they overall preferred the direct use of corpora. The participants were much interested in
finding the right answers with their own efforts, and they were very delighted that they could actually
discovered the rules. Below are excerpts from the interviews in the post-meeting.
Although it takes long to find examples, I prefer the direct use of Lextutor. As there is not only
one way to express ideas in language use, I can find a variety of different answers during the
process of searching this and that. Also I want to use it in writing later. (Young, 5/31/2011)
When I started directly using the concordancer, I fumbled around. But with the teachers little
help, I could have fun learning English and not be bored by using the computer. I prefer the direct
way. (Min, 5/31/2011)
This section demonstrated different error correction patterns and effects in corpus-based revisions
between the indirect and direct corpus uses. While the ratio of self-correction was higher in the indirect
use, most learners were more interested in the direct use as they liked to be a language detective. This
implies that students interest and motivation are not always reinforced by positive learning experiences.
While cautious in making generalizations from this case study, we found that direct corpus use seemed to
suit high-level students; they willingly changed their knowledge only when checking data on their own
efforts. On the other hand, indirect corpus use was satisfactory enough for lower-level students even when
the limited number of examples conflicted with their knowledge.
Learning Strategies in Indirect and Direct Corpus Use
As mentioned, we employed think-aloud protocols, learning journals, and observation notes to collect
qualitative data on student learning strategies through corpus use. Each learning strategy was identified,
classified into four main categories, and further sub-categorized. We identified different learning strategy
use patterns relative to the corpus use techniques and the participants English proficiency levels. Table 5
shows the frequency of learning strategies for each category.
Table 5. Classification of Learning Strategies Used in the Two Corpus Use Contexts
Category
Metacognitive strategy
Cognitive strategy
Affective strategy
Social strategy
Others
Total
Detailed strategy
Self-evaluation/monitoring
Making use of materials
Association
Grouping
Translation
Note-taking
Lowering anxiety
Self-encouragement
Question for clarification
Indirect use
Frequency
28
217
112
41
57
2
7
12
78
15
569
(%)
(4.9)
(38.1)
(19.7)
(7.2)
(10.0)
(.4)
(1.2)
(2.1)
(13.7)
(2.6)
(100)
Direct use
Frequency (%)
21
(2.5)
378
(45.5)
61
(7.3)
64
(7.7)
78
(9.4)
6
(.7)
16
(1.9)
12
(1.4)
187
(22.5)
7
(.8)
830
(100)
Learners adopted learning strategies appropriate for the particular task in the new learning situation. This
is a very personal experience to the learner who finds novel and individual ways to improvise strategies,
fortunately giving us an intimate insiders window to how the learner adjusts to the material. The total
frequency of strategy use, depicted in Table 5, is much higher in the direct corpus use, implying that
107
learners used an array of cognitive and affective skills to solve problems while engaging them in selfdirected activity.
Table 5 shows that cognitive strategy was the most frequently used main category, in both corpus
settings, which is not surprising given the nature of corpus activities. In indirect use, students used more
metacognitive strategy (4.9 percent) than in the direct use (2.5 percent), which seems be related to their
handling of novel type of learning. This appears to be evidence that participants seemed to coordinate
their own learning process (Oxford, 1990, p.136) more consciously at the first five weeks when faced
with a new type of learning, i.e. concordancing. But once they adjusted to the type of learning, they
moved to use other substantial strategies such as cognitive strategy. Of particular interest is that social
strategy was the second most frequently used learning strategy in both the corpus use settings. Clearly the
students needed the teachers guidance and mediation to continue their corpus analysis to solve their
language problems. This also conflicts with popular ideas that computer use reduces, and may even
eliminate, the need for teacher support in the classroom. If other studies uphold our finding, the evidence
will weigh in favor of a continued, active presence of teachers trained in corpus methodologies.
Now let us compare the use of the four sub-categories of strategy: making use of materials, association,
question for clarification, and translation. All of these, except for translation, were chosen because they
show the largest differences (over 5 percent) between the rates relative to the two corpus contexts. The
translation strategy was an unexpected result, because it was only a 0.7-percent difference. We had
anticipated a higher rate in the direct use where there are many more examples to translate from the
concordance outputs. The anomalous result was determined to be worth additional investigation.
Resourcing: Making use of materials
The adage, if at first you dont succeed, try, try again describes the making use of materials strategy.
In this strategy, the learners applied additional steps after their initial search process was unable to secure
answer. The frequency of this strategy varied between the direct (45.5 percent) and indirect (38.1 percent)
corpus use, which seems to be natural given the type of materials available in each setting. Indirect use
only made five to 15 corpus examples available. If the students were unsuccessful with this resource, they
moved on to other strategies to solve problems. In other words, they may or may not have been able to
solve the problem. In direct corpus use, if the learners failed to find appropriate examples, they returned
to the corpus start-up screen and typed another word to solve problems, rather than trying other strategies.
This is evidence that the resource itself is significantly important in the direct corpus use, and it implies
that the teacher can strengthen students learning skills by taking time to train them in concordancer
operations. It again exemplifies that the teachers role is critical to students successful corpus use.
Application of prior knowledge
When students could not correct errors using the initial resourcing strategies, they shifted to acquired
knowledge and logical reasoning. However, the frequencies and types of association with prior
knowledge differed between the two techniques. In the direct use (7.3 percent), the learners used the
concordancer until finding relevant data, rather than revert to their own knowledge. However, in the
indirect use (19.7 percent), students resorted to their prior knowledge because the data was too limited for
them to find answers.
Concurring with Sun (2003), we observed three manners of applying prior knowledge. First, when
learners produced the apt prior knowledge, they easily corrected mistakes and moved to the selfevaluation strategy (metacognitive strategy), while cogitating Why did I get it wrong? or I will never
make the same mistake next time. In this case, just a few examples persuaded them to correct errors, as
illustrated in the following writing example:
*..they live in the city, where is the center of the culture...(Young, 3rd composition)
108
Young said she already knew the grammatical difference between 'where' and 'which.' After using the
given example to satisfactorily correct the error, she said:
If I have a chance to produce a sentence including 'where' or 'which is' next time, I will
never make the same mistake. I already knew this grammar and I got it right on a multiplechoice exam before. Once making this mistake, I feel I didn't fully understand it before I
actually produced it.
Second, when the learners have incorrect prior knowledge, they could not correct errors for themselves,
so they tended to ask for the teachers intervention or for more examples to re-generalize. Here is an
example based on a writing sample:
* Less densed population will also remove stress (Joon, 3rd composition)
The teacher gave Joon eight examples where the word 'dense' is used as an adjective. While correcting the
error, he said:
In the data, every 'dense' word is used as an adjective. But I think it is used not only as an
adjective but also as a verb. I remember I wrote 'densed' before. I need more examples.
Can I look it up in the dictionary? Or can you give me an explanation?
Joons knowledge had become fossilized, i.e. solidified and impervious to change, even though the
concordancer supplied eight examples. It took the authority of a teacher's explanation to persuade him of
his errant knowledge.
The third case was when the learners had no relevant knowledge to the issue at hand. This
instance only occurred in direct corpus use. In this case, the teacher's intervention was necessary
to find the right examples:
* These phenomenons are dangerous...(Hyun,8th composition)
Hyun did not realize that phenomena is the plural form of phenomenon. Without the teacher's
intervention, the right examples would have eluded him because the concordancer feature would have
required him to type phenomena to access the correct usage. Or the learner could have found the plural
form phenomena by typing phenomen with the search option starts selected instead of equals in
the concordancer.
With limited data in the indirect use of corpora, the learners had to apply their prior knowledge when they
were not satisfied. This suggests that the number of examples can be a powerful attribute to compel
students to amend their errant knowledge in corpus-based contexts. Students fossilized knowledge defied
the influence of only a few examples. For indirect corpus use, then, it appears the teacher should consider
providing profuse examples when adopting the materials into classroom activities.
Translation strategy
The rate of translation strategy use differed little between direct (9.4 percent) and indirect (10.0 percent)
corpus use. This defied the researchers initial assumption that more differences would be apparent
because there were significant differences in the number of examples between the indirect (five to 15
examples) and direct method (maximum of 5,000 examples). Learners tended to first skim the data to
choose sentences they could easily understand or thought were most relevant to their errors regardless of
the number produced in the data. In both sorts of corpus use thus the number of translation strategy uses
was about the same.
A consequence of the indirect use was that learners lost interest in translation because the corpus cut off
the beginnings and ends of example sentences. Learners were confounded by the context because they
were forced to concentrate only on the target word without the whole sentence to convey meaning.
109
However, the direct use presented a different challenge that resulted in similar student indifference.
Learners were frustrated by the pressure to sort through the sheer volume of examples to translate. These
are technical issues in corpus presentation that need to be addressed for effective deployment of either
corpus type in the classroom. If corpus advocates seek to replicate a naturalistic learning environment,
then certainly the indirect use should reflect the natural written language that is likely to be produced in
full sentences. As for the direct use, it defies comprehension that a foreign speaker would ever need to
confront every form of language use in a native setting. On the other hand, perhaps more vigilant teacher
preparation is required to winnow through relevant corpus materials to select those appropriate to the
learning context that students would expect to face in the particular exercise.
Questions for clarification
The types and frequency of questions that students posed varied between the two corpus use settings. In
indirect corpus use, most learners corrected errors after using various strategies. It was only after they
failed to properly analyze the errors that they would finally ask for the teacher's explanation. Here is an
example:
* To achieve be satisfied life: Looking in the right pattern of achieve, there is always a
noun form. Isn't be satisfied life a noun? (Min, 5th composition)
After analyzing the data, Min did not recognize her errant knowledge. She then consulted the teacher who
explained the difference between the passive form and the noun form, which helped to repair Mins
knowledge.
On the other hand, the learners asked many questions in direct corpus use because they often needed the
teacher's simple scaffolding to search for right examples relevant to their errors. Once they found the
examples, they could solve most of their errors. Here is Joons error and what he described:
* I have lived in the metropolitan in Korea: Is it the error of grammar or the word
metropolitan? Do I look at the right or left pattern? (Joon, 9th composition)
In the process of finding relevant examples, students propounded many questions such as 'What type of
error is it?,' 'Do I have to look at the right side of the word or left side?,' but they did not tend to ask for
explanations about the answer.
The frequency of questions in indirect corpus use is less than in direct use because the questions are the
final method of correction, whereas the frequency is higher in direct corpus use as it occurred in the
process of finding examples (Figure 2).
110
Figure 2. The patterns of questions for clarification in corpus analysis

As the teacher was in charge of the steps to detect and find relevant examples in indirect use, the
students asked for explanations only when answers confounded them. In contrast, students had more
control over the data in direct use, so many questions arose while detecting examples. In the process, they
realized the nature of their mistake and how to correct it. Through this process they acquired language
information, and once they found relevant examples, they corrected most errors. This exemplifies that
direct use can engage learners more actively in the language searching, which in turn can lead to
development of learners cognitive skills and self-directed learning.
In addition to the findings regarding the application of corpus to the different contexts, the study also
found that the learners reacted differently in corpus analysis according to their English levels. The highlevel students, Young and Joon, occasionally defended their mistakes and explained their convictions for
writing that way. Joon was extremely strong-willed, so it was hard to reform his prior knowledge.
Whenever Joon realized he was wrong, he self-evaluated his prior knowledge by commenting Why did I
make this kind of mistake?
On the other hand, middle-level students Hyun and Min used question strategies as a way to trim their
anxiety and to elevate their confidence. They trusted the concordance data more than their preset
knowledge, so their questions were not defensive, but rather concerned their understanding about how to
find examples or to confirm the correction. Many researchers found that high-level students are more
prone to question and seek help from the teacher and their peers (Newman, 1991; Newman & Goldin,
1990; Zimmerman & Martinez-Pons, 1986). This single case study is difficult to generalize across the
board, but it does support the literature that middle-level learners tend to seek help from the teacher in
similar small-scale corpus classes, while high-level students were more willing to solve problems for
themselves with corpus materials.
In summary, this section demonstrated that learners had many opportunities to use a wide range of
learning strategies in concordancing activities. When their deployment of a certain strategy turned
fruitless, they appropriated other types of learning strategies as active language learners. Through the
process, they learned how to monitor their knowledge and to modify learning strategies to the situation.
Accordingly, the corpus consultation developed their metacognitive, cognitive, and socio-affective
strategy skills to a positive effect.
111
CONCLUSION
This case study investigated how learners corrected errors, and the strategies they employed, in L2
writing relative to indirect and direct access to concordance software. The indirect use involved teacherprepared concordance handouts supplied to students, while the direct approach gave students actual
access to the corpora. By integrating the corpus work into the revision stage of writing, students actively
found answers to correct their errors while testing out their linguistic hypotheses. This needs-based
approach (Aston, 1997; Braun, 2007) was especially effective for learners who made frequent linguistic
mistakes because the concordance data gave them instant feedback, which reinforced their learning.
Concordancing also served to develop students cognitive and metacognitive abilities by motivating
discovery learning, which enhanced autonomous learning (Johns, 1991; OSullivan, 2007). As OSullivan
(2007) argued, process-oriented corpus activities lead learners to monitor and regulate their cognitive
work by using a variety of learning strategies. As our study showed, learners can rework their writing and
process the linguistic input from corpora. To this effect, students acquired rules of language use and
restructured their prior, erroneous knowledge, which facilitated their acquisition of linguistic knowledge.
Consequently, corpus-based writing feedback can open a new way of learning language in a writing class
(Sun, 2003; Gilmore, 2009), especially in the revision stage of a process-oriented writing.
We concede our findings are tentative due to the exploratory nature and small sample size of this study.
Yet, our unique design--comparing indirect to direct corpus use--provided meaningful findings, so it may
well serve as a worthy model to emulate for future studies. This contrasts to the usual method that
examines only one corpus method in the classroom. While previous studies have been inconsistent about
the relationship between proficiency levels and the effect of corpus use (e.g. Kennedy & Miceli, 2001;
Gaskell & Cobb, 2004; McCay, 1980; Tribble, 1991; Tribble & Jones, 1990; Yoon & Hirvela, 2004), this
study revealed that corpus-based instruction can benefit different levels of students if we consider their
different patterns and strategies relative to the two corpus applications. Again, the small numbers
involved in this study may prejudice a firm conclusion, but the indirect use of corpora appeared to have
greater effectiveness in error correction for most learners. Still, the learners preferred the interactive
aspect of direct corpora use, which also raised their learning awareness. If these findings are replicated,
then it can reassure teachers with less-endowed technological infrastructures--particularly in developing
countries-- that indirect use can still benefit students linguistic acquisition in L2 writing.
Our study may have pedagogical implications because we identified specific ways that teachers can
regulate the classroom environment to enhance the learning experience of either direct or indirect corpus
use. Future studies may want to address: accuracy of students prior knowledge, levels of student
proficiency, level of research skills, students learning styles, the degree of teachers intervention, and
accessibility to corpus resources, to name a few. Our findings suggest that indirect deployment of corpus
requires the teachers to consider intrinsic factors such as students prior knowledge, language proficiency
and learning styles. The direct deployment of corpus requires evaluation of not only intrinsic factors but
also extrinsic factors such as the teachers intervention and accessibility to corpus resources. Limited
accessibility to the resources and corpus programs may impede direct use of corpora, but we mitigated
this drawback by employing a free online corpus, which may be readily accessed on the World Wide
Web. The Lextutor Web site not only enabled students easy access to corpora, but they easily handled
the data with minimal training. We argue that the use of Web resources can minimize the effects of the
digital divide that particularly confronts developing countries (and even in some economically depressed
areas within advanced economies) whose educators are tasked to prepare students for a hyper-competitive
world economy in which English is the lingua franca. Furthermore, indirect corpus is still available to
teachers, even if students lack the individual concordance resources due to fewer computer resources. Use
of either approach makes for a win-win situation for teachers and students wherever corpus-based
instruction is employed.
112
The literature shows that corpus size can be a considerable matter in terms of teachers time management
(St. John, 2001) in using indirect corpus. A teacher can spend substantial time selecting appropriate
outputs from an immense number of examples based on the target errors and students levels. Frequency
of occurrence may not be generalizable when the amount of examples is relatively small. However, smallsize corpora may not only be valuable for learners language discovery but can also familiarize learners to
use larger corpora appropriately (Aston, 1997). It is not possible to claim conclusively on the basis of our
small-scale study, but we were struck by how a relatively few examples were able to satisfy learners if the
error was simple or inadvertent. However, it was apparent that more examples are needed to correct
fossilized erroneous knowledge especially for high-level students who resisted corrections. To manage
this situation, the teacher must determine the levels of students proficiencies and the linguistic items in
the inquiry in order to select the amount of data for concordance examples. On the other hand, while
indirect corpus use is often recommended as a transitional step to moving to direct corpus use or to
independent concordance usage (Gaskell & Cobb, 2004; Johns, 1997), this study showed that indirect
corpus use can be a great learning method for lower-level students in its own right, thus assuaging any
deleterious effects of the digital divide where direct access may not be easily available in less-developed
countries. In any event, indirect use can also prepare these students for direct use if they should transfer to
a country with more advanced facilities, or if their institutions facilities should be eventually upgraded.
A more crucial issue for direct corpus use is training students about how to utilize the concordancer. In
light of the results on the frequency of strategy use, the learners used almost 50 percent of making use of
materials strategy in the direct corpus use context, which means that students rely on the material the
most. Students well-trained for using online corpora are likely to engage in autonomous browsing
(Bernardini, 2002), leading to discovery learning. This supplants the teacher as the subsequent mediator
of corpus materials and leads to student-directed learning as a true linguistic researcher and as an
independent L2 writer. This method can be rewarding for the teacher who finds it difficult to teach highly
confident students who resist correction of erroneous knowledge. Since they access concordance
resources with their own effort, they may be more likely to repair their knowledge. Our observation
comports to Chambers and OSullivan (2004), who argued that corpus consultation is good for
unlearning errors (p.168).
Regardless of the type of corpus use, the study found that teachers guidance and scaffolding was crucial
in helping to lead learners to successful experiences in corpus analysis. In this sense, we stress that
teachers must receive basic training in accessing corpora and evaluating concordancers in order to foster a
DDL-friendly environment. As Sinclair (2004) noted, a corpus is not a simple object, and it is just as
easy to derive nonsensical conclusions from the evidence as insightful ones (p.2). When training
familiarizes teachers with the corpus-based environment, they can facilitate students autonomous
learning and make them become active language detectives.
In conclusion, this study provides an in-depth but preliminary understanding of EFL students' learning
processes and strategies they will likely apply in two different uses of corpora. At the very least, we hope
our research design will be a useful model for future inquiry into this area, and that our observations may
be useful in interpreting the experiences so obtained. Corpus applications have great potential as a
learning tool as well as a linguistic resource to improve learners knowledge of language use and to
enhance discovery learning. The findings from our small sample should be carried over to a study of a
larger number of students to confirm its implications. In future studies, the teacher could select typical
types of linguistic errors most students make after essay writing and give corpus-based feedback to the
class. In this way, students can experience inductive learner-centered error correction with their own or
classmates' errors. This study introduced the indirect approach and moved to the direct approach to make
students familiar with the corpus approach. We suggest that future studies should alternate the approaches
weekly to investigate the effects. Learning from the indirect approach might accelerate the learning in the
direct approach in this study, so future studies should verify any extent, if at all, of this possible effect.
113
Finally, other studies may wish to emulate our design by expanding beyond Lexitutor as a concordance
resource. There will be other Web-based resources which may be more easily accessed with lessadvanced computing facilities in developing countries. In this way, an effective language-teaching tool
can enhance the learning of those who seek it.
NOTE
1. Pseudonyms were assigned to each participant in order to protect their confidentiality.
ACKNOWLEDGEMENTS
This work was supported by Hankuk University of Foreign Studies Research Fund. The authors are very
grateful to the editors and anonymous reviewers for their valuable insights and constructive comments on
an earlier draft.
ABOUT THE AUTHORS

Hyunsook Yoon is Associate Professor in the Department of English Education at Hankuk University of
Foreign Studies. Her primary research interests include corpus linguistics, second language writing, and
ESP/EAP.
E-mail: hsyoon3@hufs.ac.kr
Jung Won Jo received her M.Ed. degree from the Graduate School of Education at Hankuk University of
Foreign Studies. Her research interests include corpus linguistics and learning strategies.
E-mail: jazzlove83@gmail.com
REFERENCES
Aston, G. (1997). Enriching the learning environment: Corpora in ELT. In A. Wichmann, S. Fligestone,
T. McEnery, & G. Knowles (Eds.), Teaching and Language Corpora (pp. 5164). New York, NY:
Longman.
Bernardini, S. (2000). Systematising serendipity: Proposals for concordancing large corpora with
language learners. In L. Burnard, & T. McEnery (Eds.), Rethinking language pedagogy from a corpus
perspective: Papers from the Third International Conference on Teaching and Language Corpora
(pp.225234). Frankfurt, Germany: Peter Lang.
Bernardini, S. (2002). Exploring new directions for discovery learning. In B. Kettemann & G. Marko
(Eds.), Teaching and learning by doing corpus analysis. Proceedings from the Fourth International
Conference on Teaching and Language Corpora, Graz 19-24 July, 2000 (pp. 165182). Amsterdam:
Rodopi.
Bernardini, S. (2004). Corpora in the classroom: An overview and some reflections on future
developments. In J. Sinclair (Ed.), How to use corpora in language teaching (pp.1536). Amsterdam:
Benjamins.
Boulton, A. (2008). Looking for empirical evidence of data-driven learning at lower levels. In B.
114
Lewandowska-Tomaszczyk (Ed.), Corpus linguistics, computer tools, and applications: State of the art
(pp. 581598). Frankfurt: Peter Lang.
Boulton, A. (2009). Testing the limits of data-driven learning: Language proficiency and training.
ReCALL, 21(1), 3751.
Boulton, A. (2010a). Data-driven learning: On paper, in practice. In T. Harris, & M. Jan (eds), Corpus
linguistics in language teaching (pp.1752). Frankfurt: Peter Lang.
Boulton, A. (2010b). Data-driven learning: Taking the computer out of the equation. Language Learning,
60(3), 534572.
Braun, S. (2007). Integrating corpus work into secondary education: From data-driven learning to needsdriven corpora. ReCALL, 19(3), 307328.
Chambers, A. (2005). Integrating corpus consultation in language studies. Language Learning &
Technology, 9(2), 111125. Retrieved from http://llt.msu.edu/vol9num2/chambers/default.html
Chambers, A. (2007). Popularising corpus consultation by language learners and teachers. In E. Hidalgo,
L. Quereda & J. Santana (Eds), Corpora in the foreign language classroom (pp.3-16). Amsterdam, The
Netherlands: Rodopi.
Chambers, A., & OSullivan, . (2004). Corpus consultation and advanced learners writing skills in
French. ReCALL, 16(1), 158172.
Chandler, J. (2003). The efficacy of various kinds of error feedback for improvement in the accuracy and
fluency of L2 students writing. Journal of Second Language Writing, 12, 267296.
Charles, M. (2007). Reconciling top-down and bottom-up approaches to graduate writing: Using a corpus
to teach rhetorical functions. Journal of English for Academic Purposes, 6, 289302.
Cheng, W., Warren, M., & Xun-feng, X. (2003). The language learner as language research: Putting
corpus linguistics on the timetable. System, 31(2), 173186.
Cobb, T. (1997). Is there any measurable learning from hands-on concordancing? System, 25(3), 301315
Creswell, A. (2007). Getting to know connectors? Evaluating data-driven learning in a writing skills
course. In E. Hidalgo, L. Quereda & J. Santana (eds), Corpora in the foreign language classroom
(pp.267287). Amsterdam, The Netherlands: Rodopi.
Gaskell., D., & Cobb, T (2004). Can learners use concordance feedback for writing errors? System, 32,
301319.
Gabrielatos, C. (2005). Corpora and language teaching: Just a fling, or wedding bells? TESL-EJ, 8(4), AL,
137.
Gilmore, A. (2009). Using online corpora to develop students writing skills. ELT Journal. 64(3), 363
372.
Johns, T. (1991). From printout to handout: Grammar and vocabulary teaching in the context of datadriven learning. English Language Research Journal, 4, 2745.
Johns, T. (1997). Contexts: the background, development and trialing of a concordance-based CALL
program. In A. Wichmann, S. Fligelstone, T. McEnery & G. Knowles (Eds.), Teaching and language
corpora, pp.100115. London and New York: Longman.
Kennedy, C., & Miceli, T. (2001). An evaluation of intermediate student approaches to corpus
investigation. Language Learning & Technology, 5(3), 7790. Retrieved from
http://llt.msu.edu/vol5num3/kennedy/default.html
115
Kennedy, C., & Miceli, T. (2010). Corpus-assisted creative writing: Introducing intermediate Italian
learners to a corpus as a reference resource. Language Learning & Technology, 14(1), 28-44. Retrieved
from http://llt.msu.edu/vol14num1/kennedymiceli.pdf
Lee, D., & Chen, S. (2009). Making a bigger deal of the smaller words: Function words and other key
items in research writing by Chinese learners. Journal of Second Language Writing, 18, 149-165.
Lee, D., & Swales, J. (2006). A corpus-based EAP course for NNS doctoral students: Moving from
available specialized corpora to self-compiled corpora. English for Specific Purposes, 25, 56-75.
McCay, S. (1980), Teaching the syntactic, semantic and pragmatic dimensions of verbs. TESOL
Quarterly, 14(1), 17-26.
Mukherjee, J. (2006). Corpus linguistics and language pedagogy: The state of the art and beyond. In S.
Braun, K. Kohn & J. Mukherjee (Eds.), Corpus Technology and Language Pedagogy (pp.5-24). Frankfurt
am Main: Peter Lang.
Newman, R. S. (1991). Goals and self-regulated learning: What motivates children to seek academic
help? In M. L. Maehr & P. R. Pintrich (Eds.), Advances in motivation and achievement (Vol. 7, pp. 151
183). Greenwich, CT: JAI Press.
Newman, R. S., & Goldin, L. (1990). Childrens reluctance to seek help with schoolwork. Journal of
Educational Psychology, 82, 92100.
OMalley, J., & Chamot.A. (1990). Learning strategies in second language acquisition. Cambridge, UK:
Cambridge University Press.
OSullivan, . (2007). Enhancing a process-oriented approach to literacy and language learning: The role
of corpus consultation literacy. ReCALL, 19(3), 269286.
OSullivan, ., & Chambers, A. (2006). Learners writing skills in French: Corpus consultation and
learner evaluation. Journal of Second Language Writing, 15(1), 4968.
Oxford, R. (1990). Language learning strategies: What every teacher should know. New York, NY:
Newbury House.
Oxford, R. (Ed.). (1996). Language learning strategies around the world: Cross-cultural perspectives.
Honolulu, HI: University of Hawai`i Press.
Prez-Paredes, P., Snchez-Tornel, M., Calero, J., & Jimnez, P. (2011). Tracking learners actual uses of
corpora: guided vs non-guided corpus consultation. Computer Assisted Language Learning, 24(3), 233
253.
Roca de Larios, J., Murphy, L., & Marn, J. (2002). A critical examination of L2 writing process research.
In S. Ransdell, & M. Barbier (eds.), New directions for research in L2 writing (pp.1147). Dordrecht, The
Netherlands: Kluwer Academic Publishers.
Rmer, U. (2008). Corpora and language teaching. In A. Ldeling & Kyt (Eds.), Corpus linguistics: An
international handbook (pp.112130). Berlin, Germany: Mounton de Gruyter.
Rubin, J. (1975). What the good learner can teach us? TESOL Quarterly, 9(1), 4151.
Shi, L. (2006). The successors to Confucianism or a new generation? A questionnaire study on Chinese
students culture of learning English. Language, Culture and Curriculum, 19(1), 122147.
Sinclair, J. (2004). How to use corpora in language teaching. Amsterdam, The Netherlands: John
Benjamins.
Sripicharn, P. (2004). Examining native speakers and learners investigation of the same concordance
116
data and its implications for classroom concordancing with ELF learners. In G. Aston, S. Bernardini, &
D. Stewart (Eds), Corpora and language learners (pp.233245). Amsterdam, The Netherlands: John
Benjamins.
Stevens, V. (1991). Classroom concordancing: Vocabulary materials derived from relevant authentic text.
English for Specific Purposes, 10(1), 3546.
St. John, E. (2001). A case for using a parallel corpus and concordance for beginners of a foreign
language. Language Learning & Technology, 5(3), 185203. Retrieved from
http://llt.msu.edu/vol5num3/stjohn/default.html
Sun, Y. (2003). Learning processes strategies and web-based concordances: A case study. British of
Educational Technology, 34(5), 601613.
Sun, Y. (2007). Learner perceptions of a concordancing tool for academic writing. Computer Assisted
Language Learning, 20(4), 323343.
Tian, S. (2005). The impact of learning tasks and learner proficiency on the effectiveness of data-driven
learning. Journal of Pan-Pacific Association of Applied Linguistics, 9(2), 263275.
Tribble, C. (1991). Concordancing and an EAP writing programme. CAELL Journal, 1(2), 1015.
Tribble, C. (1997). Put a corpus in your classroom: Using a computer in vocabulary development. In T.
Boswood (Ed.), New ways of using computers in language teaching (pp. 266268). Alexandria, VA:
TESOL.
Tribble, C., & Jones, G. (1990). Concordances in the classroom. London, UK: Longman.
Yoon, H. (2008). More than a linguistic reference: The influence of corpus technology on L2 academic
writing. Language Learning & Technology, 12(2), 3148. Retrieved from
http://llt.msu.edu/vol12num2/yoon/
Yoon, H., & Hirvela, A. (2004). ESL students toward corpus use in L2 writing. Journal of Second
Language Writing, 13, 257283.
Zimmerman, B. J., & Martinez-Pons, M. (1986). Development of a structured interview for assessing
student use of self-regulated learning strategies. American Educational Research Journal, 23, 614628.
Zimmerman, B., & Martinez-Pons, M. (1988). Construct validation of a strategy model of student selfregulated learning. Journal of Educational Psychology, 80(3), 284290.
117

http://llt.msu.edu/issues/february2014/monteroperezetal.pdf

pp. 118-141
EFFECTS OF CAPTIONING ON VIDEO COMPREHENSION AND

INCIDENTAL VOCABULARY LEARNING
Maribel Montero Perez, iMinds - ITEC KU Leuven Kulak
Elke Peters, KU Leuven
Geraldine Clarebout, KU Leuven
Piet Desmet, iMinds - ITEC KU Leuven Kulak
This study examines how three captioning types (i.e., on-screen text in the same language
as the video) can assist L2 learners in the incidental acquisition of target vocabulary words
and in the comprehension of L2 video. A sample of 133 Flemish undergraduate students
watched three French clips twice. The control group (n = 32) watched the clips without
captioning; the second group (n = 30) watched fully captioned clips; the third group (n =
34) watched keyword captioned clips; and the fourth group (n = 37) watched fully
captioned clips with highlighted keywords. Prior to the learning session, participants
completed a vocabulary size test. During the learning session, they completed three
comprehension tests; four vocabulary tests measuring (a) form recognition, (b) meaning
recognition, (c) meaning recall, and (d) clip association, which assessed whether
participants associated words with the corresponding clip; and a final questionnaire.
Our findings reveal that the captioning groups scored equally well on form recognition and
clip association and significantly outperformed the control group. Only the keyword
captioning and full captioning with highlighted keywords groups outperformed the control
group on meaning recognition. Captioning did not affect comprehension nor meaning
recall. Participants vocabulary size correlated significantly with their comprehension
scores as well as with their vocabulary test scores.
Keywords: Video, Listening, Vocabulary, Multimedia, CALL
APA Citation: Montero Perez, M., Peters, E., Clarebout, G., & Desmet, P. (2014). Effects
of Captioning on Video Comprehension and Incidental Vocabulary Learning. Language
Learning & Technology 18(1), 118141. Retrieved from
http://llt.msu.edu/issues/february2014/monteroperezetal.pdf
Received: May 10, 2012; Accepted: April 8, 2013; Published: February 1, 2014
Copyright: Maribel Montero Perez, Elke Peters, Geraldine Clarebout, Piet Desmet
INTRODUCTION
The emergence of multimedia learning environments (Brett, 1995) and the overall accessibility of video
(DVD, YouTube, etc.) have created important platforms for enhancing second language (L2) listening
development (Vandergrift, 2011). These platforms are increasingly being used in classroom practice
(Grgurovi & Hegelheimer, 2007; Vanderplank, 2010) and provide learners with a number of listening
support options, mostly realized in the form of a technological overlay (Robin, 2007, p. 109) such as
native language (L1) subtitles (L2 video, L1 on-screen text), reversed subtitles (L1 video, L2 text), and
captioning (L2 video, L2 text). This study focuses on captioning, which maximally exposes learners to
the L2, and which has been proven to be beneficial for augmenting comprehension (Baltova, 1999;
Chung, 1999; Markham, 2001) and fostering vocabulary learning (Markham, 1999; Sydorenko, 2010).
While extensive research has addressed the potential of full captioning for video comprehension and
vocabulary learning, research studies have not provided answers to several questions. Researchers have
suggested simplifying full captions (Garza, 1991; Winke, Gass, & Sydorenko, 2010) but research that has
118
Maribel Montero Perez, Elke Peters, Geraldine Clarebout, & Piet Desmet
Effects of Captioning
attempted to reduce the textual density by offering only keyword captions has yielded inconclusive results
(Guillory, 1998; Park, 2004). Additionally, research on vocabulary learning has identified attention as
being a crucial component for vocabulary learning (Hulstijn, 2001), yet no study has looked at the effects
of salience in the captioning line for drawing L2 learners attention to target vocabulary and enhancing
incidental vocabulary learning, that is, vocabulary learning as a by-product of listening for meaning
(Gass, 1999).
This study expands on previous captioning research by investigating not only video with full captions but
also video with keyword captions and video with full captions and highlighted keywords in order to
reveal the effectiveness for L2 learners vocabulary learning and comprehension of video content in the
context of L2 French.
Literature Review
Our literature review focuses on three elements: (a) the importance of lexical coverage for successful
listening comprehension and vocabulary learning, (b) the effectiveness of captions for L2 video
comprehension, and (c) the effectiveness of captioned video for L2 vocabulary acquisition.
Research has indicated that comprehension of written or aural input is crucial for acquiring new linguistic
items incidentally (Gass, 1997; Lee & Van Patten, 2003). Although Rost (2002) identified listening as
one of the most important sources for L2 acquisition, only a handful of researchers have investigated the
relationship between listening comprehension and vocabulary learning and tried to discover which factors
affect comprehension and chances for word learning. In a series of studies, Webb (2010, 2011) and Webb
and Rodgers (2009) investigated the importance of the lexical coverage of television programs in relation
to incidental vocabulary learning and text comprehension. They found that knowledge of the 2,000 to
4,000 most frequent word families provides 95% coverage of television programs, which, according to
Nation (2006), is the coverage required for incidental learning and adequate text comprehension. Webb
and Rodgers results indicated that to achieve 98% coverage, 5,000 to 9,000 word families are necessary,
depending on the television genre. van Zeeland and Schmitt (2012) suggested that learners with a
vocabulary size between 2,000 and 3,000 word families have 95% coverage, which is sufficient for
adequate understanding of storytelling passages (audio only). Although the findings of the
aforementioned studies present slightly different figures on the number of words needed, they provide
accumulative evidence for the claim that vocabulary size is highly correlated with listening success
(Staehr, 2009) and vocabulary learning (Neuman & Koskinen, 1992). A general finding, however, is that
most L2 learners have not yet developed a sufficiently large vocabulary to understand L2 video. In spite
of the richness of video material (Baltova, 1999), it is not necessarily appropriate for adequate
comprehension and learning (Danan, 2004). One solution to help learners is to provide them with L2
subtitles or captioning.
The bulk of captioning research has focused on the effectiveness of captioning for L2 learners
comprehension of video content (Baltova, 1999; Chung, 1999; Garza, 1991; Huang & Eskey, 1999-2000;
Markham, 2001; Neuman & Koskinen, 1992; Park, 2004; Winke et al., 2010). In general, researchers
have found that captioning enhances comprehension of L2 clips. Bird and Williams (2002) suggested that
captions aid speech decoding and segmentation by helping listeners visualize the speech stream and
clearly indicating word boundaries. Captioning has therefore been characterized as a mediating device
(Vanderplank, 1988, p. 280), helping the learner when automated sound-script recognition falls short.
Markham (1999) suggested that captions also help the development of word recognition skills. By doing
so, it makes ambiguous speech clearer the next time it will be encountered (Bird & Williams, 2002;
Garza, 1991) and enables learners to successfully cope with input that is slightly above their actual
proficiency level (Danan, 2004; Neuman & Koskinen, 1992).
While it is generally acknowledged that captioning improves L2 listeners understanding (e.g. Baltova,
1999; Huang & Eskey, 1999-2000), Pujol (2002) and King (2002) have argued that the presence of
119
captions makes it difficult to conclude whether comprehension scores reflect participants listening or
reading skills. Although this claim has not been tested empirically, studies have shown that the
availability of captions does not compromise successful auditory processing (Bird & Williams, 2002;
Markham, 1999, as cited in Danan, 2004; and Vanderplank, 2010). Other researchers have tried to find
out if the amount of textual density in the captioning line can be reduced by exploring the potential of
keyword captioned video (Guillory, 1998; Park, 2004), as suggested by Garza (1991) and more recently
by Winke et al., (2010). Keywords are expected to tackle the problem of textual density (Guillory, 1998)
while still providing support. Guillorys study examined to what extent keywords can support
understanding of L2 video for beginning learners of French. The results showed that learners in the
keyword group (receiving only 14% of the full captioning text) performed as well as the full captioning
group on the comprehension questions. Moreover, the keyword group significantly outperformed the no
captioning group. Unlike Guillorys study, Parks research (2004) indicated that only the more advanced
students were able to benefit from keyword captions; lower-level students did not significantly
outperform the no captioning group. It is, however, unclear to what extent this finding can be attributed to
the keyword captions or to the video input selected for his study. Since Park used the same input for both
lower-level and higher-level learners, there might have been a mismatch between lower-level learners
proficiency and the level needed for the video (Vanderplank, 2010; Winke, et al., 2010).
Although far fewer studies have investigated the effects of captioned video on L2 learners vocabulary
learning (e.g. Baltova, 1999; Danan, 1992; Sydorenko, 2010), two general conclusions can be drawn from
the results of previous research. First of all, it has been shown that captions significantly help learners on
written form recognition (Neuman & Koskinen, 1992; Sydorenko, 2010) and aural form recognition tests
(Markham, 1999). However, Sydorenko (2010) found that learners in the video only group outperformed
the captioning group on an aural form recognition test and therefore concluded that captioning success
may depend on test modality. Yet, previous research has tended to suggest that captions help L2 learners
isolate word forms (Winke et al., 2010) and pay attention to them, which may subsequently stimulate
noticing of these forms. With regard to noticing, it is well documented in the literature on vocabulary
acquisition that noticing unknown words in the input is the first step in the acquisition process (Huckin &
Coady, 1999; Hulstijn, 2001). The crucial role of attention is also at the basis of Vanderplanks
speculative model (1990, p. 228) on language learning through captioned video. In his model, the
taking out of language from captioned video consists of both attention and adaptation. Vanderplank
defines attention as a conscious selection process that is based on systematically noting and gathering
(p. 229) information and on a reflective component in which learners notice a gap while comparing their
L2 knowledge with the captioned video input. Adaptation, the selection of linguistic elements that
learners pay attention to for [their] own purposes (Vanderplank, p. 229), aligns with Gass position that
learners have their own focus of attention (1999, p. 321).
Second, captions also help learners make form-meaning connections in the mental lexicon, which
constitutes a crucial process in the acquisition of lexical items (Van Patten, Williams, & Rott, 2004). With
regard to those form-meaning connections, studies have found that captions help learners not only to
recognize the meaning of target words (Huang & Eskey, 1999-2000; Neuman & Koskinen, 1992) but also
to provide translations (Sydorenko, 2010; Winke et al., 2010) or to produce the target forms themselves
(Baltova, 1999; Danan, 1992). While these studies provide valuable information on the potential of
captions for vocabulary acquisition, further research is needed to investigate whether captions can be
enhanced in order to stimulate vocabulary learning.
As shown in this section, previous research has lent support to the use of captioning in classroom
contexts. Yet, it also reveals a number of issues that require further research. One of the recurring
questions has been whether captions can be enhanced by providing L2 learners with salient keywords
rather than full captions (Garza, 1991; Guillory, 1998; Park, 2004), or by visually enhancing keywords in
the captioning line (Winke et al., 2010). While two studies have examined the effectiveness of keywords
120
for L2 learners understanding of video, they produced conflicting results (Guillory, 1998; Park, 2004).
Moreover, although attention has been identified as a crucial component of vocabulary learning (Hulstijn,
2001), we are not aware of studies that have focused on the effects of salience in the captioning line,
either through keywords or highlighted items in full captions, in the context of vocabulary learning. A
study comparing full captions to (highlighted) keywords may provide more information on (a) what
learners notice in the input and (b) how full captions and keywords can enhance attention and processing
quality in terms of vocabulary gains.
Research Questions
This study investigates the potential of three types of captioned video, varying in the amount of text and
the salience of the lexical items in the captioning line, for content comprehension and incidental
vocabulary learning. In particular, we have included three captioning groups (see Table 1): full captioning
(FC), keyword captioning (KC), and full captioning with highlighted keywords (FCHK) and compared
their results with those of a control group (NC).
Table 1. Overview of Experimental Conditions
Name of condition
No captioning
(NC)
Full captioning with

Full captioning Keyword
captioning (KC highlighted keywords
(FC)
(FCHK)
Participants per condition
32
30
34
37
Amount of textual support
None
Full
Partial
Full
Amount of salience
No salience
No salience
Salience
Salience of keywords
We were guided by two main research questions:

1) Does the type of captioning have a differential effect on L2 learners understanding of video
content, as measured by three comprehension tests consisting of global and detailed questions?
We hypothesize that learners in the captioning groups will outperform the NC group because of the
availability of on-screen text. With regard to the differences between FCHK, FC, and KC, we adopt a null
hypothesis as earlier studies have provided no or inconclusive results in this respect.
2) Does the type of captioning have a differential effect on L2 learners incidental learning of
target vocabulary words, as measured by a form recognition, clip association, meaning
recognition, and meaning recall test?
We expect KC and FCHK to outperform FC and NC and for FC to outperform NC on the form
recognition and clip association tests because the lexical items salience may help the groups to notice
words (Brett, 1998) and associate them with the correct clip more easily. With regard to meaning
recognition, we hypothesize that FCHK will outperform KC and FC as the availability of highlighted
keywords and full captioning may facilitate the inference of word meaning and initial form-meaning
mapping. We also expect the differences between the captioning groups to disappear for meaning recall as
the latter might be hard to achieve after the relatively short exposure to the target words. Yet, the
captioning groups are still expected to outperform NC. Table 2 summarizes our hypotheses.
121
Table 2. Research Hypotheses

Test
Hypothesis
Listening comprehension
FC, KC, and FCHK > NC
Form recognition and clip association
KC and FCHK > FC > NC
Meaning recognition
FCHK > FC and KC > NC
Meaning recall
FC, KC, and FCHK > NC
METHODS
Participants
The participants were 133 undergraduate law students (55 males, 78 females) at a Flemish university
(Mage= 17.98 years, SD = .51). All were native speakers of Dutch, except for four French-speaking
students whose data were excluded from the analysis. All students had an obligatory course in Legal
French, which focuses on communicative competence and legal vocabulary. Classes were organized in
six parallel groups of approximately 25 students. Participants were informed that attending the
experiment would count as two of the 14 obligatory hours of online training. The participants could be
considered as (high-) intermediate learners of French, as measured by the self-designed vocabulary size
test (see Instruments and Results section).
The four conditions (NC, FC, KC, and FCHK) were randomly assigned to the six groups. Four groups
were assigned to one of the four conditions, while the other two groups were divided equally over the four
conditions in order to obtain a balanced composition.
Materials
Video Selection
Since our study was embedded in a formal classroom setting, we selected three relatively short,1 but
authentic French clips2 from a Belgian and Swiss current affairs program for native speakers of French.
The clips, which were available online, had a single narrator and included short interviews in which at
least one interlocutor was shown. The first video (225, 376 words) discussed the production and export
strategy of a French brewery. The second and third clip presented the marketing strategy (424, 772
words) and history (332, 576 words) of the Lego factory. The first and second clip were used without
modifications. We removed a two-minute section of the third clip because it required too much economic
knowledge. The three clips were manually transcribed and captions were added with MAGpie
(http://ncam.wgbh.org/invent_build/web_multimedia/tools-guidelines/magpie).
Target Word Selection
An important criterion for selection was that the clips contained a number of words that were very likely
to be unfamiliar to the participants, that is, the target words (TWs) of this study. We adopted two
procedures to verify familiarity with the TWs:
First, we compiled a list of 140 possible TWs appearing in the clips and a set of pseudowords. A
representative group of students (N = 40) of the same university, with a proficiency level similar to that of
the actual participants of this study, indicated whether or not they knew each of the 140 words. We
retained all the words that were not known by at least 70% of the students, which resulted in a set of 20
TWs (see Appendix A). We did not set the cognition level at 100% because students did not need to
provide the actual translation and may have overestimated their knowledge.
Second, prior knowledge of the TWs in the present study was controlled for by means of a pretest. The
prior knowledge test was administered to the participants four weeks before the learning session and
122
contained the 20 potential TWs and 18 distracters (easier words). The test format was similar to the
Vocabulary Knowledge Scale (VKS) (Wesche & Paribakht, 1996) and measured participants depth of
vocabulary knowledge. For each item, participants indicated the corresponding level (see Instruments
section, Table 4, Test 3). We left out the fifth level (I can use this word in a sentence, write a sentence)
because it was very unlikely that the learners would be familiar with the TWs. In addition, it is hard to
determine objectively whether the sentence demonstrates knowledge of the word (Bruton, 2009; Nation &
Webb, 2011). Results of the prior knowledge test showed that learners were familiar with three TWs,
which were subsequently excluded from the TWs (see Appendix A).
Of the 17 retained TWs, there were seven nouns, four verbs, and six multiword units (verb and noun or
pronoun and verb). None of the TWs were cognates. Since we selected authentic videos, it was not
possible to control for the frequency of the TWs.3 Four TWs occurred more than once in the clip (see
Appendix A). All TWs appeared as highlighted keywords in the FCHK condition and in isolation in the
KC condition. The four TWs with a higher occurrence were keyword captioned for every occurrence in
the clip.
Five experienced lecturers of French were asked to verify the appropriateness of the clips for the target
audience and to rate the correlation between the audio and visual images in the clips. In their judgment,
the clips were appropriate and visual images were considered supportive of the dialog but did not provide
explicit information on the meaning of the TWs. There were, however, contextual clues available in the
sentences containing the TWs. These clues are important to infer word meaning (Nation, 2001) and
establish initial form-meaning connections. For example, in the sentence Lego est revenue la une aprs
avoir frl le naufrage, il y a dix ans cause des jeux lectroniques (Lego made it back to the top, after
being on the verge of disaster, ten years ago because of electronic games), the context makes it possible
to infer the meaning of the underlined TWs.
Keyword Determination Procedure
We define keywords as words that are important for the meaning of the sentence or paragraph. The
keywords were presented like the TWs: either in isolation (KC) or as highlighted words (FCHK). The
same five experienced lecturers of French were asked to highlight keywords for comprehension in the
three video transcripts (as in Guillory, 1998, and Park, 2004). We processed their selections, compared
them with our keywords and selected a final set of keywords, representing 17.11% (or 295 out of 1,724
words) of the total number of words in the three videos.
Instruments
Vocabulary Size Test
Previous research has revealed that vocabulary size is linked to listening success (Staehr, 2009; van
Zeeland & Schmitt, 2012; Webb & Rodgers, 2009) and vocabulary learning (Baltova, 1999; Webb &
Rodgers, 2009). Vocabulary size is also claimed to give a rough estimate of learners language
proficiency (Milton, Wade, & Hopkins, 2010). Therefore, we designed a 50-item4 multiple choice test
that comprised three parts corresponding to the following word frequency bands: 2,001-4,000 (21 items),
4,001-5,000 (15 items), and 5,001-7,000 (14 items), based on the Routledge (Lonsdale & Le Bras, 2009),
Verlinde (Selva, Verlinde, & Binon, 2002), and DPC corpus (Paulussen, Macken, Trushkina, Desmet, &
Vandeweghe, 2006) frequency lists. Every test item was a written multiple choice question containing
four Dutch translation options of the item (see Figure 1). Although a study on aural input should ideally
include a spoken vocabulary size test, previous studies have shown that the use of a written vocabulary
size test did not compromise the findings regarding the significant correlation between vocabulary size
and listening comprehension (e.g., Staehr, 2009; van Zeeland & Schmitt, 2012).
123
Figure 1. Sample of vocabulary size test where the first row contains the French word and its correct
translation is in italics.
Three Comprehension Tests
We developed a comprehension test for each clip. Together, the three tests included 41 items: 19 short
open-ended questions, 14 true-false items, and eight combination items (see Table 3). All participants
were native speakers of Dutch and were asked to answer the open-ended questions in their L1 (Buck,
2001). The development of the test items was inspired by Bucks competency-based default listening
construct (p. 114), which measures general understanding, detailed content, and inferencing ability. Of the
41 items, 25 items focused on understanding the main ideas and 16 items targeted more detailed yet
relevant elements. We did not include inferencing questions due to two limitations: (a) we used relatively
short clips, providing very concrete and factual information; and (b) in order to prevent a task effect on
vocabulary learning, the comprehension tests did not focus on the sections containing the TWs, posing
extra limitations on the information that could be included in the tests.
The comprehension questions targeted the keywords, which means that questions only focused on
important ideas. Moreover, if we had focused on ideas that were not represented in the keywords, we
would not have been able to investigate differences between NC and KC (see Table 3).
Table 3. Sample of Comprehension Questions
Question Type
Short open-ended
question
Comprehension Questions*
1. Explain why the French Craftworks association was established.
(video 1)
Answer:
In order to stimulate the export of French beer.
2. Explain why Lego is a classic marketing example. (video 2)

Answers: Lego has an online platform where customers can present their creations.
By posting pictures of their creations on the online platform, they
provide inspiration for new Lego sets.
3. According to the interviewee, what caused the crisis at Lego? (video 3) (with picture
of the interviewee).
Answer:
Lego wanted to grow so fast that they started to manufacture new

124
products. They did however not have the expertise to do so.

True/False
questions
Indicate if the statement is true or false and, if false, correct the sentence.
Annick states that the brewerys export rate is satisfactory since they
export more than the national average. (video 1)
Lego refunds purchases when customers put pictures or videos of their
Lego constructions on the Internet. (video 2)
Combination task
Who says what? Combine each statement with the corresponding picture**
(video 2).
Il y a une rvolution dans la brique de mon enfance
(picture of toy store communication manager)
On a un laboratoire de 20.000, 200.000 utilisateurs
(picture of marketing specialist)
Cest un jouet quon garde longtemps (picture of Lego retailer)
Note: *We checked the internal consistency of the 41 items and found acceptable reliability (N = 133, Cronbachs alpha = .73).
** The pictures are not reproduced due to copyright restriction. The words in italics were keyword captioned for KC and FHCK.
Four Vocabulary Tests

It is generally assumed that vocabulary acquisition is not an all or nothing phenomenon (Laufer, Elder,
Hill, & Congdon, 2004, p. 209) but an incremental process with noticing as initial step (Hulstijn, 2001).
Therefore, we assessed vocabulary learning by means of multiple tests, as suggested by Nation and Webb
(2011). We first measured three aspects of word knowledge: form recognition, clip association, and
meaning recall.5 The form recognition test assessed whether learners were able to recognize the TWs, a
selection of keywords that were not TWs, and some distracters. Learners ticked off yes if they thought
the word appeared in the clips, and no if they did not (see Table 4, Test 1). This test did not include
non-words but we controlled for guessing by also presenting a clip association test. The clip association
test contained the same TWs, keywords, and distracters as the form recognition test and built on learners
answers on the previous test. If learners had checked yes on the form recognition test, they were asked
to indicate in which clip the word had occurred (Brewery or Lego). This allowed us to check if they could
associate the word with a vague meaning (see Table 4, Test 2). The form recognition and clip association
tests were combined with a VKS identical to the one used in the prior knowledge test (see Table 4, Test
3). The VKS enabled us to check whether learners were able to translate the TWs into their L1 (meaning
recall).
Table 4. Target Word "amertume" in the Four Vocabulary Tests
Test
1
Form recognition
Word used in the clips? Yes No
Clip association
If yes, in which clip? Brewery Lego
VKS (meaning recall)
Amertume
I dont remember having seen this word before.
I have seen this word before but I dont know the meaning.
I think I know the meaning: (provide translation).
125
I am sure that this word means:... (provide translation).

4
Meaning recognition
Amertume
1. evenwicht (balance)
2. ontgoocheling (disappointment)
3. bitterheid (bitterness)
4. deskundigheid (expertise)
In addition to the first test, we also administered a multiple choice meaning recognition test that included
only the 17 TWs in order to check if learners were able to recognize the translation of the TWs among
four Dutch translation options (see Table 4, Test 4). The tests in Table 4 were all administered
immediately after the treatment.6
Questionnaire
Participants completed a questionnaire, consisting of a five-point Likert scale, including statements on
video comprehension (research question 1), vocabulary learning (research question 2), and the usefulness
of captions for both tasks. The questionnaire data were used to help us clarify the findings of our
quantitative research.
Procedure
Three months before the experiment, the materials and procedures were pilot tested with a group of 22
students in their last year of secondary school who were enrolled in a French summer course. The
participants completed all the tests used in the present study and evaluated the materials by means of a
questionnaire. Our main finding was that the comprehension tests were too easy because many
participants achieved maximum scores. However, the questionnaire results showed that they considered
the clips challenging. As a result, some questions were rewritten for the present study and extra openended questions were added. The results of the pilot test did not reveal problems with the vocabulary tests
or the TWs.
One month prior to the learning session, participants in the present study completed a vocabulary size and
prior knowledge test (see Figure 2). They were told that such tests are typically administered at the
beginning of the academic year. However, they were not informed about the aim of the experiment.
Experimental phase - 90 minutes
Pre-learning phase
- Vocabulary size test
- Vocabulary prior
knowledge test
Clip 1(2x)
Comprehension test 1
Clip 2(2x)
(1 month before 30
minutes)
Clip 3(2x)
Vocabulary test 1
- Noticing
- Clip association
- Meaning recall
Questionnaire
Vocabulary test 2
- Meaning
recognition
Figure 2. Overview of procedures.

The learning session took place in three computer rooms with a PC and headset for each participant.
Students were told that the exercises were part of a research study on the use of video in the L2
classroom. We explained that they were going to watch three short clips. Learners completed a
126
comprehension test after viewing each clip twice (see Figure 2). Students were prompted to focus on the
meaning of the clips and were forewarned of the comprehension tests. Participants were informed that
each comprehension test contained short, open-ended questions, true-false questions, and combination
tasks. They were not allowed to make notes, either during the clips or in between viewings. In order to be
able to gauge incidental learning, we did not inform participants about the vocabulary tests, as suggested
by Hulstijn (2001), who states that the incidental character is ensured at the level of test announcement.
At the end of the learning session, all participants were debriefed about the aim of the experiment
We tested all the students in two consecutive sessions of 90 minutes. The supervisors made sure that
participants involved in the first session did not see participants of the second session and could thus not
inform them about the procedures, as this might have compromised the incidental nature of the
vocabulary acquisition this experiment aimed to measure. All the tests were pencil-and-paper tests.
Scoring
Comprehension Tests
Full credit (1 point) was given for exact answers. Partial credit (0.5) was given for partially correct
answers to open-ended questions. For example, partial credit was given when learners provided only one
of the two parts of the answer to the question Explain why Lego is a classic marketing example.
[Lego has an online platform where customers present their creations.]/[The customers provide
inspiration for new Lego sets.]
Each true-false item consisted of two tasks - (a) indicate whether the sentence is true or false and (b) if
false, correct the sentence - and had a maximum score of 2 points.
Vocabulary Tests
One point was given for each correct answer on the form recognition, clip association, and meaning
recognition tests. The VKS could have been scored in two ways (Bruton, 2009, p. 294): we could have
either used the level number or recoded answers into known words (i.e., correctly translated words, 1
point) and unknown words (0 points). Because we used the VKS to measure meaning recall rather than
progress in the scale (Nation & Webb, 2011), we chose to recode into binomial scoring (see Table 5).7
Table 5. Scoring of VKS
VKS
Level
Binomial scoring
I dont remember having seen this word before.
I have seen this word before but I dont know what it means.
Or: wrong answer at level 3 or 4.
I think I know the meaning of the word and correct translation.
I am sure that this word means and correct translation.
Although most TWs were unknown to students, we wanted to account for their minimal prior knowledge.
We did so by calculating retention scores (Horst, Cobb & Meara, 1998; Peters, 2006, p. 77) for the
meaning recognition and meaning recall test (see Equation 1):
(1)
(testscore priorknowledge)
(max test priorknowledge)
As shown by the equation, we calculated the difference between the number of correct responses on the
127
test (meaning recognition and meaning recall) and the number of known words, as measured by the prior
knowledge test, and then divided this number by the total number of TWs (max test) minus the prior
knowledge score. Since known words were dismissed from the analysis, the result is a retention score
representing the percentage of the actual learning gains.
Analyses
We conducted a multivariate analysis of covariance (Tacq, 1997) for each of the two research questions.
The independent variable was type of captioning (NC, KC, FC, and FCHK); the covariate was the score
on the vocabulary size test. There were two sets of dependent variables: the comprehension tests (research
question 1) and the vocabulary tests (research question 2). We set the significance level of the p-value at
.05 in all statistical analyses.
As an effect size measure, we used partial eta squared (p), which refers to the proportion of total
variance explained by an effect in which the effects of other independent variables and interactions are
partialled out (Richardson, 2011, p. 135). Effect size values have the advantage of being independent of
sample size. We used Cohens rules of thumb (1988) for interpretation: small, p > .0099, medium, p >
.0588, and large, p > .1379.
RESULTS
Vocabulary Size Test
Table 6 summarizes the test scores for each of the three frequency bands and shows that participants were
most proficient on the first part, which assessed knowledge of the 2,001 to 4,000 most frequent words.
Scores decreased as the word frequency level increased (see Table 6). Since vocabulary size is an
estimate of language proficiency, scores indicate that participants proficiency level in the current study
ranged from intermediate to high intermediate. This is also the required level at the end of high school
education according to the criteria of the Common European Framework of Reference (Council of
Europe, 2011). A one-way ANOVA revealed that the participants in the four conditions did not differ
significantly in terms of vocabulary size, F(3, 129) = 0.01, p = .998, p < .01. The 50-item size test had
an acceptable reliability index (N = 133, Cronbachs alpha = .78) and was used as a covariate in further
analyses in order to control for learner variables.
Table 6. Mean Scores and Standard Deviations on Vocabulary Size Test
All students
(N=133)
SD
NC
(n = 32)
M
FC
(n = 30)
SD
KC
(n = 34)
SD
FCHK
(n = 37)
Word Frequenc Items
SD
SD
2,0014,000
21
16.47
2.63
16.81
2.09
16.17
3.12
16.41
2.35
16.46
2.91
4,0015,000
15
9.87
2.20
10.03
2.07
9.83
2.51
9.97
2.10
9.68
2.21
5,0017,000
14
7.83
2.56
7.44
2.99
8.27
2.63
7.68
2.54
7.97
2.12
Total score
50
34.17
6.13
34.28
6.20
34.27
7.12
34.06
5.65
34.11
5.89
Research Question 1
We checked the correlations between the three comprehension tests and found that the scores of
comprehension test 1 correlated significantly with test 2 (r = .39, p < .001) and test 3 (r = .30, p = .001);
test 2 correlated significantly with test 3 (r = . 35, p < .001). In order to take into account the relationship
between the dependent variables, we performed a MANCOVA (Tacq, 1997).
Table 7 summarizes mean scores on the comprehension tests. The results of the MANCOVA contradicted
128
our initial hypotheses and showed type of captioning had no significant effect of on comprehension
scores, Wilkss lambda, F(9, 301.93) = 0.40, p = .935, p = .01.
Table 7. Mean Scores and Standard Deviations on the Comprehension Tests
Comprehension 1
Comprehension 2
Comprehension 3
All students
(N = 131)
NC
(n = 32)
SD
11.21
4.81
12.11
2.63
1.33
2.88
10.95
4.78
12.19
FC
(n = 28)
SD
2.76
1.30
2.75
KC
(n = 34)
SD
11.70
5.11
12.50
2.32
1.46
3.68
FCHK
(n = 37)
SD
10.79
4.63
11.91
2.28
1.43
2.48
SD
11.43
4.78
11.92
3.02
1.19
2.73
Note. The maximum scores of comprehension tests 1, 2, and 3, were 16, 7, and 18, respectively.
The covariate vocabulary size was significantly related to the comprehension scores, Wilkss lambda,
F(3, 124) = 8.81, p < .001, and had a large effect: p = .18. The positive b-values for each comprehension
test (see Table 8), indicated a positive relationship between vocabulary size and comprehension: the
larger ones vocabulary, the better the comprehension score.
Table 8. F-statistics for Vocabulary Size
Test
Comprehension 1
Comprehension 2
Comprehension 3
Source
Voc size (covariate)
Error
df
10.11
1
1
.07
.12
.05
.05
**
.15
.18
.002
5.93
.016
21.59
p
*
<.001
126
Note. * p < .05. **p < .001.
Research Question 2
We checked the correlations between the four dependent variables measuring vocabulary learning (cf.
procedure research question 1). Table 9 displays significant correlations between all tests.
Table 9. Correlations Between the Four Dependent Variables Measuring Vocabulary Learning
Form
recognition
Form recognition
1.00
Clip
association
.85*
Clip association
.85*
1.00
Meaning recognition
.46*
.52*
Meaning recall
.40*
.43*
Meaning
recognition
Meaning
recall
.46*
.40*
.52*
.43*
1.00
.45*
.45*
1.00
Note. *p < .001.
We found the highest mean scores on the form recognition test (see Table 10). The retention scores were
highest for the meaning recognition test. Results of the MANCOVA in Table 11 indicate that type of
captioning significantly affected vocabulary learning, Wilkss lambda, F(12, 301.91) = 4.63, p < .001, p
129
= .14. Vocabulary size was significantly related to vocabulary learning, Wilkss lambda, F(4, 114) =
16.64, p < .001, p = .37.
Table 10. Mean Scores on the Vocabulary Tests
All students
(N = 122)
NC
(n = 30)
SD
FC
KC
(n = 27) (n = 31)
SD
SD
FCHK
(n = 34)
M
SD
SD
Form
recognition
9.87
3.17
7.13
3.26
11.07
2.54
11.16
2.33
10.15
2.82
Clip
association
7.63
2.90
5.73
3.17
8.78
2.52
8.23
2.17
7.85
2.84
Meaning
recognition
.59
.14
.53
.14
.60
.14
.62
.09
.63
.17
Meaning
recall
.14
.20
.13
.22
.16
.22
.13
.17
.14
.19
Note. The smaller number of participants on the vocabulary tests is due to the fact that some participants accidentally skipped
one or two pages due to double-sided printing of the test, which consisted of 9 pages. The maximum score on the form
recognition and clip association test was 17.
The analysis of the between-subjects effects (see Table 11) showed that type of captioning significantly
affected the first three components of vocabulary knowledge tested, that is, form recognition, clip
association, and meaning recognition. Vocabulary size scores were significantly related to all four
vocabulary components, and b-values reported in Table 11 indicate a positive relationship: the larger
ones vocabulary, the more vocabulary gains.
Table 11. Results of MANCOVA on Vocabulary Acquisition
Test
Source
Form recognition
Type of captioning
Clip association
Type of captioning
Meaning recognition
Meaning recall
Type of captioning
df
3
1
3
1
3
F
14.46
7.91
7.51
14.92
4.85
<.001**
.06
<.001
**
.16
<.001
**
.11
.11
**
.29
.003
46.90
<.001
Type of captioning
.03
.993

Error
37.27
.27
.006
<.001
b
.12
.16
.01
.00
**
.24
.02
117
Note. * p < .05. **p < .001.
The results of the post hoc analysis (Bonferroni correction) partially confirmed our hypotheses: captioned
groups significantly outperformed the control group on form recognition and clip association. However,
no differences between the captioning groups were found. For meaning recognition, only KC and FCHK
significantly outperformed NC. No significant differences between KC, FCHK, and FC were found.
Contrary to our hypothesis, we found no differences between the captioned groups and the control group
130
on the meaning recall test.

The effect sizes showed a decreasing effect of type of captioning (FC, KC, FHCK, or NC) as the
vocabulary component tested increased in difficulty. The effect size of type of captioning was largest for
form recognition (p = .27) and clip association (p = .16) and smaller for meaning recognition (p =
.11). The opposite pattern was found for the covariate (vocabulary size): the deeper the knowledge
component tested, the larger the effect size of the covariate (see Table 11).
Questionnaire
The descriptive statistics for each questionnaire statement are listed in Table 12. Results of the
comprehension statements revealed that all conditions gave similar scores regarding the understanding
and difficulty level of the questions.
Table 12. Questionnaire Results
All students
(N = 133)
M
SD
NC
(n = 32)
M
FC
(n = 30)
SD
KC
(n = 34)
SD
SD
FCHK
(n = 37
M
SD
Comprehension (RQ1)
1. I didnt understand the French
video clips and I had no idea
what to answer to the
comprehension questions.
2.00
.86
2.03
.82
1.83
2. I think viewing the video twice

was important to help me
understand it.
3.38
1.26
3.53
1.22
2.97
1.33 3.44 1.24
3.51 1.24
3. I would have preferred seeing

the video three times.
2.77
1.30
3.12
1.31
2.33
1.24 2.71 1.22
2.86 1.36
Video 1
1.86
.76
1.97
.82
1.70
.79 1.79
.85
1.95
.58
Video 2
1.79
.75
1.75
.72
1.83
.75 1.65
.85
1.92
.68
Video 3
2.22
.87
2.13
.91
2.13
.86 2.12
.98
2.46
.69
5. I was able to infer word

meaning from the textual
context.
3.49
.88
3.66
.97
3.37
.96 3.50
.83
3.43
.80
6. The video images made it

possible for me to infer the
meaning of unknown words.
3.13
.99
3.06
1.16
3.27
.94 3.06
.92
3.14
.95
7. If I had known that vocabulary

exercises were to follow, I
would have paid more attention
to the words.
3.82
1.24
3.84
1.08
3.53
.87 1.91
.87
2.19
.88
4. The comprehension questions

were easy (1); average (2);
difficult (3); very difficult (4).
Vocabulary (RQ 2)
1.36 3.94 1.35
3.92 1.16
Usefulness of (keyword) captioning

(RQs 1 & 2)
131
8. I did not need extra support to

understand the video.
2.96
1.16
9. If I had been able to activate

captioning, I would definitely
have done that.
3.66
.90
4.25
1.05
2.60
1.10 2.38
.99
2.22 1.08
10. I managed to understand the

videos because (keyword)
captioning was available.
3.60
1.05
3.90
.89 3.00 1.10
11. I was distracted by the presence

of captioning. Consequently, I
focused less on the audio.
2.23
1.15
1.73
1.05 2.62 1.02
2.27 1.22
12. Captioning helped me to notice

unknown words.
3.51
1.19
3.60
1.28 3.68 1.04
3.30 1.24
13. I learned new words thanks to

the availability of (keyword)
captioning.
3.30
1.10
3.52
1.15 3.18 1.14
3.24 1.01
3.92
.89
Note. 1= I do not agree at all, 5 = I completely agree.
With regard to the results of the vocabulary statements, we observed that all groups considered the textual
context and images helpful for word meaning inferences. Moreover, participants would have focused
more on the words if they had been forewarned of a vocabulary test. Learners in the NC group were most
confident about not needing extra support, but would activate captions if available.
When asked about the usefulness of captions, the FC and FCHK groups considered captioning more
useful than the KC group. The latter also provided the highest, yet still average, mean score for the
distraction induced by the support. The captioning groups were equally positive about the usefulness of
captions for recognizing and learning new words.
DISCUSSION
This study was motivated by the need to explore variations on captioning, either by providing the learner
with more salience and/or less textual density. The results suggest that (a) the type of captioning did not
affect comprehension scores; (b) learners vocabulary size positively correlated with comprehension and
incidental vocabulary learning; (c) the presence of on-screen text supported receptive vocabulary
learning; and (d) salience induced by KC or FCHK did not result in greater vocabulary gains than
captioning without salience (FC).
Research Question 1
The analyses revealed that participants in all conditions achieved similar scores on the comprehension
tests, which is at variance with our hypothesis. Thus, our findings only provide limited support for the use
of captions and keyword captions and therefore contradict previous research (e.g. Baltova, 1999; Chung,
1999; Markham, 2001). An important explanation for the similar test scores might lie in the information
targeted by the comprehension questions. First of all, the comprehension questions did not focus on the
TWs, in order to prevent an effect of the comprehension tests on vocabulary retention. Because learners
did not need the TWs to answer to the comprehension questions, the differences between the four
conditions might have been minimized. Perhaps the parts containing the TWs did cause problems for the
NC group, but this was not reflected in the comprehension test scores as the questions did not require the
TWs occurring in the captions. Second, because of the nature of the clips, which presented concrete
factual information, we could only ask literal comprehension questions, rather than a combination of
literal and inferencing questions (Buck, 2001). Third, it is possible that captioning groups decoded the
132
clips better than the NC group (Bird & Williams, 2002), a hypothesis that is supported by the higher
scores of the FC, KC, and FCHK groups on the form recognition test (cf. research question 2). Yet, our
findings suggest that a better decoding does not necessarily result in a better understanding of the
message.
It could be hypothesized that the clips were too easy for the participants and therefore minimized the
comprehension differences between the groups. Yet, what needs to be stressed is that the groups achieved
only intermediate scores on the comprehension tests, which shows that content understanding was
challenging for the participants. Although the usefulness of captions was not reflected in higher
comprehension scores, questionnaire results indicated that participants in the FC and FCHK groups
considered captions useful for understanding the clips (see Table 12, question 10). While the NC group
indicated that they did not need extra support to understand the videos, they unambiguously indicated that
they would use full captions if they had been available (see Table 12, question 9).
With respect to the keyword captions, we found that salience of the keywords for the KC and FCHK
groups did not help learners to outperform FC and NC on the comprehension test. Our results are in line
with Parks findings for lower intermediate and intermediate learners (2004) and therefore contradict
research on reading comprehension in which it has been demonstrated that salience and attention
allocation are important for successful text comprehension (Gaddy, van den Broeck, Sung, 2001).
Moreover, learners in the KC group reported lower mean scores on the usefulness of keyword captions
when compared to the FC and FCHK group (see Table 12, question 10). In order to reveal the potential of
keyword captioning, further research is necessary.
Finally, learners scores on the vocabulary size test were significantly related to their comprehension
scores, and had a large effect size. Our result provides additional evidence in support of previous research
on this relationship (e.g., Staehr, 2009; van Zeeland & Schmitt, 2012).
Research Question 2
Form Recognition and Clip Association
The results of our analyses partially confirm our hypothesis and reveal that the captioning groups (FC,
KC, and FCHK) significantly outperformed the control group on the form recognition test. Although
mean scores were generally lower for clip association, we found a similar pattern. Overall, our results
confirm the usefulness of captions at the decoding level and their capacity to help learners to isolate
words (Winke et al., 2010) and pay sufficient attention to these items. Our results are consistent with
Vanderplanks model (1990), in which the role of subtitles was considered crucial to taking out words
from captioned input. This ties in with previous research that has found positive effects of captions for
enhancing written form recognition (e.g. Neuman & Koskinen, 1992; Sydorenko, 2010). On the other
hand, we did not find evidence for the hypothesis that the salient KC and FCHK groups would
outperform the FC group on the form recognition and clip association test. Interestingly, the availability
of on-screen text, rather than salient input, had an overall positive effect on form recognition and clip
association. This finding was confirmed by our questionnaire results (see Table 12, questions 12 and 13)
which showed that the FC, KC, and FCHK groups found the availability of captioning equally useful for
recognition and word learning.
How can we explain that salience did not enhance noticing? In his model, Vanderplank (1990) claimed
that the taking out of language consists of a combination of attention and adaptation. More particularly,
he stated that learners might pay attention to words and select language attended to for own purposes
(p. 229). Although keywords were not made salient for the FC group, learners might have noticed these
words for different reasons.
One possibility is that unfamiliarity with the TWs drew learners attention and produced a conscious
focus on form which occurs particularly when new or striking expressions are used (Vanderplank,
133
1988, p. 276). By comparing the language used in the captioned video with their own lexis, learners might
have noticed gaps in their knowledge (Gass, 1997).
Another reason may be found at the level of test announcement, that is, the comprehension test might
have directed the learners attention towards particular words even though they were not textually salient.
Hulstijn (2001, p. 268) suggested that, on the one hand, learners can serve under an intentional
condition when trying to understand the text with the subsequent comprehension test in mind. On the
other hand, they may serve under an incidental condition in that they are being exposed to unfamiliar
words and are not aware that a vocabulary test will follow (p.268). Because learners were prompted to
focus on the meaning of the clips and were only forewarned of the comprehension test, learners in the FC
group might have recognized words as well as the salient groups because they considered them important
for the test and the meaning of the clips. This explanation seems plausible since the majority of the TWs
were considered keywords for comprehension by our experienced lecturers. Moreover, this finding
confirms the claim that learners can simultaneously attend to form and meaning when the form is
important for the meaning (Baltova, 1999; Van Patten, 1990).
Meaning Recognition and Recall
In order to determine the quality of processing, we included a meaning recognition and recall test.
Although we hypothesized that participants in the FCHK group would outperform participants in the
other conditions on the meaning recognition test, the group scored only slightly, but not significantly,
higher than the KC and FC groups. The results of the meaning recognition test differed from the previous
two vocabulary tests in that KC and FCHK significantly outscored only the NC group and therefore seem
to have processed the TWs more elaborately than the NC group. The FC group did not differ significantly
from the NC group. This finding might be explained by the fact that the TWs were better isolated for the
salient keyword groups. Prince (1996, p. 489) underlined the importance of isolating the word from the
context, so that context provides the means to identify the meaning of the new word. Yet, since the
salient keyword groups did not significantly outperform the FC group, this result should be interpreted
cautiously.
In order to find out how captioning might have helped word meaning inferences, learners were asked
about the usefulness of contextual and visual clues. Results revealed that learners in all conditions
considered the textual context most helpful and reported very similar scores (see Table 12, questions 5
and 6). A similar appreciation was found for the support of visual clues, although scores were slightly
lower. Given the rather abstract nature of most TWs, the audio-video correlation could indeed only
provide scarce visual clues.
Generally speaking, we found that participants made initial form-meaning connections for a series of
TWs. But which conditions fostered the most qualitative vocabulary processing? The results of the
meaning recall test were low and, contrary to our hypothesis, no differences were found between the
groups. These findings do not support previous research (e.g. Danan, 1992; Sydorenko, 2010; Winke et
al., 2010), which showed beneficial effects of captioning on meaning recall. Our study differs, however,
from the cited articles in that they focused on low intermediate and beginning students respectively.
Plausible explanations for the low gains might be the following: (a) Since captioning as such does not
provide concrete information on word meaning, learners are required to construct meaning based on
inferring processes. Yet, the meaning of difficult words may remain unknown even after two or more
viewings. (b) Because of the real-time nature (Buck, 2001, p. 6) of listening, learners are left with very
little time to infer word meaning from context, which might prevent them from hearing the rest of the
video (Goh, 2000). (c) It has been shown that inferring word meaning is a difficult, slow, and often an
unsuccessful process (Liu & Nation, 1985). Because we wanted to make sure that words were unfamiliar,
we were forced to focus on low frequency words, which might have a higher learning burden (Laufer,
2005, p. 234). Thus, the meaning recall test was a very demanding test because participants were expected
134
to watch the clips, read the captions, remember the content, derive the meaning of unknown words
successfully, and finally remember the meaning of a set of TWs after only two viewings of the clips.
Vocabulary Size
Our study corroborated the findings of previous research on the importance of vocabulary size for
successful vocabulary learning (Webb, 2010; Webb & Rodgers, 2009). We found medium and significant
effect sizes of vocabulary size for form recognition and clip association. The larger ones vocabulary size,
the less decoding load a video presents and the more time learners can spend on specific lexical items
(Goh, 2000; Pulido, 2007). Yet, the effect sizes for vocabulary size were considerably lower than the
effect size for type of captioning, which suggests that captioning is crucial to form recognition and clip
association.
For meaning recognition and recall, we found very large effect sizes for vocabulary size; these effects
were considerably higher than the effect sizes for type of captioning. Indeed, it is well documented that
greater lexical knowledge seems to facilitate guessing from context (Liu & Nation, 1985) and it has been
characterized as a critical factor (Nation, 2001, p. 233) for successfully inferring word meaning.
Pedagogical Implications
When teachers intend to use video for stimulating vocabulary acquisition, they should be encouraged to
use captioning because it might facilitate students recognition of unknown words and their making initial
form-meaning connections. Our results provide limited support for the use of on-screen text for
improving content comprehension. We therefore recommend that teachers adopt a staged video
approach (Danan, 1992) in which they gradually decrease the amount of text. Teachers could show the
same video first with FCHK or FC, then with KC and finally in a NC mode. By doing so, they might help
learners to progressively decrease the amount of support while at the same time optimizing word
recognition and chances for word learning.
Teachers should also be encouraged to address the importance of vocabulary size as it has a positive
effect not only on successful comprehension but also on vocabulary learning. Teachers could provide
opportunities for learners to enlarge their vocabulary size through exposure to captioned video or by
means of other incidental (e.g., extensive multimedia reading) or intentional (explicit vocabulary
learning) tasks.
CONCLUSION AND LIMITATIONS
Although this study has tried to respond to the need to explore variations on standard captioning, it also
presents a number of limitations. First of all, we could not assess understanding of the complete clips
because our comprehension tests had to focus on the parts that did not contain the TWs. Moreover, the
clips were short, which presents limitations in terms of the type of information targeted by the questions.
Our results therefore provide only limited insight into how different types of captioning can improve
comprehension. Second, this study has focused on a set of predefined TWs. Yet, learners might have
noticed and established form-meaning connections of other words in the videos they were not familiar
with. Because our vocabulary tests only included predefined TWs, we are unaware of any other learning
gains (Pulido, 2007). Third, we assessed noticing using an indirect measure (i.e., the form recognition
test). More accurate data on what exactly induced form recognition, taking into account potentially
moderating variables such as frequency of occurrence and the part of speech of the TWs (Webb, 2007),
might reveal when and why learners notice certain items. Finally, the written format of the vocabulary
posttests may have favored the captioning groups to some extent. Unlike the control group, they also
encountered the target words in their written form.
As a conclusion to this paper, we would like to indicate three future research directions. First of all, future
studies might investigate the extent to which captioning, and keyword captioning in particular, can be
135
enhanced by adding access to the target words meaning, such as a gloss containing the L1 translation
(Sydorenko, 2010; Webb, 2010). Research questions could focus on the use of such glosses and their
effectiveness for comprehension and vocabulary learning. Second, current research on captioning has
almost exclusively measured the value of short clips, by means of isolated experiments. As one reviewer
suggested, it may be interesting to look at the potential of captioning when used with full-length TV
programs (Rodgers & Webb, 2011). We should also encourage longitudinal research in order to
investigate whether systematic and long-term exposure to captioned video affects learners use of
captions and the effectiveness of these captions in terms of improved listening comprehension. Third, the
availability of on-line measures such as eye-tracking might shed more light on the time learners spend on
the captioning line (Winke, Gass, & Sydorenko, 2013) and provide an objectified measure to study the
role of attention in vocabulary learning (Godfroid, Housen, & Boers, 2010). Further research on the topics
mentioned above will undoubtedly allow us to gain a greater insight into how and when captioning can
help language learning.
APPENDIX A. List of 20 Target Words

# encounters in
clip
Target word
Type of word
Clip
Larguer
verb
Lego (2)
faire un tabac
multiword
Lego (2)
Assaut
noun
Lego (1)
fermentation
noun
Bnifontaine
a cartonne
multiword
Lego (1)
*berceau
noun
Lego (2)
amertume
noun
Bnifontaine
Houblon
noun
Bnifontaine
*solidifier
verb
Lego (2)
gravir les chelons
multiword
Lego (2)
Mrir
verb
Bnifontaine
Malt
noun
Bnifontaine
tre fond
multiword
Lego (1)
se disperser
verb
Lego (2)
*rcompenser
verb
Lego (1)
doper les ventes
multiword
Lego (2)
frler le naufrage
multiword
Lego (1)
Brasser
verb
Bnifontaine
divertissement
noun
Lego (1)
Levure
noun
Bnifontaine
Note. Words marked with * were not included in the analysis because too many participants were already familiar with the word
before the experiment (as measured by the vocabulary prior knowledge test).
136
NOTES
1. Because a formal learning context differs from normal viewing situations, a number of practical
limitations such as time constraints inevitably played a role in the selection of clips. In Flemish high
schools for example, one learning session lasts only between 50 and 60 minutes, which would make it
difficult to use long clips and exploit them in a meaningful way. In this context, short clips present
ecologically valid L2 viewing and practicing opportunities. The length of the clips in this study was also
similar to the clips used in previous studies on captioning effects (e.g. Park, 2004; Sydorenko, 2010;
Winke et al., 2010).
2. Video and clip are used interchangeably.
3. The differences in occurrences of the TWs may be considered a limitation of the clips used. Although
frequency of the TWs may have played a role for vocabulary learning in this study, as suggested by
previous research (Webb, 2007), results show that learners were able to recognize more than the four
TWs with higher occurrence (see Table 10). Nonetheless, results do not imply that they necessarily
noticed the four TWs with higher occurrence correctly. A detailed analysis of the aspects that induced
form recognition and vocabulary learning might clarify this aspect but was beyond the scope of the
present study.
4. The initial vocabulary size test consisted of 53 items. As the reliability test indicated that three items
correlated negatively, these items were left out of the analysis.
5. The concepts of meaning recognition and meaning recall are not clearly defined. We adopt the
distinction made by Laufer et al. (2004) between active and passive recognition and recall, whereby
passive recognition consists in choosing the meaning of the target word from the four options provided
(p. 207) and passive recall consists in translating the target form.
6. This study did not include delayed post-testing. Hulstijn (2003) argued that a study measuring the
effectiveness during a learning session in which words are presented for the first time, requires only an
immediate posttest (p. 372) because it is difficult to ascertain whether learners scores on the delayed
posttest should be ascribed to the experimental treatment or to other learning opportunities which
occurred in the period between the learning session and the delayed test (Hulstijn, 2003; Nation & Webb,
2011). Nation and Webb pointed out that this problem could be avoided by using nonsense words instead
of real TWs. Yet, the use of nonsense words in a study based on authentic video seems inconvenient as it
would considerably affect the original speech stream.
7. We used the VKS to measure meaning recall rather than the progress made. Using the VKS as
described in Nation and Webb (2011) might have offered more precise data concerning learners
progress.
ACKNOWLEDGEMENTS
We are grateful to the LLT reviewers for their insightful comments and suggestions. All remaining errors
are our own. We would also like to thank Dr. Carmen Eggermont for allowing us to conduct this study in
her classes (KU Leuven Kulak) and the team of Roeland (Saint-Dizier 4, 2011) for their help in
organizing the pilot study for this experiment. Finally, we would also like to thank the Media and
Learning Unit (KU Leuven) and Dr. Hans Paulussen for developing the video player.
137
ABOUT THE AUTHORS

Maribel Montero Perez obtained her PhD in 2013 and works within the iMinds - ITEC research group at
the KU Leuven Kulak. Her research interests concern the effectiveness of video for L2 listening and
vocabulary acquisition and the use of eye-tracking for investigating vocabulary learning from watching
captioned video.
E-Mail: maribel.monteroperez@kuleuven-kulak.be
Elke Peters is an assistant professor of English at the Department of Language and Communication at the
KU Leuven. Her research interests include educational language policy and instructed second language
acquisition, particularly vocabulary acquisition and learning of collocations.
E-Mail: elke.peters@arts.kuleuven.be
Geraldine Clarebout is head of the Center for Medical Education of the faculty of medicine, KU Leuven.
Her PhD research focused on the use of tools in learning environments and the optimization of tool use.
She is currently involved in quality assurances processes in education.
E-Mail: geraldine.clarebout@med.kuleuven.be
Piet Desmet is full professor of French and Applied Linguistics and Foreign Language Methodology at
KU Leuven Kulak and coordinator of the iMinds - ITEC (Interactive Technologies) research group. His
research takes a particular interest in adaptive learning environments, mobile language learning, serious
gaming, parallel corpora for CALL and writing aids.
E-Mail: piet.desmet@kuleuven-kulak.be
REFERENCES
Baltova, I. (1999). The effect of subtitled and staged video input on the learning and retention of content
and vocabulary in a second language. Unpublished doctoral dissertation. University of Toronto, Toronto,
Canada.
Bird, S. A., & Williams, J. N. (2002). The effect of bimodal input on implicit and explicit memory: An
investigation into the benefits of within-language subtitling. Applied Psycholinguistics, 23(4), 509533.
Brett, P. (1995). Multimedia for listening comprehension: The design of a multimedia-based resource for
developing listening skills. System, 23(1), 7785.
Brett, P. (1998). Using multimedia: A descriptive investigation of incidental language learning. Computer
Assisted Language Learning, 11(2), 179200.
Bruton, A. (2009). The vocabulary knowledge scale: A critical analysis. Language Assessment Quarterly,
6(4), 3741.
Buck, G. (2001). Assessing Listening. Cambridge, UK: Cambridge University Press.
Chung, J. (1999). The effects of using video texts supported with advance organizers and captions on
Chinese college students listening comprehension: An empirical study. Foreign Language Annals, 32(3),
295308.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Erlbaum.
Council of Europe. (2011). Common European framework of reference for languages: Learning,
teaching, assessment (CEFR). Retrieved from
http://www.coe.int/t/dg4/linguistic/Source/Framework_FR.pdf
138
Danan, M. (1992). Reversed subtitling and Dual Coding Theory: New directions for foreign language
instruction. Language Learning, 42(4), 497527.
Danan, M. (2004). Captioning and subtitling: Undervalued language learning strategies. META, 49(1),
6777.
Gaddy, M. L., van den Broeck, P., & Sung, Y.C. (2001). The influence of text cues on the allocation of
attention during reading. In T. Sanders, J. Schimperoord, & W. Spooren (Eds.), Text representation:
Linguistic and psycholinguistic aspects (pp. 89110). Amsterdam: John Benjamins.
Garza, T. J. (1991). Evaluating the use of captioned video materials in advanced foreign language
learning. Foreign Language Annals, 24(3), 239258.
Gass, S. M. (1997). Input, interaction, and the second language learner. Mahwah, NJ: Lawrence Erlbaum
Associates.
Gass, S. M. (1999). Discussion: Incidental vocabulary learning. Studies in Second Language Acquisition,
21, 319333.
Godfroid, A., Housen, A., & Boers, F. (2010). A procedure for testing the Noticing Hypothesis in the
context of vocabulary acquisition. In M. Ptz & L. Sicola (Eds.), Cognitive processing in second
language acquisition (pp. 169197). Amsterdam: John Benjamins.
Goh, C. C. (2000). A cognitive perspective on language learners listening comprehension problems.
System, 28(1), 5575.
Grgurovi, M., & Hegelheimer, V. (2007). Help options and multimedia listening: Students use of
subtitles and the transcript. Language Learning & Technology, 11(1), 4566. Retrieved from
http://llt.msu.edu/vol11num1/pdf/grgurovic.pdf
Guillory, H. G. (1998). The effects of keyword captions to authentic French video on learner
comprehension. CALICO Journal, 15(13), 89108.
Horst, M., Cobb, T., & Meara, P. (1998). Beyond a Clockwork Orange: Acquiring second language
vocabulary through reading. Reading in a Foreign Language, 11, 207223.
Huang, H.-C., & Eskey, D. E. (19992000). The effects of closed-captioned television on the listening
comprehension of intermediate English as a second language (ESL) students. Journal of Educational
Technology Systems, 28(1), 7596.
Huckin, T., & Coady, J. (1999). Incidental vocabulary acquisition in a second language. Studies in Second
Language Acquisition, 21(2), 181193.
Hulstijn, J. H. (2001). Intentional and incidental second language vocabulary learning: A reappraisal of
elaboration, rehearsal and automaticity. In P. Robinson (Ed.), Cognition and second language instruction
(pp. 258286). Cambridge: Cambridge University Press.
Hulstijn, J. H. (2003). Incidental and intentional learning. In C. Doughty, & M. H. Long (Eds.),
Handbook of second language acquisition (pp. 349381). Malden, MA: Blackwell.
King, J. (2002). Using DVD feature films in the EFL classroom. Computer Assisted Language Learning,
15(5), 509523.
Laufer, B., Elder, C., Hill, K., & Congdon, P. (2004). Size and strength: Do we need both to measure
vocabulary knowledge. Language Testing, 21(2), 202226.
Laufer, B. (2005). Focus on form in second language vocabulary learning. EUROSLA Yearbook, 5(1),
223250.
139
Lee, J. F., & Van Patten, B. (2003). Making communicative language teaching happen. New York, NY:
McGraw Hill.
Liu, N., & Nation, I. S. P. (1985). Factors affecting guessing vocabulary in context. RELC Journal, 16(1),
3342.
Lonsdale, D., & Le Bras, Y. (2009). A frequency dictionary of French core vocabulary for learners.
London, UK: Routledge.
Markham, P. (1999). Captioned videotapes and second-language listening word recognition. Foreign
Language Annals, 32(3), 321328.
Markham, P. (2001). The influence of culture-specific background knowledge and captions on second
language comprehension. Journal of Educational Technology Systems, 29(4), 331343.
Milton, J., Wade, J., & Hopkins, N. (2010). Aural word recognition and oral competence in English as a
foreign language. In R. Chacn-Beltrn, C. Abello-Contesse, & M. del Mar Torreblanca-Lpez (Eds.),
Insights into non-native vocabulary teaching and learning (pp. 8398). Clevedon, UK: Multilingual
Matters.
Nation, I.S.P. (2001). Learning vocabulary in another language. Cambridge: Cambridge University
Press.
Nation, I. S. P. (2006). How large a vocabulary is needed for reading and listening? The Canadian
Modern Language Review, 63(1), 5982.
Nation, I. S. P., & Webb, S. (2011). Researching and analyzing vocabulary. Boston, MA: Heinle.
Neuman, S. B., & Koskinen, P. (1992). Captioned television as comprehensible input: Effects of
incidental word learning from context for language minority students. Reading Research Quarterly,
27(1), 95106.
Park, M. (2004). The effects of partial captions on Korean EFL learners listening comprehension.
Unpublished doctoral dissertation. University of Texas at Austin, Austin, TX.
Paulussen, H., Macken, L., Trushkina, J., Desmet, P., & Vandeweghe, W. (2006). Dutch parallel corpus:
A multifunctional and multilingual corpus. Cahiers de lInstitut de Linguistique de Louvain, 32(14),
269285.
Peters, E. (2006). Learning second language vocabulary through reading: Three empirical studies into the
effect of enhancement techniques. Unpublished doctoral dissertation. University of Leuven, Leuven,
Belgium.
Prince, P. (1996). Second language vocabulary learning: The role of context versus translations as a
function of proficiency. The Modern Language Journal, 80, 478493.
Pujol, J.T. (2002). CALLing for help: Researching language learning strategies using help facilities in a
web-based multimedia program. ReCALL, 14(2), 235262.
Pulido, D. (2007). The relationship between text comprehension and second language incidental
vocabulary acquisition: A matter of topic familiarity? Language Learning, 57(1), 155199.
Richardson, J. (2011). Eta squared and partial eta squared as measures of effect size in educational
research. Educational Research Review, 6, 135147.
Robin, R. (2007). Commentary: Learner-based listening and technological authenticity. Language
Learning & Technology, 11(1), 109115. Retrieved from http://llt.msu.edu/vol11num1/pdf/robin.pdf
Rodgers, M.P.H., & Webb, S. (2011). Narrow viewing: The vocabulary in related television programs.
140
TESOL Quarterly, 45(4), 689717.

Rost, M. (2002). Teaching and researching listening. London, UK: Longman.
Selva, T., Verlinde, S., & Binon, J. (2002). Le Dafles, un nouveau dictionnaire lectronique pour
apprenants du franais. In A. Braasch (Ed.), Proceedings of the Tenth EURALEX International Congress,
(pp. 199208). Copenhagen, Denmark: CST.
Staehr, L. S. (2009). Vocabulary knowledge and advanced listening comprehension in English as a
foreign language. Studies in Second Language Acquisition, 31, 577607.
Sydorenko, T. (2010). Modality of input and vocabulary acquisition. Language Learning & Technology,
14(2), 5073. Retrieved from http://llt.msu.edu/vol14num2/sydorenko.pdf
Tacq, J. (1997). Multivariate analysis techniques in social science research: From problem to analysis.
London, UK: Sage.
Van Patten, B. (1990). Attending to form and content in the input: An experiment in consciousness.
Studies in Second Language Acquisition, 12, 287301.
Van Patten, B., Williams, J., & Rott, S. (2004). Form-meaning connections in second language
acquisition. In B. Van Patten, J. Williams, S. Rott, & M. Overstreet (Eds.), Form-meaning connections in
second language acquisition (pp. 126). Mahwah, NJ: Lawrence Erlbaum.
van Zeeland, H., & Schmitt, N. (2012). Lexical coverage in L1 and L2 listening comprehension: The
same or different from reading comprehension? Applied Linguistics, 33(5), 124.
Vandergrift, L. (2011). Second language listening: Presage, process, product, and pedagogy. In E. Hinkel
(Ed.), Handbook of research in second language teaching and learning (pp. 455471). New York, NY:
Routledge.
Vanderplank, R. (1988). The value of teletext sub-titles in language learning. English Language Teaching
Journal, 42(4), 272281.
Vanderplank, R. (1990). Paying attention to the words: Practical and theoretical problems in watching
television programmes with unilingual (CEEFAX) sub-titles. System, 18(2), 221234.
Vanderplank, R. (2010). Dj vu? A decade of research on language laboratories, television, and video in
language learning. Language Teaching, 43(1), 137.
Webb, S. (2007). The effects of repetition on vocabulary knowledge. Applied Linguistics, 28(1), 4665.
Webb, S. (2010). Using glossaries to increase the lexical coverage of television programs. Reading in a
Foreign Language, 22(1), 201221.
Webb, S. (2011). Selecting television programs for language learning: Investigating television programs
from the same genre. International Journal of English Studies, 11(1), 117136.
Webb, S., & Rodgers, M. P. H. (2009). Vocabulary demands of television programs. Language Learning,
59(2), 335366.
Wesche, M., & Paribakht, T. S. (1996). Assessing second language vocabulary knowledge: Depth vs.
breadth. The Canadian Modern Language Review, 53, 1340.
Winke, P., Gass, S., & Sydorenko, T. (2010). The effects of captioning videos used for foreign language
listening activities. Language Learning & Technology, 14(1), 6586. Retrieved from
http://llt.msu.edu/vol14num1/winkegasssydorenko.pdf
Winke, P., Gass, S., & Sydorenko, T. (2013). Factors influencing the use of captions by foreign language
learners: An eye-tracking study. The Modern Language Journal, 97(1), 254275.
141

http://llt.msu.edu/issues/february2014/mcneil.pdf

pp. 142159
ECOLOGICAL AFFORDANCE AND ANXIETY IN AN ORAL

ASYNCHRONOUS COMPUTER-MEDIATED ENVIRONMENT
Levi McNeil, Sookmyung Womens University
Previous research suggests that the affordances (van Lier, 2000) of asynchronous
computer-mediated communication (ACMC) environments help reduce foreign language
anxiety (FLA). However, FLA is rarely the focus of these studies and research has not
adequately addressed the relationship between FLA and the affordances that students use.
This study explored sources of FLA in an oral ACMC environment, and how the
affordances perceived by students in this environment correlated with FLA. One class of
Korean EFL university students (n = 15) completed voiceboard tasks for eight weeks.
Affordance data were collected with a questionnaire, and FLA was measured qualitatively
and by employing an adapted version of the foreign language classroom anxiety scale
(Horwitz, Horwitz, & Cope, 1986). The results suggest that students experience FLA in
particular ways using the voice board, and that some sources of anxiety are similar to
those reported in face-to-face contexts and others are unique to ACMC contexts.
Additionally, this study found a moderate correlation between the total use of affordances
and FLA, with some affordances being associated with reduced anxiety and some
associated with higher anxiety. The study discusses these findings and identifies avenues
for future research examining the interplay between the ACMC environment and FLA.
Key words: Affordance; Foreign Language Anxiety; Asynchronous CMC; Voiceboards;
Ecological Linguistics.
APA Citation: McNeil, L. (2014). Ecological affordance and anxiety in an oral
asynchronous computer-mediated environment. Language Learning & Technology 18(1),
142159. Retrieved from http://llt.msu.edu/issues/february2014/mcneil.pdf
Received: December 27, 2012; Accepted: June 20, 2013; Published: February 1, 2014
Copyright: Levi McNeil
INTRODUCTION
Research from the past 20 years has consistently demonstrated an inverse relationship between foreign
language anxiety (FLA) and second language (L2) achievement (see Horwitz, 2001 for an overview).
Although anxiety can be conceptualized as a general personality trait exhibited across many situations
(i.e., trait anxiety), or related to a particular moment in time (i.e., state anxiety), many view FLA as a
situation specific anxiety, which occurs consistently over time within a well-defined situation (MacIntyre
& Gardner, 1991). In this regard, FLA is the feeling of tension and apprehension specifically associated
with second language contexts (MacIntyre & Gardner, 1994, p. 284) that some liken to public speaking
(Horwitz, 2001), stage fright, or test anxiety (Horwitz, 2010). As a situation specific construct, FLA can
be explored from ecological perspectives of language learning (van Lier, 2000, 2004) that examine the
interconnectedness of human cognition and the environment.
While FLA disrupts cognitive processing (MacIntyre and Gardner, 1994) and potentially limits one
meditational tooloral collaboration with others (Baker & MacIntyre, 2000)computer-mediated
communication (CMC) environments are alternative mediating sources. CMC research suggests that these
contexts play important roles in reducing FLA (e.g., Chun, 1994; Kern, 1995). However, as Baralt and
Gurzynski-Weiss (2011) correctly point-out, many CMC studies fail to directly measure FLA.
Additionally, studies examining voiceboardsasynchronous platforms that allow learners to record and
review audio files and post them for private or group viewing and commentingoffer anecdotal evidence
142
Levi McNeil
Ecological Affordance and Anxiety
that students experience decreased levels of FLA while communicating with them (Poza, 2011; Song,
2009; Sun, 2009). Yet, in these contexts we still know little about how anxiety functions and how aspects
of these contexts perceived by learners relate to FLA.
The current descriptive study addressed these gaps by exploring student perceptions of anxiety and, by
drawing from the concept of affordance within ecological linguistics, the relationship between student use
of the asynchronous CMC (ACMC) environment and FLA. Before describing the methodology and
discussing the results of this study, relevant literature is reviewed examining: (a) FLA, language learning,
and affordance; and (b) studies of voiceboards in L2 classrooms.
LITERATURE REVIEW
FLA, Language Learning, and Affordance
A number of scholars focus on cognitive processing to illustrate the negative impact of FLA in language
learning. For example, Krashen (1982) posits that high levels of anxiety block input from reaching that
part of the brain responsible for language acquisition (p. 31). Others, such as MacIntyre and Gardner
(1994), show that FLA can affect not only input, or encoding, but also language storage and retrieval
processes. According to MacIntyre (1995), anxiety hinders these processes by creating scenarios in which
anxious students divide their cognitive resources between the task at hand and worry, whereas those who
do not experience anxiety will be able to process the information more quickly, more effectively, or both
compared to those who are distracted by task-irrelevant cognition (p. 92). These cognitive views
describe how FLA might interfere with information processing, but computational metaphors are
criticized for, among other reasons, overlooking the context in which thinking occurs.
At this point in the evolution of SLA theory, arguments abound in the literature (e.g., Atkinson, 2011;
Firth & Wagner, 1997, 2007) detailing the dangers of separating mind from situation. Instead of focusing
only on the individual, an ecological perspective of language learning examines the entire situation and
asks, what is it in this environment that makes things happen as they do? ( ... ) Ecology therefore involves
the study of context (van Lier, 2004, p. 11). This approach builds upon the tenets of sociocultural theory
(Vygotsky, 1978). Sociocultural theory (SCT) explains how all forms of human mental activity are
mediated by culturally constructed auxiliary means (Lantolf & Thorne, 2006, p. 59), which include
physical, social, and mental forms of mediation (Lantolf, 2011). In SCT, language learning is a process of
moving from other-regulation to self-regulation that co-occurs with changes in the quality and forms of
assistance (Aljaafreh & Lantolf, 1994). In this process, dialogue serves as a major mediating source for
cognitive development (e.g., McNeil, 2012; Swain & Lapkin, 1998). However, as demonstrated by
studies investigating willingness to communicate, FLA can limit social interaction (Baker & MacIntyre,
2000; MacIntyre & Charos, 1996). From a SCT perspective, the lack of communication due to anxiety
confines engagement in the co-construction of linguistic and content knowledge. In short, if students do
not communicate, there exist limited opportunities to receive the assistance from others that supports
language development.
While FLA potentially limits one meditational toolcollaborative dialogueecological approaches
highlight the diverse meaning making sources in the immediate environment that may facilitate
interaction in the face of FLA. By accounting for both mediated and immediate tools, ecological
perspectives acknowledge a wider range of contextual supports than traditional SCT. This is
accomplished by examining, the totality of relationships of an organism with all other organisms with
which it comes into contact (van Lier, 2004, p. 3). These other organisms include not only people but
also other symbolic and material objects in the physical and social world. In this way, the linguistic
environment immediately increases in complexity when we envisage a learner physically, socially, and
mentally moving around a multidimensional semiotic space (van Lier, p. 93). Greeno (1994) explains the
underlying epistemological stance of ecological approaches:
143
Levi McNeil
The framing assumption of ecological psychology is that cognitive processes are analyzed
as relations between agents and other systems. This theoretical shift does not imply a
denial of individual cognition as a theoretically important process. It does, however,
involve a shift of the level of primary focus of cognitive analyses from processes that can
be attributed to individual agents to interactive processes in which agents participate,
cooperatively, with other agents and with the physical systems that they interact with (...)
Research in ecological psychology has focused mainly on relations of agents with physical
systems and environments. (p. 337)
It follows, then, that language learning from an ecological perspective involves the intermixing of an
individuals mind and the resources in the environment that support engagement in the learning process.
As van Lier (2004) explains, language emerges from participation in the social world through a process
that begins with the learner perceiving and using objects in the environment to create rudimentary
meaning, which are then rendered into linguistic and symbolic meaning through further action. The
environmentally available resources form the basis for the concept of affordance, which originated from
Gibson (1979). He explains that, affordances of the environment are what it offers the animal, what it
provides or furnishes, either for good or ill (p. 127). In terms of language learning, these could include
gestures, body movement, gaze, or other items in the physical classroom or online environment. Thus,
affordances might, as van Lier argues, best be seen as pre-signs, that is, they may be the fuel that get
sign-making going (p. 93). Recognizing that the physical environment plays a vital role in language
learning has important implications for CMC research.
The material affordances of online environments have not escaped the attention of the CMC community.
For example, Lamy and Hampel (2007) discuss the affordances related to time and mode for different
ACMC tools. They do not discuss voiceboards explicitly, but since voiceboards share qualities with both
audio conferencing tools and asynchronous written forums, clear connections to the affordances of
voiceboards can be made. These affordances include time to reflect, comprehend, construct responses,
and use outside sources. In regards to mode, voiceboards promote learners analytic and metacognitive
skills and accuracy, while also allowing learners to use intonation, stress patterns, and expressive
vocalization.
Situating the learner in an environment with tools at ones disposal casts FLA into a new light.
Traditionally, FLA is linked to the struggles of expressing oneself with limited L2 linguistic and cultural
resources (Horwitz, Horwitz, & Cope, 1986). Ecological approaches, however, extend the meaning
making resources beyond just the learners linguistic and cultural abilities to include resources in the
context. Considering the potential affordances outlined above, voiceboards may offer students the
possibilities of using dictionaries and writing scripts. Thus, FLA may be qualitatively different across
contexts, depending on the potentiality for linguistic action presented by the local environment.
It is clear that the structural affordances of ACMC hold some potential in reducing FLA. However, the
physical affordances of ACMC alone are not enough to determine its impact. The affordances of the
environment must be perceived by the learner. Lamy and Hampel (2007) capture this idea:
Concerning affordances, user perceptions are more pertinent than the object itself. So it is
not just the material affordances of CMC that play a role in enhancing or limiting
communication, but also how people see them and the different practices that result from
their different perspectives. (p. 43)
It is the learners view that prompts Norman (1999) to offer the term perceived affordance. To him,
perceived affordance more accurately reflects the notion of affordance. Norman posits that the issue is
whether the user, not the designer or researcher, envisions the possibility of an action. In regards to the
ACMC voiceboards then, while they provide particular time and mode affordances, which increase
linguistic action potential, student perceptions of those affordances and how they are used to support
144
Levi McNeil
activity play important roles in adding or reducing feelings of anxiety. Studies examining the use of
voiceboards in L2 classrooms present some evidence that students perceive the affordances available with
this tool.
Voiceboard Studies in L2 Classrooms
Much of the voiceboard research focuses on development of speaking skills, although some data collected
to measure students overall perceptions of voiceboards help demonstrate a possible connection between
this tool and anxiety.
Hsu, Wang, and Comac (2008) conducted their study at a university in the United States. The study aimed
to help an instructor of an advanced English conversation course integrate voiceboards to extend student
oral practice and to evaluate student listening and speaking skills. The researchers also collected
qualitative data regarding the 22 students perceptions of voiceboards. Each week, the instructor posted
assignments, such as pronunciation practice and listening exercises with comprehension questions. The
results showed that the instructor believed that the voiceboard helped meet her needs to evaluate students
oral skills. Additionally, both the instructor and students responded that the voiceboard was easy to use.
Furthermore, in addition to enjoying voiceboard activities and reporting that their speaking abilities
improved, students felt that they had more confidence in English after completing the voiceboard tasks.
Since confidence is defined by some as the absence of anxiety, (i.e. Clement, 1987), this study indirectly
relates voiceboard use and language anxiety.
In regards to students speaking skills, Choi and Jung (2006) found that participating in voiceboard
activities enhanced perceived oral competence, but that this effect may not be observed for all learners. In
the study, nine Korean elementary EFL students were divided into three proficiency levels. Outside of
class, the students recorded onto voiceboards either dialogues from the course book or created their own
language output. Choi and Jung found that although the top two proficiency groups felt that they
improved in their speaking performance, the lower group did not. Furthermore, the lower group
occasionally avoided posting on the voice board, and when they did post, recorded dialogues directly
from the book. A question that remains from this study is whether these low-proficiency learners
perceived that they could use the affordance of extended time and the aid of other outside materials to
construct responses in addition to the course book. Perhaps perceiving these affordances could have
helped them complete the speaking tasks.
Sun (2009) reported more promising results. She investigated 46 Taiwanese university EFL students
processes of creating voiceboard posts and perceptions of voiceboards. Students completed 30 postings
and 10 responses to classmates throughout the semester. The data showed that students go through a
number of steps when constructing posts, including planning, rehearsing, and evaluating their potential
posts. In addition to finding that most students believed that voiceboards helped enhance their
communication and presentation skills, students expressed the anxiety-lowering potential of voiceboards.
For example, one student stated that voiceboards, really help me reduce speaking anxiety, and another
participant added that, unlike the classroom environment, blogs make me feel relaxed and thus speak
more fluently. I feel that I perform better on the voice blog than in face-to-face situations (p. 97). While
the focus of the study was not to investigate anxiety, the findings document how students use the
affordances of voiceboards and, peripherally, establish a relationship between this use and reduction in
FLA for some students.
Songs (2009) study further suggests that voiceboards may help anxious students. Song investigated the
impact of voiceboard use on student oral performance and collected data regarding student perceptions of
voiceboards. Thirty female university EFL learners recorded five minute diary posts six days a week for
10 weeks. The results showed that students significantly improved their oral performance when post-tests
were compared to pre-tests. Importantly, qualitative data indicate that these activities lowered anxiety. For
example, Song stated that students felt much more comfortable and natural speaking English (p. 137).
145
Levi McNeil
In the study, one student commented, I still cannot speak in English very fluently, but at least do not feel
the anxiety towards speaking English that much (p. 137). Although this study demonstrates at some level
that voiceboards facilitate feelings of lower FLA, it neither includes a direct measure of FLA nor offers
leads as to which affordance mediated these feelings.
The studies above provide smatterings of evidence to suggest a relationship between FLA and student
activity on voiceboards. Aside from Suns (2009) study, however, it is still not known how and to what
extent certain affordances are perceived in these contexts. Additionally, Baralt and Gurzynski-Weiss
(2011) make the point that FLA is rarely measured directly. In their study they found that learners
experience comparable levels of state anxiety when completing face-to-face (F2F) and synchronous CMC
tasks. Since the focus of the study was state anxiety in synchronous CMC, there are obvious differences
between their study and the current one. However, their observation regarding the limited number of
studies assessing anxiety holds for most of the voiceboard studies, although one voiceboard study has
measured FLA.
Poza (2011) aimed to compare anxiety in F2F and voiceboard contexts utilizing adapted versions of the
foreign language classroom anxiety scale (FLCAS, Horwitz et al., 1986) that reflected both environments.
English L1 speakers enrolled in two sections (n = 48) of the same Spanish course participated in the
study. For six weeks, students completed discussion-based assignments that required them to respond to
the instructors questions and reply to a classmates post. From the FLCAS data, Poza found that there
were few differences between anxiety in the classroom and anxiety on the voiceboard. Interview data
from four students, however, showed that they felt more relaxed in the online environment than the F2F
classroom due to the extra time to construct and edit their responses. In regards to negative evaluation, the
results showed that students perceived the two environments as mostly equal concerning anxiety related
to making mistakes. There were some differences in favor of the voiceboard environment for items
measuring anxiousness and feedback from the teacher and anxiety from comparisons to classmates L2
abilities. This study is important in that it directly measures FLA in an ACMC context and explores
qualitatively the students perceived affordances. However, this qualitative data came from only four
students, which may not accurately show how the voiceboard was used by most students. Additionally,
the study compared F2F to CMC environments, which often raises questions about validity due to the
difficulties in controlling a wide range of variables (Blake, 2009; Garrett, 2009). Indeed, the study does
not adequately detail the F2F tasks that were being compared to the voiceboard task. Instead of making
comparisons between the two settings, we need to understand anxiety in the voiceboard context itself.
Studies in F2F contexts have explored the classroom environment to uncover the sources and causes of
FLA. For example, Young (1991) identified six main causes of FLA. First, personal and interpersonal
aspects strongly influence anxiety. This category includes low self-esteem, competitiveness among
classmates, and social anxiety resulting from prospects of being evaluated by others during performance.
Second, FLA is caused by learner beliefs about language learning. This involves the contrast between
unrealistic language learning goals and the realities of language attainment within given timeframes
(Horwitz, 1988). In addition, the instructors beliefs about language can increase anxiety. Here,
intimidating students, not allowing them to work in smaller groups or pairs, and providing error
correction in a strict manner heighten student tension. In a similar vein, classroom procedures that
commonly require students to perform in front of the class or large groups of people facilitate
anxiousness. Finally, language testing, particularly tests that fail to create a close relationship between
what was taught in the course and what is included on tests, is a source of FLA. Yan and Horwitzs
(2008) study extends this list. They carried out a qualitative study that aimed to develop a groundedtheory model of the relationships among FLA and multiple factors. After interviewing 21 Chinese
university EFL participants with differing levels of anxiety, Yan and Horwitz found that although anxiety
is peripherally intertwined with a range of factors, such as parental influence, regional differences among
students, teacher characteristics, and class arrangement, three directly influenced FLA: (a) learning
146
Levi McNeil
strategies, (b) comparison with peers, and (c)language learning motivation and interest. These studies
raise questions regarding whether, how, and to what extent these sources operate in voiceboard contexts.
Understanding how these situational causes of FLA function in ACMC is important for teachers and
researchers in order to better structure CMC tasks.
The voiceboard literature seems to suggest that this tool holds potential for lessening anxiety. Yet we still
do not know the sources of anxiety in these contexts, what affordances learners perceive, or how the
affordances perceived directly relate to FLA. The current study aimed to address these issues, and it was
guided by the following research questions:
1. How and from what sources do students experience FLA in a voiceboard environment?
2. What is the relationship between the affordances perceived in a voiceboard environment and
FLA?
METHODS
Participants and Data Site
The participants in this study were second-semester university students enrolled in a required English
communication course at a mid-sized university in Seoul, South Korea. The class consisted of 15 (nine
female; six male) Korean EFL learners from different majors. These students had already completed one
required English course the previous semester. The final exam from that course is used by the university
to group students for second-semester English classes. This phone-pass test is administered by an
outside institution. To complete the test, students go to a computer lab on campus and participate in four
activities: reading sentences, repeating sentences, responding to questions, and organizing groups of
words into sentences. Responses are assessed for situation appropriateness, fluency, and grammatical
accuracy. After the university compiles the results for all participants, the scores are rank-ordered, and
students that score within 10 points of each other are assigned to groups and courses. In total, 219
students were assigned to 14 separate classes. The maximum score for the test is 80 and scores on the test
ranged from 20 to 78. The scores for the participants in the current study ranged from 43 to 51.
The teacher of this course, Mr. O, has taught at the university for 12 years and described the students
English proficiency as low-intermediate. The main goal of the course was to develop student
conversational abilities. To accomplish this, the department selects a textbook, and instructors are
required to follow the tasks in the textbook. According to Mr. O, these tasks consisted of both individual
listening activities and conversational pair work such as structuring and performing dialogues and asking
each other scripted questions.
The class met for two 75-minute sessions per week; one on Monday and one on Friday. There were two
different types of homework assignments. The first was a department-required online assignment that
consisted of listening comprehension and worksheet activities. These assignments were graded for course
points. The second type of assignment was completed on a class voiceboard outside of class, which was
integrated into the curriculum the last eight weeks of the semester. The voiceboard assignments were
implemented by Mr. O to provide additional conversational activities for students in the course. Due to
rigid structuring of the course syllabus at the institutional level, the completion of the voiceboard
assignments could not be assessed for points.
Using the Voice Board
During the seventh week of the semester, Mr. O introduced the voiceboard, Voxopop
(http://www.Voxopop.com/), in class. He demonstrated for students how to set up individual accounts
that they would later use outside of class and how to connect to the class private talk group. Voxopop
allows users to record audio directly to the website with a microphone. Users first record their speech and
147
Levi McNeil
then have the option of reviewing it. When users are satisfied with the product, they publish the post.
Thus, the affordances of Voxopop include time to replay posts to enhance comprehension and to monitor
and revise language production, both of which can be supported by outside resources, such as dictionaries,
thesauruses, and textbooks. Additionally, since posts are comprised of spoken language, users are
afforded opportunities to utilize acoustic information (e.g., intonation, stress) to construct meaning.
Voxopop posts are displayed linearly in the discussion thread (see Figure 1, which has been edited to
protect participants identities). The teacher modeled the process of recording, reviewing, publishing, and
listening to other posts several times during class. After introducing the voiceboard, students were
assigned a homework task (i.e., self-introduction) to demonstrate that they could use the technology. All
students completed this assignment and reported no problems.
Figure 1. Screenshot of the Voxopop interface.

For the remaining eight weeks of the semester, students were assigned one weekly voiceboard
assignment. The homework assignments followed a similar structure. For example, each encouraged
students to state an opinion about a topic that was related to the one introduced in class. In-class and
homework topics came from the course textbook. Units in the textbook were thematically organized and
consisted of a main topic (e.g., changes in the modern workforce) and several related sub-topics (e.g.,
stay-at-home dads). The main topics and some sub-topics were discussed in the classroom. Sub-topics
that were not discussed in class were considered for use as homework. The instructor then selected from
the remaining sub-topics those that he felt would appeal to student interests. The homework assignments,
then, were thematically similar, but not exactly the same, as those that took place earlier in the physical
classroom. In addition to providing their own opinions, students were instructed to respond to a
classmates opinion. Commenting was an important part of the assignment because it supported some of
the instructional goals of the course, such as constructing utterances for specific audiences and making
contextually appropriate responses. There were no length requirements for the posts. However, the
average length of student posts was 1 minute 14 seconds.
148
Levi McNeil

Two sources of data were used in the study. The first was an adapted version of the FLCAS (Horwitz et
al., 1986). The FLCAS is considered by many as the standard measure of language anxiety (Horwitz,
2010, p. 158), with valid and reliable psychometric properties (Horwitz, 1986). It includes 33 items with
5-point Likert-type response options for each item, and it is comprised of three sub-components:
communication apprehension, fear of negative evaluation, and test anxiety. An adapted FLCAS
(AFLCAS, see Appendix A) was employed in order to reflect the context and goals of the current study.
The final nine items on the AFLCAS were the result of a two-step selection process, which began with
eliminating items dealing with test anxiety since this was not related to the focus of the study. Then,
among the remaining items pertaining to communication apprehension and fear of negative evaluation,
those that were not responsive to either the speaking task the participants completed or the oral ACMC
environment were eliminated. For example, excluded items were those that were too general (e.g., It
wouldnt bother me at all to take more language classes; I worry about the consequences of failing my
foreign language class), related only to in-class environments (e.g., When I am on my way to language
class I feel very sure and relaxed; During language class, I find myself thinking about things that have
nothing to do with the course), and did not reflect the situational use of the voiceboard (e.g., I feel that
my language teacher is ready to correct every language mistake I make; I would probably feel
comfortable around native speakers of the foreign language). The remaining items were then
operationalized by changing the setting referred to within each item from in class to on Voxopop. All
items were translated into Korean, and the AFLCAS was given to the students during the last week of the
semester, immediately after students completed the eight voiceboard assignments.
The AFLCAS was scored by assigning a value to each of the five response options (e.g., 5 points for
Strongly Agree; 1 point for Strongly Disagree). One statement, item 4, was reversed scored because it is
stated in positive terms whereas the other items are negatively stated. Scores for each item were then
input into SPSS 17. Mean scores for each AFLCAS item served as one data source to answer research
question one. Additionally, AFLCAS data were used to compute a Spearmans rho (rs) correlation
coefficient to answer research question twothe relationship between affordances perceived and FLA.
Spearmans rho is a non-parametric test that is recommend when data are at the ordinal level (LarsonHall, 2010), such as the AFLCAS data in this study. Cronbachs alpha for the AFLCAS was .89.
The second data source was the voiceboard anxiety and affordance questionnaire (VAAQ, see Appendix
B), which was given two days after the AFLCAS. The questionnaire consisted of nine open-ended
questions and aimed to tap student perceptions of FLA in voiceboard environments (items 1, 9), and
affordances perceived (items 2 - 8). Since initial analyses of the AFLCAS showed that each participant
indicated elevated levels of FLA for at least one item (i.e., responses of agree or strongly agree),
questions 1 and 9 invited students to elaborate and characterize these feelings. The affordances of
voiceboards discussed in the literature (e.g., Lamy & Hampel, 2007; Sun, 2009) were consulted to create
questions related to the affordances perceived. The resulting questions asked students how many times
they used each affordance. While wording the questions in this way might give students the impression
that using affordances frequently was expected of them, it is important to note that participants responded
to the questions on the VAAQ anonymously. Responding anonymously provided at least some freedom to
state that they did not use an affordance. Such responses were not uncommon, a point that is reflected in
the relatively low use of some affordances which are presented and discussed later in the study. The
VAAQ was provided to the students in Korean and they were allowed to write their answers in Korean.
All responses were translated separately into English by two Korean-English bilinguals.
The VAAQ was analyzed in two ways. First, responses that elicited numbers from students (e.g., On
average, how many times per assignment did you use a dictionary or other source to help you understand
a classmates Voxopop post?) were copied directly from the VAAQ to SPSS 17 for correlation analyses.
On some occasions, students responded to these questions by providing a range of numbers, for example,
149
Levi McNeil
2-3. In these instances, the average of the two numbers was recorded (e.g., 2.5). The remaining items on
the on the VAAQ were analyzed following Bogdan and Biklens (2006) coding method in order to answer
research question one. First, responses for each item on the questionnaire were read for open-coding,
which produced salient categories that served as a primary framework for initial coding. These initial
categories were revised during subsequent rounds of analyses so that they reliably accounted for the data.
Once all responses could be codified, a final round of coding was completed.
RESULTS
Research Question 1How and from what sources do students experience FLA in a voiceboard
environment?
Data from the AFLCAS and VAAQ were collected to answer this question. Data from the AFLCAS show
that, overall, the participants responded that they felt moderate levels of FLA in the ACMC environment.
The highest possible AFLCAS score is 45, and the mean score on the AFLCAS for the participants was
22.85 (SD = 4.14). Dividing the mean score by the total number of items on the AFLCAS indicates that
most responses fell between the options of Disagree and Neutral. The precise means for each AFLCAS
item were calculated and are presented in Figure 2. The results show that responses for two items moved
towards responses of Agree. Students felt most anxious when comparing themselves to other students
(item 5), and they did not feel too confident when speaking English on the voiceboard (item 4).
Figure 2. Mean scores for each AFLCAS item.

On the VAAQ, students responded to two questions that asked them about anxiety in the oral ACMC
environment. The results for each question are shown in Table 1, which includes three columns. The first
identifies the question to which students responded and the second presents the qualitative categories
most frequently applied. The third column shows the number of responses coded for each category.
150
Levi McNeil
Table 1. Qualitative Responses of Anxiety on the Voiceboard

Question
Categories
Similar
responses
What made you feel nervous / anxious /

uncomfortable speaking English on Voxopop?
Pronunciation
Lack of immediate
feedback/non-verbal cues
Classmates negative evaluations
Grammar
Voxopop did not provide nonverbal cues
Voxopop offered more chances

to be negatively evaluated by
classmates
In what ways are the feelings of nervousness /

anxiousness different between speaking English
in class and speaking English on Voxopop?
The results for these questions indicate that pronunciation and grammar are sources of anxiety for
students in the oral ACMC environment. For example, comments such as, I am constantly concerned
about grammar and I always worry about my non-native pronunciation were common.
Other responses suggest that concerns about pronunciation relate to fears of being evaluated by
classmates. Some students explicitly referenced anxiety related to pronunciation:
On Voxopop, I record after I hear others recordingstheir pronunciation is superior to
mine and this makes me feel uncomfortable.
Others can listen to my recordings over and over. They probably dont care about the
contents but care more about pronunciation just like I do.
These results support the finding from the AFLCAS that anxiety can result from students comparing
themselves to each other (item 5). Although some participants did not specify precisely why they worried
about what peers thought of their performanceI always wonder what my classmates think of my
recordingsthe statements above suggest that pronunciation was a key concern.
Additionally, responses to both questions on the VAAQ show that the lack of non-verbal cues was
perceived as anxiety-inducing. Participants commented about the absence of non-verbal cues generally, I
feel more comfortable when speaking in class because non-verbal cues such as facial expressions or
gestures help me understand the communication, and when producing the target language:
In class I get facial expression clues to know whether they [classmates] are following what
I am saying or not.
In class I can at least use gestures as an aid to make others understand what I attempt to
say!
In sum, the results for research question 1 show that grammar, the absence of non-verbal cues,
comparisons of oneself to other classmates, and pronunciation can be sources of anxiety. Data also
suggest that pronunciation concerns are related to comparisons with classmates.
151
Levi McNeil
Research Question 2What is the relationship between affordances perceived and FLA?
This question was answered by correlating data from the AFLCAS and data from the participants
regarding the ways they used the affordances of the voiceboard. Table 2 presents the average number of
times each affordance was used by the participants when completing one voiceboard task, with the
exception of writing the response. For this question, students provided the number of times they wrote
responses for the eight assignments.
The results in Table 2 show that students rarely used other sources to understand classmates posts. On
the other hand, using dictionaries or outside resources to construct responses was the affordance
perceived most frequently, with students consulting these sources four times per assignment on average.
Around three times per assignment, students re-recorded their responses before publishing them to the
class discussion. When asked in what ways their post was revised when re-recording it (VAAQ question
7), the overwhelming response was pronunciation (10), followed by grammar (6), and fluency (5).
Table 2. The Average Frequency of Affordances Perceived
Affordance
Mean
SD
Replay to understand
2.54
1.32
Use source to understand
.54
.90
Use source for response
4.00
1.97
Re-record response
3.33
1.78
Write response
4.63
1.14
The frequencies of affordances perceived were then paired with the AFLCAS scores to compute
correlation coefficients. Table 3 presents the correlation between FLA in the oral ACMC environment
and the total affordances perceived, and FLA and each individual affordance. There exists a moderate
(rs = -.380, p = .223) inverse relationship between the total affordances perceived and FLA. This suggests
that as the use of affordances increased, FLA decreased to a limited degree. Among the individual
affordances perceived, the strongest relationship (rs = .550, p = .051) is between using outside sources to
understand classmates posts and FLA. The positive direction of this relationship suggests that the more
students used sources to help them understand, the higher their FLA was. Replaying to understand
classmates postings and using sources to produce responses had the strongest relationships associated
with decreases in FLA, although these were moderate and not statistically significant.
Table 3. Correlations Between FLA and Affordances Perceived
Total
Affordances
Replay to
Understand
Sources to
Understand
Sources for
Response
Re-record
Response
Write
Response
FLA
-.380
-.461
.550
-.359
-.008
-.061
Sig.
.223
.113
.051
.252
.979
.851
DISCUSSION, LIMITATIONS, AND IMPLICATIONS

The main goals of this study were to explore anxiety in an oral ACMC environment and to investigate the
relationship between affordances perceived and FLA. In regards to anxiety-related experiences, data from
the AFLCAS suggests, in a general way, that the students were not overly anxious when completing
152
Levi McNeil
voiceboard assignments. This finding supports previous oral ACMC studies that found that students
performing in voiceboard environments experience low levels of FLA (Poza, 2011; Song, 2009; Sun,
2009). However, while this finding could be interpreted to suggest that voiceboards offer more relaxed
environments than F2F classrooms, this interpretation is not warranted in the context of the current study.
For instance, the voiceboard was integrated into the curriculum as an extension of the classroom, where
the topics discussed were related to those already introduced in class. Furthermore, F2F data were not
collected and the voiceboard task differed from many of the F2F tasks in marked ways; the online posts
were not tied to course credit, which likely influenced both motivation and anxiety, and teacher feedback
was not provided. Moreover, a closer look at the AFLCAS and VAAQ data provided evidence that
students experienced FLA in particular ways. Data showed that pronunciation, grammar, the absence of
non-verbal cues, and comparisons to, and evaluations of, classmates were sources of FLA.
Studies in F2F contexts have also found many of these sources of anxiety and primarily explain them as
properties of the individual. For example, Yan and Horwitz (2008) and Young (1991) found that personal
and interpersonal characteristics such as competitiveness among classmates and nervousness associated
with evaluations from peers resulted in anxiety. Additionally, Youngs finding that learners beliefs about
language and unrealistic language learning goals produce anxiety could explain the participants
preoccupation with pronunciation in the current study. Indeed, recent views of English pronunciation hold
native-like models as an unrealistic language learning goal (e.g., Jenkins, 2002). Without dismissing the
relevance of individual characteristics, ecological perspectives examine the relations between the learner
and the context (Greeno, 1994), and the oral ACMC context in general, and its affordances in particular,
help explain the findings of anxiety in this study.
The voiceboard context that students worked within to complete the discussions lacked non-verbal cues.
These cues include facial expressions, gestures, and back-channels, and for lower proficiency students,
such as the ones in this study, these are likely valuable compensatory resources commonly employed
during the meaning making process. The absence of these resources in this environment led to feelings of
anxiousness, which may also relate to the other ways participants in this study experienced anxiety. For
instance, the lack of non-verbal communication cues may have further intensified the focus on
pronunciation and grammar. While Lamy and Hampel (2007) explain that voice modalities offer the
possibilities for intonation and inflection to amplify meaning, without the aid of non-verbal cues, this
modality alone may have placed even greater emphasis on the articulatory aspects of speech and syntactic
processing. In other words, without the broader semiotic resources of non-verbal cues to offer semantic
support, learners operated in scenarios that required them to closely examine pronunciation and grammar.
This explanation may be one reason why the participants used the replay function as often as they did.
Replaying the posts helped to understand classmates, but it also provided multiple opportunities to make
comparisons to others language abilities. The implications of this finding extend to both research and
practice. Researchers and teachers need to consider integrating and examining asynchronous video blogs
(vlogs), which offer many of the same affordances as voiceboards, but with the added feature of nonverbal cues. Additionally, as Yan and Horwitz (2008) suggested, creating small, similar-ability groups
may reduce anxieties related to comparisons to others, which could be done in online environments by
making separate discussion threads for each small group.
The finding that students experienced anxiety when comparing themselves to others contrasts the findings
in Poza (2011). In that study, students did not place importance on the fact that classmates would listen to
their posts, a point illustrated by a student during an interview, I didnt even think about it, honestly (p.
50). The contrast in student perceptions between that study and the current one could be explained in a
number of ways, such as differences in sociopolitical contexts. For example, since English often functions
in Korea as a factor in determining university acceptance and job attainment, classmates in this context
might view each others L2 performances differently than native English speakers in the United States
learning Spanish. Another explanation centers on the order in which the voiceboard tasks were completed
153
Levi McNeil
in conjunction with the classroom tasks. In Pozas study, students finished the voiceboard assignments
before having similar discussions in class, whereas in the current study these were completed in the
reverse order. While having the online discussions after the F2F in-class discussions offers students
opportunities to incorporate language that might have been learned during class, it might also place more
emphasis on this language. Future studies might address how the ordering of topically similar F2F and
voiceboard tasks relate to anxiety.
The findings from the correlation results further illustrate the complex interplay between the environment
and FLA. While anecdotal evidence reported in other studies links FLA with the ACMC environment, the
current study established an empirical one by directly investigating the affordances perceived in these
contexts. This study found that the relationship between the overall frequency of affordance use and FLA
was moderate, and many of the correlations for particular affordances were moderate or weak. These
findings can be explained in a number of ways.
First, it is important to recall that affordance includes both the good or ill offered by the environment
(Gibson, 1979), or as van Lier (2004) explains, the opportunity for or inhibition of action (p. 4).
Understanding this premise helps to explain the overall moderate correlation between FLA and the total
affordances perceived, and the extremely weak correlation between FLA and re-recording posts. The
environment afforded students a way to attend to pronunciation by revising their recordings, which they
did often in comparison to other aspects, but, as mentioned earlier, the replay affordance enabled
participants to make comparisons among each other. Likewise, although replaying classmates postings
provided opportunities for these comparisons, it also allowed students in this reduced cue context a way
of understanding each other, which is why we see a stronger correlation for this affordance and FLA.
These findings show that the affordances of oral ACMC environments relate to FLA in both positive and
negative ways.
Additionally, the co-dependency between individual abilities and affordance explain how some
affordances in this study related more strongly with FLA than others. As Greeno (1994) makes clear, the
concept of affordance cannot be understood without accounting for both. He argues that an affordance
relates attributes of something in the environment to an interactive activity by an agent who has some
ability, and an ability relates attributes of an agent to an interactive activity with something in the
environment that has some affordance (p. 338). Revisiting the perceived affordance of replaying others
posts demonstrates how affordance is intricately tied to ability. This affordance was moderately related
with lower levels of FLA, while, conversely, the use of outside sources to understand classmates
postings corresponded strongly with higher levels of anxiety. When considering individual ability, it
could be that the possibility for action provided by the environmentreplaying others posts
sufficiently melded with the students comprehension abilities, thus leaving the option of using outside
sources to understand unnecessary. This explanation is supported by a students comment. After
indicating that she never used outside resources when listening to her classmates, she explained that, I
just use my own knowledge to understand what they say. On the other hand, the affordance of using
outside resources to construct responses seems to have benefited FLA. In light of the voiceboard task in
this study, which prompted students to state their opinion to a question and respond to one classmates
opinion, participants relied upon their own abilities and replay to understand classmates. In order to
respond, however, the affordance of using outside resources was needed, and this explains the moderate
relationship between this affordance and decreases in FLA.
The findings are further accounted for in light of the methodological limitations of this study. For
example, the overall moderate correlation between FLA and the total affordances perceived could be due
to the difficulties in capturing a wide range of affordances. Some students noted that time and the physical
absence of others when performing helped ease FLA:
I felt more confident using it [voice board] because I didnt have to do speaking in real-
154
Levi McNeil
time situations because Im not good at improvising and correcting grammar and
vocabulary.
I am not too nervous on Voxopop because I dont have an audience that is present right in
front of me.
These affordances are challenging to capture quantitatively and underscore the importance of qualitative
data in examining affordance, while also demonstrating the need for further research in this area.
Using items from the literature to present affordances to participants in a fixed way also limited this
study. For example, the VAAQ did not illuminate how the environment where the technology was
accessed related to FLA. Additionally, while studies have suggested that preparing written scripts eases
FLA, which served as the basis for VAAQ item 8, it was not strongly correlated to FLA, and the question
may not accurately reflect how scripts function to lower anxiety. Students noted, for example, that it was
not writing the script that reduced FLA but reading itI dont feel nervous because I usually have a
script that I can simply read off of. Opening the response format would help uncover more of the
affordances perceived by the learners in these contexts and allow researchers to more accurately relate
this use to FLA. In a similar vein, collecting self-reported questionnaire data operationalizing perceived
affordances as the affordances used raises the question as to whether participants perceived certain
affordances and chose not to use them, or if they did not recognize certain possibilities in the ACMC
environment. Future studies collecting screen-capture and think-aloud data might help address this issue
by showing which affordances were signaled from the environment, which were acted upon, and why.
Findings from such data would assist teachers in knowing which functions of a tool need to be explicitly
taught, and which can be perceived directly from the environment.
CONCLUSION
Previous studies show that FLA can disrupt cognitive processing (MacIntyre & Gardner, 1994) and limit
social interaction (Baker & MacIntyre, 2000). Oral ACMC research has provided some evidence that this
environment might play a role in reducing FLA (Poza, 2011; Song, 2009; Sun, 2009). The current study
adopted the notion of affordance within ecological linguistics in order to examine in greater detail the
interplay between FLA and the potential for action signaled within the ACMC context. In line with
previous ACMC research, this study found that students are not overly anxious when completing
speaking tasks in this environment. Importantly, however, this study also found that integrating
voiceboards does not automatically eliminate foreign language anxiety. Instead, the environment and its
affordances are related with decreased and increased FLA. To explain this finding, it was posited that the
students abilities, the task, and the environment worked in consort to produce or ease anxiety. If this
explanation is accurate, it complicates the literature by suggesting that simply completing voiceboard
tasks does not entirely rid all foreign language anxiety and underscores the need for more research framed
by ecological perspectives. This study has pointed-out methodological concerns regarding how to more
accurately elicit and capture affordance data, which might assist future researchers examining the
affordances perceived by students in oral ACMC environments. Reliably accounting for the affordances
perceived in voiceboard contexts will allow us to investigate how they relate to FLA and, ultimately, the
ways in which they promote and impede language learning.
155
Levi McNeil
APPENDIX A. Adapted FLCAS

1.
I never feel quite sure of myself when speaking English on Voxopop.

Strongly Disagree
2.
Strongly Agree
Disagree
Neutral
Agree
Strongly Agree
Disagree
Neutral
Agree
Strongly Agree
Disagree
Neutral
Agree
Strongly Agree
Disagree
Neutral
Agree
Strongly Agree
Disagree
Neutral
Agree
Strongly Agree
I am afraid that other students will laugh at me when I speak in English on Voxopop.
Strongly Disagree
9.
Agree
I get nervous and confused when I am speaking English on Voxopop.

Strongly Disagree
8.
Neutral
I feel very self-conscious about speaking English in front of other students on Voxopop.
Strongly Disagree
7.
Disagree
I always feel that other students speak English better than I do on Voxopop.
Strongly Disagree
6.
Strongly Agree
I feel confident when I speak English on Voxopop.

Strongly Disagree
5.
Agree
It embarrasses me to volunteer my answers in English on Voxopop.

Strongly Disagree
Neutral
I start to panic when I have to speak English without preparation on Voxopop.

Strongly Disagree
3.
Disagree
Disagree
Neutral
Agree
Strongly Agree
I get nervous on Voxopop when the English teacher asks questions which I havent
prepared in advance.
Strongly Disagree
Disagree
Neutral
Agree
Strongly Agree
APPENDIX B. Voiceboard Anxiety and Affordance Questionnaire

Answers to the questions below will help me make the best possible English activities for students. Your
honest feedback and suggestions are greatly appreciated. You may answer the questions in Korean. You
are not required to write your name on this form; your responses will have no bearing for your score in
this class.
1.
What made you feel nervous/anxious/uncomfortable speaking English on Voxopop?
2.
On average, how many classmates postings did you listen to per assignment (1 Voxopop
discussion)?
3.
On average, how many times per assignment did you replay a classmates Voxopop post in order to
help you understand it?
4.
On average, how many times per assignment did you use a dictionary or other source to help you
understand a classmates Voxopop post?
5.
On average, how many times per assignment did you use a dictionary or other source to help you
make your posting on Voxopop?
6.
On average, how many times per assignment did you re-record your response on Voxopop before
you posted it?
7.
When you re-recorded you response, what aspects of your response did you change?
8.
For how many of the 8 Voxopop assignments did you write your response before recording it?
9.
In what ways are the feelings of nervousness/anxiousness different between speaking English in
class and speaking English on Voxopop? Why do you think so?
156
Levi McNeil
ABOUT THE AUTHORS

Levi McNeil (PhD Washington State University) is an associate professor in the Graduate School of
TESOL at Sookmyung Womens University. His research interests include CALL teacher education,
(new) literacies, and dialogic interaction.
E-mail: levi.mcneil@gmail.com
ACKNOWLEDGMENT
I thank the reviewers for their valuable comments. This study was supported by Sookmyung Womens
University Research Grant 2012.
REFERENCES
Aljaafreh, A., & Lantolf, J. P. (1994). Negative feedback as regulation and second language learning in
the zone of proximal development. The Modern Language Journal, 78, 465483.
Atkinson, D. (2011). Introduction: Cognitivism and second language acquisition. In D. Atkinson (Ed.),
Alternative approaches to second language acquisition (pp. 123). New York, NY: Routledge.
Baker, S. C., & MacIntyre, P. D. (2000). The role of gender and immersion in communication and second
language orientations. Language Learning, 50(2), 311341.
Baralt, M., & Gurzynski-Weiss, L. (2011). Comparing learners state anxiety during task-based
interaction in computer-mediated and face-to-face communication. Language Teaching Research, 15(2),
201229.
Blake, R. J. (2009). The use of technology for second language distance learning. The Modern Language
Journal, 93, 822835.
Bogdan, R. C., & Biklen, S. K. (2006). Qualitative research for education: An introduction to theories
and methods. Boston, MA: Pearson Education Group.
Choi, H., & Jung, B. (2006). The effects of the use of voice bulletin board on elementary school students
English speaking skills and affective domains: A case study. Multimedia-Assisted Language Learning,
7(3), 283311.
Chun, D. M. (1994). Using computer networking to facilitate the acquisition of interactive competence.
System, 22(1), 1731.
Clement, R. (1987). Second language proficiency and acculturation: An investigation of the effects of
status and individual characteristics. Journal of Language and Social Psychology, 5, 271290.
Firth, A., & Wagner, J. (1997). On discourse, communication and some fundamental concepts in SLA
research. The Modern Language Journal, 81, 285300.
Firth, A., & Wagner, J. (2007). Second/foreign language learning as a social accomplishment:
Elaborations on a reconceptualized SLA. The Modern Language Journal, 91, 798817.
Garrett, N. (2009). Computer-assisted language learning trends and issues revisited: Integrating
innovation. The Modern Language Journal, 93, 719740.
Gibson, J. J. (1979). The ecological approach to visual perception. Hillsdale, NJ: Lawrence Erlbaum.
Greeno, J. G. (1994). Gibsons affordances. Psychological Review, 101, 336342.
157
Levi McNeil
Horwitz, E. K. (1986). Preliminary evidence for the reliability and validity of a foreign language anxiety
scale. TESOL Quarterly, 20, 559562.
Horwitz, E. K. (1988). The beliefs about language learning of beginning university foreign language
students. The Modern Language Journal, 72, 182193.
Horwitz, E. K. (2001). Language anxiety and achievement. Annual Review of Applied Linguistics:
Language and Psychology, 21, 112126. Retrieved from
http://leighcherry.wikispaces.com/file/view/Anxiety+and+Language+Achievement+--+Horwitz.pdf
Horwitz, E. K. (2010). Foreign and second language anxiety. Language Teaching, 43(2), 154167.
Horwitz, E. K., Horwitz, M. B., & Cope, J. (1986). Foreign language classroom anxiety. The Modern
Language Journal, 70, 125132.
Hsu, H. Y., Wang, S. W., & Comac, L. (2008). Using audio-blogs to assist English-language learning: An
investigation into student perception. Computer Assisted Language Learning, 21(2), 181198.
Jenkins, J. (2002). A sociolinguistically based, empirically researched pronunciation syllabus for English
as an international language. Applied Linguistics, 23(1), 83103.
Kern, R. (1995). Restructuring classroom interaction with networked computers: effects on quantity and
characteristics of language production. The Modern Language Journal, 79, 457476.
Krashen, S. (1982). Principles and practice in second language acquisition. New York, NY: Pergamon.
Lamy, M-N., & Hampel, R. (2007). Online communication in language learning and teaching.
Basingstoke, UK: Palgrave Macmillan.
Lantolf, J. (2011). The sociocultural approach to second language acquisition: Sociocultural theory,
second language acquisition, and artificial L2 development. In D. Atkinson (Ed.), Alternative approaches
to second language acquisition (pp. 2447). New York, NY: Routledge.
Lantolf, J., & Thorne, S. (2006). Sociocultural theory and the genesis of second language development.
New York, NY: Oxford University Press.
Larson-Hall, J. (2010). A guide to doing statistics in second language research using SPSS. New York,
NY: Routledge.
MacIntyre, P. D. (1995). How does anxiety affect second language learning? A reply to Sparks and
Ganschow. The Modern Language Journal, 79, 9099.
MacIntyre, P. D., & Charos, C. (1996). Personality, attitudes, and affect as predictors of second language
communication. Journal of Language and Social Psychology, 15, 326.
MacIntyre, P. D., & Gardner, R. (1991). Methods and results in the study of anxiety in language learning:
A review of the literature. Language Learning, 41, 85117.
MacIntyre, P. D., & Gardner, R. (1994). The effects of induced anxiety on three stages of cognitive
processing in second language learning. Studies in Second Language Acquisition, 16, 117.
McNeil, L. (2012). Using talk to scaffold referential questions for English language learners. Teaching
and Teacher Education, 28(3), 396404.
Norman, D. A. (1999). Affordances, conventions, and design. Interactions, 3842.
Poza, M. I. C. (2011). The effects of asynchronous computer voice conferencing on L2 learners speaking
anxiety. The International Association for Language Learning Technology Journal, 41(1), 3363.
Retrieved from
http://www.iallt.org/iallt_journal/the_effects_of_asynchronous_computer_voice_conferencing_on_l2_lear
158
Levi McNeil
ners_speaking_anxiet
Song, J. W. (2009). An investigation into the effects of an oral English diary using a voice bulletin board
on English spoken performance. Multimedia-Assisted Language Learning, 12(1), 125150.
immersion students working together. The Modern Language Journal, 82, 320337.
Sun, Y. C. (2009). Voice blog: An exploratory study of language learning. Language Learning &
Technology, 13(2), 88103. Retrieved from http://llt.msu.edu/vol13num2/sun.pdf
van Lier, L. (2000). From input to affordance: Social-interactive learning from an ecological perspective.
In J. Lantolf (Ed.), Sociocultural theory and second language learning (pp. 245259). New York, NY:
Oxford University Press.
van Lier, L. (2004). The ecology and semiotics of language learning: A sociocultural perspective. Boston,
MA: Kluwer Academic Press.
Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes. Cambridge,
MA: Harvard University Press.
Yan, J. X., & Horwitz, E. K. (2008). Learners perceptions of how anxiety interacts with personal and
instructional factors to influence their achievement in English: A qualitative analysis of EFL learners in
China. Language Learning, 58(1), 151183.
Young, D. J. (1991). Creating a low-anxiety classroom environment: What does language anxiety
research suggest? The Modern Language Journal, 75(4), 426439.
159

http://llt.msu.edu/issues/february2014/aydinyildiz.pdf

pp. 160180
USING WIKIS TO PROMOTE COLLABORATIVE EFL WRITING

Zeliha Aydn, zyein University
Senem Yldz, Boazii University
This study focuses on the use of wikis in collaborative writing projects in foreign language
learning classrooms. A total of 34 intermediate level university students learning English
as a foreign language (EFL) were asked to accomplish three different wiki-based
collaborative writing tasks, (argumentative, informative and decision-making) working in
groups of four. Student wiki pages were then analyzed to investigate the role of task type
in the number of self and peer-corrections as well as form-related and meaning-related
changes. In addition, focus-group interviews and questionnaires were conducted to find
out how students would describe their overall experience with the integration of a wikibased collaborative writing project in their foreign language learning process. The results
revealed that the argumentative task promoted more peer-corrections than the informative
and decision-making tasks. In addition, the informative task yielded more self-corrections
than the argumentative and decision-making tasks. Furthermore, the use of wiki-based
collaborative writing tasks led to the accurate use of grammatical structures 94% of the
time. The results of the study also suggest that students paid more attention to meaning
rather than form regardless of the task type. Finally, students had positive experiences
using wikis in foreign language writing, and they believed that their writing performance
had improved.
Keywords: Online Teaching & Learning, Virtual Environments, Writing, Collaborative
Learning, CALL, Learners Attitudes, Second Language Acquisition
APA Citation: Aydin, Z., & Yildiz, S. (2014). Using wikis to promote collaborative EFL
writing. Language Learning & Technology 18(1), 160180. Retrieved from
http://llt.msu.edu/issues/february2014/aydinyildiz.pdf
Received: January 31, 2012; Accepted: July 31, 2013; Published: February 1, 2014
Copyright: Zeliha Aydin and Senem Yildiz
INTRODUCTION
Writing instruction in foreign language classes is especially important since good writing requires the
acquisition of a range of linguistic abilities, including grammatical accuracy, lexical knowledge, syntactic
expression and a range of planning strategies such as organization, style and rhetoric. Writing instructors
are not only responsible for emphasizing accuracy in formal language but they should also attend to the
establishment of meaning by providing their learners with meaningful contexts and authentic purposes for
writing. Although writing is an individual act, it is also a social and interactional process during which the
writer tries to express a purpose through responding to other people and texts. As Hyland (2003) argues,
including formal elements into writing instructions to achieve grammatical accuracy and ensuring that
students use those structures appropriately for specific purposes in a variety of writing contexts can be a
demanding task for second language writing teachers. Research shows that collaborative writing, both in
the first language (L1) and second language (L2) during which time learners jointly produce a text,
creates a sense of community among student writers and requires reflective thinking. The exchange of
feedback among students during a joint project allows them to notice linguistic and organizational
problems in their writing and would lead to error correction and grammatical accuracy (Donato, 1988;
Storch, 2002, 2005; Swain and Lapkin, 1998).
In the past few decades, the integration of technology in writing instructions, especially the development
160
Zeliha Aydn and Senem Yldz
Using Wikis to Promote Collaborative EFL Writing
of computer-supported social tools such as wikis and blogs, offer new ways of teaching by allowing
authoring, information sharing, knowledge building and easier collaboration opportunities among
learners. Previous studies that investigated the revision behaviors of learners during collaborative wiki
projects reveal that students pay attention to both form and meaning in their writing when writing
contexts based on carefully-designed collaborative writing assignments are employed (Kost, 2011; Lee,
2010).
The present study was conducted to contribute to the existing literature as regards of using wikis for
collaborative writing purposes by systematically examining the role of three different meaning-focused
tasks. These encourage the negotiation of meaning by providing learners with an aim to convey a
message to an audience, and on the interaction of learners and their revision behaviors in small groups
from the perspective of form- versus meaning- focused changes and self- versus peer-corrections while
jointly constructing texts. The research questions include:
1. In a wiki-based collaborative writing project, what is the role of task type in the number of
form-related changes and meaning-related changes?
2. What is the role of task type in the number of self-corrections and peer-corrections?
3. To what extent will the participants be accurate in making these self- and peer-corrections?
4. What are students perceptions of using wikis in collaborative projects?
Literature Review
Collaboration
There has been increasing attention in recent years towards a social cognitive perspective, which posits
that meaningful social interaction is fundamental for language learning since learning a language is
considered the outcome of a process of co-constructing ones L2 knowledge with peers rather than as a
result of an individuals construction of knowledge (Benson, 2003; Donato, 2000; Hauck & Youngs,
2008; Lee, 2002; Swain & Lapkin, 1998; van Lier, 1996). These ideas are grounded in Vygotskys (1978)
sociocultural theories of learning of and specifically his notion of the Zone of Proximal Development.
This theory describes learning as a social process, and emphasizes the fundamental role of social
interaction among learners in which a more knowledgeable peer provides scaffolding to the less
knowledgeable peer while completing a shared task. In L2 classrooms, collaborative tasks are expected to
engage learners and to provide scaffolding on each others use of language (Storch, 2002; Swain &
Lapkin, 1998). It is through this collaborative scaffolding that learners improve their linguistic and
cognitive capacities. According to Swain (2000), when interlocutors are engaged in a collaborative
activity, the language they use (whether spoken or written) mediates a process of joint constructive
interaction. She calls this collaborative dialoguea dialogue that constructs linguistic knowledge in
which what learners contribute becomes an object for reflection, receives peer feedback, addresses
linguistic problems and encourages modified output. However, writing is more than simply linguistic
accuracy. Swain (2000) also argues that tasks that engage students in collaborative dialogue might be
particularly useful for learning strategic processes as well as grammatical aspects of language (p.12).
From this perspective, collaborative interaction helps L2 learners in writing, especially when they are
asked to construct texts jointly and do peer-editing (Storch, 1999; Swain & Lapkin, 1998) by providing
opportunities for learners to focus on various aspects of writing such as grammatical accuracy, lexis and
discourse (Donato, 1994; Kim, 2008; Hirvela, 1999; Storch, 2002, 2009; Swain & Lapkin, 1998;
Wigglesworth & Storch, 2009). While working on a single text by taking group responsibility, learners
generate ideas, and pay attention to their language use and the organization of their ideas. Furthermore,
they become engaged in collaborative scaffolding by giving and receiving feedback, which promotes the
consideration of alternative uses of language and elaboration of ideas. Therefore, collaborative writing is
161
a powerful method of writing that encourages cooperation, critical thinking, peer learning and active
participation towards an end product (Hernandez, Hoeksema, Kelm, Jefferies, Lawrence, Lee & Miller,
2008).
Role of Tasks in Language Learning
Nunan (1992) defines task as a piece of classroom work which involves learners in comprehending,
manipulating, producing or interacting in the target language while their attention is principally focused
on meaning rather than on form (p.10). It is widely accepted that the nature of both oral and written
interaction is affected by the type of task (Cohen, 1994; Skehan, 1996). Pica, Kanagy, and Falodun (1993)
present a typology for communicative tasks according to interactional activities and communication
goals. According to their taxonomy, tasks which promote the greatest opportunities for learners to
experience the comprehension of input, feedback on production, and interlanguage modification are those
tasks which require each interactant to hold a different portion of the information to reach the task
outcome, Both interactants request and supply this information through the same or convergent goal, and
only one acceptable outcome. Other research shows that open-ended tasks in which learners co-construct
a piece of discourse, such as essays or reports, tend to encourage an increased amount of lexical and
morphosyntactic negotiations (Pellettieri, 2000; Storch, 2005; Storch & Wigglesworth, 2007). In
particular, tasks that require learners to use vocabulary, ideas and concepts that are beyond their
immediate knowledge are found to increase opportunities for interaction (Blake, 2000; Foster, 1998;
Pellettieri, 2000; Peterson, 2008; Pica, Kanagy, & Falodun, 1993). A review of task-based research by
Ellis (2003) reveals that those tasks which are non-familiar, require information exchange, and have twoway information gap, closed outcome, human/ethical topic, no contextual support, and narrative discourse
type promote the most meaning negotiation among learners. According to Skehans (1998) and Skehan
and Fosters (2001) limited attentional capacity model, learners need to prioritize whether to give their
attention to meaning or form. If a task demands too much attention to its content due to its complexity,
the learners attention will be allocated to its meaning; they will pay less attention to the language since
humans have a limited capacity to process information. In other words, tasks which are cognitively
demanding in their content are likely to draw attentional resources away from language forms,
encouraging learners to avoid more attention-demanding structures in favour of simpler language
(Skehan & Foster, 2001, p.189). In conclusion, the nature and type of task is expected to have an
influence on the writers focus on form versus content, and the amount and type of interaction among
writers during collaborative writing. This present study aims to further investigate this influence.
Research on Wikis
The development of new technologies offers new ways for language teachers to promote and enhance
collaboration in foreign language education. With the advent of Web 2.0 tools, more potential for
collaborative writing in the L2 classroom has emerged. Wiki is a web-based collaboration tool which can
be easily created, viewed and modified using any web browser. The asynchronous online collaboration
function offers language teachers new opportunities to combine all the essential parts of writing
instruction such as grammatical accuracy, appropriate use of grammatical forms in different contexts,
audience awareness, and multiple drafting and revising (Lund, 2008).
Wikis were found to provide a rewarding experience for students (Arnold, Ducate, & Kost, 2009; Ducate,
Anderson & Moreno, 2011; Kost, 2011; Lee, 2010; Lund, 2008; Mak & Coniam, 2008), supporting
learner autonomy (Kessler, 2009; Kessler, Bikowski & Boggs, 2012; Lee, 2010), resulting with an
aggregated output (Kost, 2011; Mak & Coniam, 2008) and providing more focus on structure and
organization (Elola & Oskoz, 2010; Mak & Coniam, 2008). Yet, in some studies, the issues of text
ownership and a reluctance to edit the contributions of peers were raised. The interview responses in
Lunds study (2008) revealed students concerns about inexpert editing and abuse, while in Kesslers
(2009) study, students were more willing to edit their peers work than their own. However, those peer
162
edits were found to be focused more on form rather than content as students felt they did not have the
right to change the content of the contributions of others. Unequal contribution by group members to the
collective product was another concern raised in Arnold, Ducate and Kosts 2009 study.
Several studies that have been conducted specifically investigated the types of revisions students made. In
most of those studies, topic choice and task type were found to affect the degree to which students engage
in collaborative writing as well as the degree of focus on form and the amount of writing production.
Arnold et al. (2009) examined the revision behaviors of intermediate German students in three different
classes using wikis collaboratively. One of the classes was structured and received instructions on how
to edit their contributions, and revisions focused more on form than meaning. In contrast, in the
unstructured class, revisions focused more on meaning than form. In all three classes, stylistic changes
came third after form- and meaning-related changes. Kesslers (2009) analysis of revision behaviors of 40
non-native English speaking pre-service teachers in a Mexican university as they collaboratively-defined
and revised the word culture using a wiki showed that they were willing to collaborate in this
autonomous environment, and that they were more willing to edit their peers work than their own. The
task was initiated by the teacher but was completely left to the students to develop: no feedback, revisions
or elaborations were provided by the teacher. Students focused more on meaning during the production,
worked on improving content and did not strive for perfect grammatical accuracy as long as errors did not
impede meaning. These findings are in line with Kessler et al.s (2012) study, which investigated the
collaborative writing behaviors during the production of research reports of their own choice by 30
highly-proficient non-native English speakers using Google Docs. Students focused more on meaning
than form and the grammatical edits they made were more accurate than inaccurate. However, contrary to
these findings, Kost (2011), who analyzed the number of formal changes versus meaning-preserving
(stylistic) changes made by students in a German language class, found that formal changes were much
more frequent than meaning-preserving (stylistic) changes (89% vs. 11%) and that students were very
successful in repairing grammatical errors. Similarly, Spanish language learners in Lees (2010) study
attended to language errors at the sentence or word level during meaning-driven activities as they worked
together. The open-ended tasks and topics that were broad enough and gave freedom to students to
incorporate their personal interests while at the same time requiring them to focus on form, motivated the
learners and resulted in a high degree of collaborative exchange in her study.
Although several studies have sought to address the effects of using tasks in writing instructions, few
studies have examined the role of tasks on self-corrections and peer-corrections. The current study
contributes to the literature by examining whether task type has an effect on the number of form-related
and meaning-related changes, number of self- and peer-corrections, and by investigating the accuracy of
self- and peer-corrections learners make during wiki-based collaborative writing tasks in an EFL context.
It further seeks to understand learners perceptions towards the use of wikis.
METHODOLOGY
Participants
Data for this study was collected from 16 female and 18 male non-native speakers of English from
various educational backgrounds studying in a preparatory program at a private university in Istanbul.
Participants had an average age of 19.2 years, and were studying in two different classes, each class
consisted of 17 students. Two instructors taught each class. While one of the instructors was one of the
researchers in both classes, the second instructor varied. All participants in this study shared the same
native language, Turkish. They had already completed levels A1, A2 and B1 of the Common European
Framework (CEF) before starting the B2-level module, and had 24 hours of English instruction each
week. As such, they were considered independent users of the target language. In an interview prior to the
study, all participants considered themselves to be competent users of Web 1.0 technology, including
browsing the Internet and using email and text chat. However, none of the students had used a wiki before
163
the study.
Tasks
Learners participated in three different meaning-focused tasks (Table 1) that were selected to engage them
in the collaboration and negotiation of both meaning and form as they produced texts in a wiki-based
environment. Meaning-focused tasks can be defined as tasks in which students have an aim to convey a
message to an audience thereby encouraging them to focus on the content of the text they produce.
However, a form-focused task, such as drills or gap filling exercises, could be described as a task which
encourages the learners to focus on the formal elements of the language. In line with the suggestions
given in the literature, learners were immersed in open-ended and authentic tasks which were based on
real life situations, including a communicative aim that intended to engage them in meaningful interaction
and collective production through shared decision making, while at the same time allowing them to pay
attention to form (Lee, 2010; Skehan, 1998; Swain, 2000).
Table 1. Distribution of the number of MRCs in All Three Tasks
Type of MRC
Argumentative
Task
89
Informative
Task
36
Decision-making
Task
46
New Information
33
51
14
98
Picture
22
48
79
Deleted Information
22
12
41
Synthesis of Information
17
Reorganizing
Video
Link
Clarification / Elaboration of
Information
Total
171
While all tasks required the use of higher-order thinking skills in line with Skehans (1998) suggestions,
task topics were selected from among familiar and meaningful topics for students to balance their
cognitive load. The learners in the current study had just completed the B1 level of the CEF and were
competent in writing simple connected text on topics which are familiar or of personal interest and
writing personal letters describing experiences and impressions (Teachers Guide to the Common
European Framework, n.d., p.8). Considering Hess (2011) Cognitive Rigor Matrix for reading and
writing, the argumentative and decision-making tasks were selected to be cognitively and linguistically
more demanding than the informative task, since the first two required the learners to apply skills such as
devising an approach among many alternatives, developing a logical argument, and articulating a new
voice. The informative task, on the other hand, mainly required such skills as recalling or locating basic
facts, details, definitions and events, and describing the features of a place. All tasks were designed to be
convergent in terms of goal orientation and required the learners to try to reach a common goal out of
multiple outcome options. See Appendix A for a description of the tasks used in this study. Participants
were continuously encouraged to use their own words, and were reminded beforehand that their text
production would be monitored by the instructor against any form of plagiarism, especially because Tasks
1 and 2 seemed conducive to copying and pasting from various resources. The non-error-free nature of
the texts co-constructed by participants in the wiki pages led researchers to assume that plagiarism had
not been a concern for the results of the study.
164

Procedure
The study took place during the second semester, which started in February 2010 and continued for seven
weeks. At the beginning of this nine-week semester, the instructor/researcher set up a class wiki for each
class and held a training session before learners started to work on their projects. Learners were provided
with detailed criteria regarding the grading of their wikis, which included an assessment of both
individual and collaborative working skills. The grade learners received from the project constituted 5%
of their final course grade. Intially, learners worked on a non-graded task in which they collaboratively
wrote definitions for specific concepts determined together in a class discussion whose aim was to assist
the students in their familiarity with using the wiki. After the non-graded task, they were asked to
complete three different tasks in a row. The first and second tasks took two weeks to complete, while the
final task took one due to time restrictions of the nine-week academic semester.
Prior to each task, learners were randomly assigned to groups of four prior to each task and therefore they
worked in a different group for each task. This was done purposefully since in Arnold et al.s (2009)
study, some students complained about unequal participation and poor communication within their groups.
The researchers wanted the learners in this study to have a different group dynamic for each task and have
a chance to be able to interact with different peers in their class. After the completion of the tasks, the
content created by the learners in all the wiki pages was analyzed. A questionnaire was given to the
students and a focus group interview was conducted in the seventh week of the study.
In order to examine the role of task type in the number of meaning-related and form-related changes, the
history pages of all tasks were analyzed and the number of meaning-related and form-related changes was
calculated separately for each task by the researchers. For the argumentative task, there were 31.75
history pages on average. The average number of history pages for the second task was 17.5. The final
task generated 13 history pages on average. In the present study, to identify form-related changes, all
sentences including grammatical corrections were analyzed by using Kesslers (2009) categorization as a
starting point. However, only the incidents observed in the data became part of the categorization, and an
analysis was based on the categories that emerged from the data as shown in Appendix B.
To identify meaning-related changes, all sentences including at least one meaning-related change (MRC)
were examined. Kessler and Bikowski (2010) define MRC as any meaning-related change a student
makes such as changing a letter, word, sentence, paragraph or the entire wiki (p.45). Kessler and
Bikowskis (2010) coding category was adapted to examine meaning-related changes in the data.
However, the change of a letter, for example, the change of a misspelled word such as improvment to
improvement, was coded as a form-related change unless it led to a change in the meaning of a sentence.
The last three categories in Appendix C were added by the researchers as they emerged in the data and
includes a description of each category. In order to examine the role of task type in the number of selfand peer-corrections, all history pages of all tasks were analyzed and the number of self-corrections and
peer-corrections was calculated separately for each task by one of the researchers and one native English
speaking teacher (NES) who was also teaching with one of the researchers at the same institute. Both selfand peer-corrections were defined as any changes in the form of a grammatical structure and did not
include any meaning-related changes. Peer-corrections were defined as the corrections made to one
participants contribution by another member of the group that he/she worked with, while self-corrections
referred to the corrections made to ones own contributions to the wiki pages. The number of self- and
peer-corrections was compared to explore which task yielded more self-correction or peer-correction.
Corrections by the students were judged as correct or incorrect by the researchers and the
aforementioned NES teacher. The number of correct and incorrect changes was calculated along with
their percentages. All correct and incorrect edits were noted and counted separately for each task.
165
The Questionnaire.
In the seventh week of the study, students were asked to fill out a questionnaire that included questions
about their overall learning, motivation, group interaction and use of technology (see Appendix D). The
aim of the questionnaire was to describe the learners overall experience with the use of wikis and to
evaluate the effectiveness of the project. There were a total of 39 statements in the questionnaire: 35
Likert-type statements and four open-ended questions. All of the statements were adapted from Lee
(2010) and Hazari, North, and Moreland (2009). The response rate was 67.64% for the questionnaire and
its reliability was calculated as =.98. Descriptive statistics were used to analyze the results of the
questionnaire. Open-ended questions were carefully read and recurring themes were grouped by the
researchers.
The interview.
A face-to-face semi-structured focus-group interview was conducted with six randomly chosen
participants in the seventh week of the study regarding their experiences in using wikis for collaborative
writing tasks (see Appendix E). The interview was conducted at the institution of one of the researchers
and was tape-recorded and transcribed. Common threads in the responses were identified and the results
were used to triangulate the data obtained through the content analysis of the wiki entries and
questionnaire results. The interview enabled the researchers to gain insight into the participants overall
experience regarding this wiki project.
RESULTS
In this section, the results of the analysis of the wiki pages will be presented to answer the first three
research questions, and the questionnaire and interview responses will be discussed in relation to the
fourth research question in the study.
The Effects of Task Type on Form-related vs. Meaning-related Changes
The content analysis of the wiki pages revealed that there were more meaning-related changes than formrelated changes across all three tasks (Figure 1). In the argumentative task, 57% of all changes were
meaning-related whereas 43% were form-related. In the informative task, the meaning-related changes
constituted 77% of all changes while 23% were form-related; and in the decision-making task, 54% of the
changes were meaning-related and 46% were form-related. A Chi-square test indicated that those
relationships were statistically significant (X2=26.371, p<0.05).
Figure 1. Distribution of changes in all three tasks.

The distribution of the number of meaning-related and form-related changes in all three tasks was
analyzed using descriptive statistics as summarized in Tables 1 and 2.
166
Table 2. Distribution of the number of FRCs in All Three Tasks

Type of FRC
Argumentative Task
Informative Task
Decision-making Task
Total
Word Choice
29
22
56
Spelling
23
13
45
Coordination
13
23
Singular/Plural
12
19
Articles
16
Tense
16
Capitalization
12
13
Verb Form
12
Part of Speech
Subject/Verb Agreement
Prepositions
Unnecessary Word
Word Order
Punctuation
Verb-Verb Agreement
Relative Clauses
Active/Passive
Modals
Superlatives
Double negation
The most common meaning-related change was the clarification/elaboration of information followed by
adding new information and adding a picture to the text. Word choice was found to be the most common
form-related change followed by spelling. The results revealed that students attended to meaning more
than form across all three tasks.
The Role of Task Type on Self- vs. Peer-Corrections
The content analysis of the wiki pages revealed more peer-corrections (a total of 203) than selfcorrections (a total of 54) in total as illustrated in Figure 2; the argumentative task resulting in the largest
number of corrections, followed by the decision-making and informative tasks. It was only in the
informative task that self-corrections outnumbered peer-corrections. In the argumentative task, 89% of all
the corrections were made by peers while 11% were self-corrections. In the informative task, 32% of all
the corrections were peer-corrections whereas 68% were self-corrections. Finally, in the decision-making
task, peer-corrections constituted 88% of the total number of corrections with the remaining 12% being
self-corrections. The results of the Chi-square analysis in SPSS revealed that those differences are
statistically significant (X2=71.197, p<0.05).
167
200
150
Form-related
100
Meaning-related
50
Task 1
(Argumentative)
Task 2 (Informative)
Task 3 (Decisionmaking)
Form-related
137
44
76
Meaning-related
178
147
90
Figure 2. Distribution of corrections in all three tasks.

The Extent of Accuracy of Self- and Peer-corrections
A total of 203 peer-corrections in all three tasks were observed, and while 94.5% of them were correct,
5.5% were incorrect. Table 3 illustrates the distribution of these corrections across all three tasks.
Table 3. Number of Correct and Incorrect Corrections in Peer-correction in Each Task
Peer
Corrections
Correct
122
116
(95%)
(5%)
14
10
(71%)
(29%)
Task 3 (Decision-making)
67
66
(98.5%)
(1.5%)
203
192
(94.5%)
11
(5.5%)
Task 1 (Argumentative)
Total
Incorrect
A total of 54 self-corrections in all three tasks were observed, 92.5% of which were correct and 7.5%
were incorrect. Table 4 illustrates the distribution of these corrections across all three tasks.
Table 4. Number of Correct and Incorrect Corrections in Self-correction in Each Task
SelfCorrections
Correct
15
13
(86%)
(14%)
30
29
(96%)
(4%)
(89%)
(11%)
54
50
(92.5 %)
Total
Incorrect
(7.5%)
When the total number of corrections was examined, out of 242 corrections made by participants in the
wiki pages, 94% were accurate. As Table 5 illustrates, the number of accurate corrections was higher than
the number of inaccurate corrections in all three tasks. The highest level of accuracy was found in the
decision-making task, as 97% of the corrections resulted in grammatical accuracy.
168
Table 5. Level of Accuracy in the Corrections

Correct
Percentage
129
94%
6%
39
89%
11%
74
97%
3%
242
94%
15
6%
Total
Incorrect
Percentage
Students Perceptions Regarding Using Wikis in Collaborative Projects

Appendix D illustrates the averages of all the items in the questionnaire. As can be seen, the statements
with the highest mean average were Use of wiki-based collaborative writing tasks helped to improve my
foreign language writing skills (M=3.9), I liked the topics used in the tasks (M=3.9), and I started to
view other English language learning methods more positively after using the wiki. (M=3.9) while the
statements with the lowest mean average were Use of the wiki enhanced my interest in the course
(M=3) and Doing the assignments through the wiki encouraged me to study more regularly (M=3).
All six students who participated in the interview stated that they had a positive experience using wikis
for collaborative writing tasks. This finding was also supported by the questionnaire results; 52.2% of the
respondents stated that they agreed or strongly agreed with the statement Overall, I had a positive
experience with the use of wiki-based collaborative writing tasks (M=3.5). However, the questionnaire
results showed that nearly 48% of the participants did not particularly enjoy these tasks. During the focus
group interviews, those who did not enjoy the tasks explained that their discontent was mainly due to the
compulsory nature of the project. They argued that they were not motivated because this project was a
course requirement. Nevertheless, all interviewees concurred that the wiki provided a condusive
environment for group work and was therefore useful. Moreover, 60.9% of the respondents agreed or
strongly agreed with the statement I liked working together with my friends while creating wiki pages
(M=3.7). Furthermore, students felt that their writing skills improved as stated by one interviewee:
When you read your peers edits, you see different sentences used to express different
ideas. This contributes to your existing knowledge of vocabulary.
The feeling of improvement in the students foreign language writing skills is also reflected in the
questionnaire as Use of wiki-based collaborative writing tasks helped to improve my foreign language
writing skills was one of the most popular statements with 69.8% of the respondents who agreed or
strongly agreed with it. The value of the collaboration was evident in the questionnaire as 39.1% of the
respondents agreed or strongly agreed with the statement I learned more because of my friends
contributions to the wiki (M=3.3). Similarly, the statement I learned new things while reading and
editing my peers contributions had a mean average of 3.5 with 43.5% of the respondents agreed or
strongly agreed with it. Finally, 73.9% of the respondents who agreed or strongly agreed with the
statement Doing assignments on the wiki helped me to learn from my own mistakes (M=3.8). This
result was pointed out in the interview as:
If I had a weak or a simple sentence in the text the teacher would probably correct the
grammatical mistakes in it. However, while we are working on the wiki my peers rewrite
the whole sentence, change some words or add some words into it so that I can see how to
make my sentences look more complex and ideas sound stronger.
DISCUSSION
Both quantitative and qualitative data provided multiple sources of information regarding the use of wikis
in collaborative writing projects and the role of task type during this process. In this section, the results of
169
the study will be discussed in light of this data and in relation to theory and previous research.
The Effects of Task Type on Form-related vs. Meaning-related Changes
In this study, learners were engaged in three different meaning-focused tasks that encouraged them to
attend to both content and form while collaboratively producing texts. An analysis of the learners
contributions in the wiki pages showed instances of both form-related and meaning-related changes in
their writing. However, learners paid more attention to meaning than form, regardless of the task type in
these wiki-based collaborative writing tasks. This result may be attributed to the design of the tasks; that
is, in line with the suggestions in the literature (Ellis, 2003; Pica, Kanagy & Falodun, 1993), all tasks
were designed requiring interactants to request and supply information in order to achieve the same or
convergent goal such as defending an idea, introducing a city, or offering a solution to a problem. For the
completion of the tasks, negotiation of meaning was crucial while learners tried to convey a message to
the readers. This finding supports the findings of Kessler (2009), Arnold et al. (2009), and Kessler,
Bikowski and Boggs (2012) which revealed that students paid more attention to meaning than form in
unstructured wiki-based and web-based collaborative writing tasks that encourage students to focus on
content with an aim of conveying a message rather than focusing on the grammatical structures of the
target language. In line with Kesslers (2009) findings, the meaning focus of the tasks may have led
students to overlook the grammatical errors as long as they did not interfere with the comprehension of
the intended message. Yet, the presence of a high number of form-related changes, especially in the
argumentative and decision-making tasks, showed us that attention to form was also considered important
during these meaning-based activities. This finding is consistent with Lees (2010) study in which
students collectively attended to grammatical inaccuracies.
The most frequent forms of MRCs were found to be the clarification/elaboration of old information and
new information. In Kesslers and Bikowskis (2010) study, the most frequent forms of MRCs were new
information and deleted information. Similar to Kessler and Bikowskis (2010) findings, participants in
this study did not engage in MRCs which required higher order critical thinking skills such as
synthesizing and reorganizing in their wiki changes. As Kessler and Bikowski (2010) argue without the
extensive use of synthesis, it is difficult to succeed at collaborative writing in a wiki setting (p. 52), and
what was found is this study,, participants preferred to expand on a currently introduced topic or add
completely new information rather than synthesizing pieces of information that existed in the wiki. This
finding is also in line with Lees (2010) study in which participants showed reluctance to editing their
peers postings and showed more willingness to adding text rather than to editing existing writing.
Adding a picture to the text was another frequent MRC and this finding is also consistent with Lees
(2010) study during which it was observed that students embedded multimedia sources extensively to
support the content of their postings. The distribution of FRCs in the current study is parallel to the
distribution of FRCs in Kesslers (2009) study as word choice and spelling were the most frequent types
of FRCs he found in his study.
As mentioned earlier, the argumentative and decision-making tasks were selected to be cognitively and
linguistically more demanding for the participants than the informative task. Skehan (1998) argues that
when a task demands too much attention to its content due to its complexity, learners will pay more
attention to its meaning and less attention to its language. This did not necessarily apply here. The
participants in this study allocated their attentional resources to form during the more challenging tasks
and paid more attention to meaning during a task which they felt more familiar with. However, as can be
seen in Table 4, the majority of the MRCs in the informative task was in the form of adding new
information, a picture and the clarification/elaboration of information. In addition, only one instance of
synthesizing was observed whereas the argumentative and decision-making tasks promoted more use of
this skill. It should also be noted that the informative task included the least number of peer corrections
among the three tasks. 68% of all changes made in the informative task were self-corrections made by the
participants: therefore most of these MRCs were actually done by participants on their own postings.
170
The Role of Task Type on Self- vs. Peer-Corrections

When all changes made by the learners were analyzed, more peer-corrections than self-corrections were
observed in the wiki pages. Task type was also found to have a significant effect on the number of
corrections. While the argumentative task promoted the largest number of peer-corrections, both the
argumentative and decision-making tasks promoted more peer-corrections than the informative task. This
meaningful difference may be attributed to the fact that when students deal with cognitively and
linguistically related rigorous tasks, and when they try to convey their own ideas, they tend to be more
engaged in collaborative dialogue which Swain (2000) describes as joint problem solving and
knowledge building (p.102). During the informative task, learners were more familiar with the content as
they had done some research on the cities they had chosen prior to preparing a visitors guide. They had
organized and presented information from memory and written sources in their own words: thus, the
informative task lacked the mutual problem-solving orientation found in the other two tasks. Furthermore,
during the interview, some learners stated that they preferred to divide the work for the informative task
because the nature of the task suited a division of responsibilities since there were different parts in the
visitors guide. In contrast, during the argumentative and decision-making tasks, all members in the
groups tried to either defend their own ideas against an opposite idea to convince the reader or expressed
their own opinions in how to solve a problem. As a result, the participants of the study did not feel that
they needed to correct one anothers contributions in the informative task, which resulted with less
engagement in collaborative dialogue. One of the learners mentioned this during the interview:
in the informative task, I felt I did not have the right to change what my peers had
written since the information presented was an obvious fact known by everyone.
This study supports previous research (Lee, 2010) which shows that open-ended tasks offering
opportunities for learners to mutually engage in problem solving results in learners paying more attention
to each others contributions, and thus increases collaboration.
The Extent of Accuracy of Self- and Peer-corrections
The findings of the present study suggest that although there is no statistically significant relationship
between correction type and grammatical accuracy, wiki-based collaborative writing tasks lead to the
accurate use of grammatical structures most of the time (94%), as has been observed in previous studies
(Arnold et al., 2009; Kessler et al., 2012; Kost, 2011). This finding is also consistent with Elola and
Oskoz (2010) and Lee (2010) who conclude that wiki-based collaborative writing activities foster
grammatical accuracy. An important feature of using wikis for collaborative writing tasks, as shown in
this study, is that it allows opportunities for learners to notice their linguistic gaps by drawing attention to
form for the improvement of linguistic accuracy. Learners had the opportunity to revise and improve their
writing through feedback from their peers, an opportunity that they may not have had in the absence of
wiki technology.
Students Perceptions as regards Using Wikis in Collaborative Projects
The results of the focus-group interview and the questionnaire suggest that the learners had a positive
experience and enjoyed working collaboratively on the wiki-pages; most felt that their writing skills
improved as a result of the project. This finding is in line with the findings of recent studies (Arnold,
Ducate, & Kost, 2009; Lee, 2010). Furthermore, the interview responses show that the use of wikis
promoted collaborative work by allowing the participants to contribute to the projects without any time or
space restrictions. This was also argued by other researchers (i.e., Engstrom & Jewett, 2005; Keith, 2006;
Lamb & Johnson, 2007). Moreover, wikis offer an alternative way to extend collaboration outside the
classroom and provide learners with opportunities to do extra writing practice, which is also supported by
earlier research (Farabaugh, 2007; Lamb, 2004).
However, the participants feelings of inhibition to edit each others postings, similar to the ones observed
171
in previous studies by Lund (2008), Kessler (2009) and Lee (2010), were also revealed during the focusgroup interviews. Apparently, the self-confidence of students in their language skills got in the way of
their correcting grammatical mistakes in the wiki pages, as is evident in the following quote during the
focus group interview:
Since we are not completely proficient in English, it was difficult for us to notice all the
grammatical mistakes in our peers work. Even if we thought something needed to be
expressed in a different way, we had to search from other sources and be sure that our
edits were accurate.
The interview responses also showed that participants felt more comfortable to comment on and edit each
others contributions more during open-ended tasks which required them to exchange opinions, rather
than fact, and engaged them in mutual problem-solving activities.
Limitations and Suggestions for Further Research
Although the present study has revealed important insight for collaborative writing in foreign language
teaching and learning, some limitations need to be acknowledged. First of all, the study was conducted in
seven weeks, which is a short duration for such a study. Although all the participants were computer
literate, it was the first time they had used wikis as a component of their foreign language learning
courses. Therefore, the novelty effect of the tool may have affected student participation in the wiki-based
tasks. Hence, the study could be replicated within a longer time span by allowing the participants to get
fully accustomed to the technical features of the tool and also to acquire full appreciation of how to
collaborate using wikis. Another limitation was that new groups were formed for each task in this study.
This was done purposefully to prevent the familiarity effect on the students performance and to enable a
fair evaluation of their performance as some unresponsive group members may have affected the
performance of others in the tasks. Needless to say, working in a new group for each task may have
affected the performance of some students. Therefore, the study could be replicated without changing the
groups throughout the tasks in order to eliminate the possible negative effect of group dynamics on
student performance in the project. Task types were another limitation to the study. An argumentative, an
informative and a decision-making task were designed to examine the role of the task type. The study
could be replicated using other types of tasks to explore the role of different task types in wiki-based
collaborative writing projects.
Unfortunately, during this study, researchers did not have access to the interactions learners had outside
of the wiki environment. All evidence of collaboration and cooperation in this study was based on the
history pages and groups final products on the class wiki. Having access to their written or oral
discussions and interactions regarding the tasks could have provided valuable insight concerning their
collaboration process.
CONCLUSION
It is evident that collaborative writing environments foster an opportunity for L2 learners to co-construct
knowledge and become engaged in collaborative dialogue with their peers through scaffolding on each
others language use as well as to improve their linguistic capacity. It is especially important to provide
EFL students with collaborative learning environments outside the classroom, as there are fewer
opportunities to practice the target language in social contexts in an EFL setting. Learners in this study
showed a willingness to learn from each other through scaffolding and feedback. Therefore, this study
showed that the affordances of wikis offer an alternative way to extend collaborative interaction and
scaffolding beyond the classroom through carefully chosen collaborative tasks, to allow learners
opportunities to notice linguistic gaps and organizational problems in their writing, and to have great
potential in providing an enjoyable foreign language learning experience.
172
However, the characteristics of the tasks chosen and types of activities learners engage in clearly bear
great importance in constructing the amount of collaborative dialogue and scaffolding among learners:
therefore, they should be chosen carefully. Our research indicates that open-ended and meaning-focused
tasks encouraged learners to attend to meaning more than form, and tasks that were designed to be
cognitively more demanding resulted in more peer-corrections which finds support in relevant literature
(Ellis, 2003; Pica, Kanagy, & Falodun, 1993).
Although, participants seemed to attend to meaning more during the informative task, which was selected
as a less rigorous task, a significant number of the meaning-related changes were in the form of simply
adding more information to existing information, which is not necessarily an indicator of collaboration
among learners. One important reason for this was found to be that the informative task was based on
more factual information and students could have a division of labor. On the contrary, tasks that do not
have clear-cut divisions of responsibilities and that require interactants to request and supply opinions and
information in order to mutually solve a problem or reach a goal, promote the greatest opportunities for
giving and receiving feedback, and produce a text collectively and collaboratively.
Evidently, task based instruction is very important in fostering interaction and collaboration among
learners and the type of task affects the way learners interact with each other. As was mentioned in
previous research, in a wiki environment, it is the task that promotes collaborative interaction among the
learners and not the technology itself (Lee, 2010; Lund, 2008). Future research that explores more
structured wiki environments for peer feedback and the role of language teachers during collaborative
writing tasks will certainly contribute to the interpretation of the results of this study.
APPENDIX A. Tasks Designed for the Study

Task 0:
Writing
Definitions
Choose 5 concepts and write definitions for them. Explain what they mean to you.
Task 1:
Argumentative
Task
Choose one of the prompts below and write an argumentative essay with your
partners.
a)
Restrictions should be placed on the use of mobile phones in public areas

like restaurants and theaters.
b)
Censorship is necessary.
c)
Traditional male role has changed in Turkey over the last 20 years.
d)
Advertising means manipulation.
e)
The mass media, including TV, radio, newspapers have a great influence
on people and especially on the younger generation. It plays an important
role in shaping the opinions and positions of the younger generation.
f)
Global climate change is man-made.
g)
Parents should let teenagers make their own decisions.
h)
Age does not matter in relationships.
Task 2:
Informative
Task
(Visitors
Guide)
Choose one of the cities in Turkey and prepare a visitors guide for people who
want to visit that city. The guide should include general information on the city,
accommodation, food, places to see, and things to do, etc. in that city. You may add
photos or videos in it. You may get ideas from the websites below:
http://www.visitorsguide.is/
http://www.seattlepi.com/visitorsguide/
Task 3:
Decisonmaking Task
(Dear Abby)
You are working for a website called 'Dear Abby' on which people write about
their problems and ask for advice. Read the posting and try to help the person by
offering advice on how to solve his/her problem.
173
APPENDIX B. Coding Categories and Descriptions of Form-Related Changes

Coding category
Description of Category
Data Sample
Word Choice
Student changes a word another

student has used.
People should learn how to make true correct

decisions
Coordination
Student changes/adds a coordinating

conjunction to a sentence.
Finally, I want to mention that your mother...
Spelling
Student changes the spelling of a

word.
The most crucial alteration is that a lot of men exhance

enhance and change their perspectives on life.
Part of Speech
Student changes the form of a word.
These days, life conditional living conditions are

becoming more and more difficult.
Singular/Plural
Student changes the singularity or

plurality of a noun.
Woman Women should work.
Articles
Student adds/changes the article of a

noun.
Living conditions are becoming more and more

difficult due to the economic crisis.
Prepositions
Student adds/changes a preposition to

a sentence.
It is assumed that men are superior from to women...
Subject/Verb
Agreement
Student changes the subject or the

verb of the sentence to maintain
subject-verb agreement.
On the other hand the father were was more

comfortable...
Unnecessary
Word
Student deletes an unnecessary word

from a sentence.
...because also women generally do the housework...
Tense
Student changes the tense of a

sentence.
...because also women generally do did the housework

and take took care of children...
Punctuation
Student adds/changes a punctuation

mark to a sentence.
Thus, unlike woman, men start to accept

responsibility....
Word Order
Student changes the word order of a

sentence.
Is the second largest port after Istanbul Izmir Port.

izmir port is the second largest port after Istanbul Port.
Capitalization
Student changes the caption of a

letter.
...I belive that traditional male role has changed in

turkey Turkey over the last 20 years
Verb Form
Student chages the form of a verb.
...now women start share to share different

responsibilities...
Active/Passive
Student changes an active sentence

into passive or a passive sentence into
active voice.
...before saying things which are hurt your mother...
Superlatives
Student adds/corrects a superlative

structure.
Becoming a teenager is very the most dangerous

period of life.
Relative Clauses
Student adds/corrects a relative

clause to a sentence.
Individual's occupational self-selection which means a

particular work environment...
Negation
Student corrects a double negation in

a sentence.
...nobody is not the same...
Modals
Student adds/changes a modal to a

sentence.
...and you should offer to study together again.
Note. Cross out words show the data deleted from the text. Bold words show the data inserted i
174
APPENDIX C. Coding Categories and Descriptions of Meaning-Related Changes

Coding
Category
Description of
Category
Data Sample
Explanations
New
information
Student writes
about a subtopic not
previously
discussed
stanbul is the most populated and vivacious city in Turkey.

It is located at the northwest of Turkey and costs of the sea
of Marmara and Blacksea. stanbul is a perfect choice who
want to go a place which is in the heart of history, art,
natural beauties and technology. stanbul consists of two
sides which are connected with each other by two bridges.
HISTORICAL PLACES OF ISTANBUL
You can see almost everywhere in stanbul traces of old
civilizations such as Ottoman Empire and Byzantine Empire.
Every avenue, building even paving stones carry a historical
and magical atmosphere especially districts like Eminn,
skdar, Beyolu, etc.
The title
Historical
Places of
Istanbul was
added as a new
piece of
information to
the existing body
of the text.
Deleted
information
Student deletes
information,
ranging from
one word or
piece of
punctuation to
the entire body
of the wiki
There is no point brooding over it, you sould talk to your

friend face to face and explain your feelings and concerns
about your friendship and her attitude towards you, so you
can feel comfortable.
The crossed out

part was deleted
from the text.
Clarification
/ elaboration
of information
Student adds to
a sub-topic that
had already
been introduced
You can see almost everywhere in stanbul traces of old

civilizations such as Ottoman Empire and Byzantine Empire.
Every avenue, building even paving stones carry a historical
and magical atmosphere especially districts like Eminn,
skdar, Beyolu etc. Moreover, if you want to learn more
abot stanbul and its past, you should visit Topkap palace,
Blue Mosque, Hagia Sophia etc. Topkap Palace is a very
rich museum where the important objects belonging to
Ottoman Empire mostly are kept. Blue Mosque and Hagia
Sophia also are of interest to a wide range of visitors.
The underlined
part was added
to the paragraph
to clarify /
elaborate on the
existing
information.
Synthesis of
information
Student writes a
sentence or
paragraph that
ties together
previously
written
information
In retrospect, the traditional male role has undergone

massive changes in terms of social, domestic and business
life. The traditional notion of male and female roles has been
redefined in our country. In the light of the aforementioned
ideas, we can say that men's perspective and the point of
view of the society to men's role have improved with the aid
of getting educated and disposing of gender bias. There is no
question that the man has gradually given up his patriarchal
authority. Therefore, the traditional position of women in the
society has considerably changed over the last 20 years and
as a result of this, one of men has too.
In this paragraph
student wrote a
conclusion to the
argumentative
essay that ties
together what
had already been
written.
Link
Student adds a
link.
If you want to have information this issue, they can should

look link
http://en.wikipedia.org/wiki/Interpersonal_relationship
Reorganization
Student changes
the place of a
sentence or a
whole
Official statistics indicate that women now represent almost

fifty percent of the workforce. These days, living conditions
are becoming more and more difficult due to the economic
crisis. When this effect is taken into consideration, women
The crossed out

sentence shows
the previous
location of the
175
paragraph.
Picture
Student adds a
picture.
Video
Student adds a
video.
should work. This condition affects the authority of man

because earning money is his sole power on woman in some
period of male life so, when the woman learn how to stand
on her own legs, traditional male role disappear easily. These
days, living conditions are becoming more and more difficult
due to the economic crisis. When this effect is taken into
consideration, women should work.
part which was

reorganized by
the student.
Student deleted
the sentence
from its original
location and
pasted it to a
different location
in the text
without making
any other
editions.
Note. Adapted from Developing collaborative autonomous learning abilities in computer mediated language learning: Attention
to meaning among students in wiki space, by G. Kessler & D. Bikowski, 2010, Computer Assisted Language Learning, 23(1),
41-58.
APPENDIX D. Questionnaire & Mean Results

This questionnaire was designed to examine students opinions on the use of wikis in English language teaching.
Please circle the best option which states how much you agree with the following statements. This questionnaire is
not going to affect your grade in the course.
Mean
Questions
3.8
1.
The wiki interface and features were overall easy to understand.
3.6
2.
I liked seeing other students interaction with material I posted in the wiki.
3.3
3.
I would prefer classes that use wikis over other classes that do not use wikis.
3.7
4.
Editing information in the wiki was easy.
3.9
5.
Use of wiki-based collaborative writing tasks helped to improve my foreign language writing
skills.
3.3
6.
I stayed on task more because of using the wiki.
3.1
7.
I would like to see wikis used in other courses when I go to my faculty.
3.4
8.
Benefit of using the wiki is worth the extra effort and time required to learn it.
3.7
9.
I participated in the assignment more because of using the wiki.
3.5
10.
Benefits of using the wiki outweighed any technical challenges of its use.
3.4
11.
Use of the wiki for the assignment helped me interact more with students.
3.5
12.
Technical features in the wiki helped me improve my writing skills in English.
3.3
13.
Because of using the wiki, my group was able to come to a consensus faster.
3.5
14.
I will retain more material as a result of using the wiki.
3.8
15.
I would recommend classes that use wikis to other students.
3.7
16.
Compared to other discussion boards and forums, the wiki was easier to use.
3.7
17.
Use of the wiki promoted collaborative learning.
3.3
18.
I learned more because of my friends contributions to the wiki.
3.0
19.
Use of the wiki enhanced my interest in the course.
176
3.5
20.
Wiki projects should be used more often in education.
3.8
21.
Wiki tasks were completely related to the course objectives.
3.9
22.
I liked the topics used in the wiki tasks.
3.7
23.
I liked working together with my friends while creating wiki pages.
3.4
24.
I often used the History page to see the previous changes before I edit something on the wiki.
3.1
25.
I found the Discussion page useful to communicate with my friends and share my comments.
3.5
26.
I learned new things while reading and editing my peers contributions.
3.7
27.
I felt comfortable while editing my peers work.
3.6
28.
I would rather write on the wiki to traditional essay writing.
3.4
29.
Contributing to the wiki tasks helped me write better essays in the classroom.
3.5
30.
Overall, I had a positive experience with the use of wiki-based collaborative writing tasks.
3.0
31.
Doing the assignments through the wiki encouraged me to study more regularly.
3.3
32.
Doing assignments on the wiki enabled me to evaluate my own performance.
3.8
33.
Doing assignments on the wiki helped me to learn from my own mistakes.
3.8
34.
Working on the wiki projects improved my research skills.
3.9
35.
I started to view other English language learning methods more positively after using the wiki.
36.
What did you like most about the wiki assignments?
37.
What did you find the most challenging about the wiki assignments?
38.
39.
What would you recommend to improve the use of wikis in the future?
If there is anything else you would like to mention, please write it below.
Note. The items were scored on a scale of 1, 2, 3, 4 or 5 points (5 = Completely agree,

Disagree, 1 = Completely disagree).
4 = Agree, 3 = Not sure, 2 =
APPENDIX E. Focus Group Interview Questions

1.
Could you please explain your overall experience with the wiki tasks?
2.
What do you think about the topics in the tasks? Which task did you like the most and the least? Why?
3.
What was the most important and interesting aspect of working on a wiki? Why?
4.
What was the most challenging aspect of working on a wiki? Why?
5.
What is the contribution of the wiki to the group work?
6.
Was it easy for you to change/edit your peers writings? Did you feel comfortable while editing your peers
work?
7.
How did the use of wikis affect your overall opinion on foreign language writing?
8.
What would you suggest to make the use of wikis more effective?
ABOUT THE AUTHORS

Zeliha Aydn works as an instructor of English at zyein University and holds an MA degree in Foreign
Language Education. Her research interests include collaborative learning, task-based learning, and use of
technology in language teaching.
E-mail: zelihaydin@gmail.com
177
Senem Yldz is an Assistant Professor at Boazii University, Foreign Language Education Department.
Her research interests focus around the use of new technologies in language teaching materials design and
development, and classroom practice.
E-mail: senem.yildiz@boun.edu.tr
REFERENCES
Arnold, N., Ducate, L., Kost, C. (2009). Collaborative writing in wikis: Insights from culture projects in
intermediate German classes. In L. Lomicka & G. Lord (Eds.), The next generation: Social networking
and online collaboration in language learning, (pp.115144). San Marcos, TX: CALICO.
Benson, P. (2003). Learner autonomy in the classroom. In D. Nunan (Ed.), Practical English language
teaching (pp. 289308). New York, NY: McGraw Hill.
Blake, R. (2000) Computer mediated communication: A window on L2 Spanish interlanguage. Language
Learning & Technology, 4(1), 120136. Retrieved from http://llt.msu.edu/vol4num1/blake/default.html
Cohen, E. G. (1994). Restructuring the classroom: Conditions for productive small groups. Review of
Educational Research, 64, 135.
Donato, R. (1994). Collective scaffolding in second language learning. In J.P Lantolf & G. Appel (Eds.)
Vygotskian approaches to second language research. (pp. 3356). Norwood, NJ: Ablex.
Donato, R. (1988). Beyond group: A psycholinguistic rationale for collective activity in second-language
learning. Unpublished doctoral dissertation. Newark, DE: University of Delaware.
Donato, R. (2000). Sociocultural contributions to understanding the foreign and second language
Classroom. In J. Lantolf (Ed.), Sociocultural theory and second language learning (pp. 2750). Oxford,
UK: Oxford University Press.
Ducate, L., Anderson, L., & Moreno, N. (2011). Wading through the world of wikis: An analysis of three
wiki projects. Foreign Language Annals, 44(3), 495524.
Ellis, R. (2003). Task-based language learning and teaching. New York, NY: Oxford University Press.
Elola, I. & Oskoz, A. (2010). Collaborative writing: Fostering foreign language and writing conventions
Engstrom, M. E. & Jewett, D. (2005). Collaborative learning the wiki way. TechTrends: Linking
Research & Practice to Improve Learning, 49(6), 1216.
Farabaugh, R. (2007). The isle is full of noises: Using wiki software to establish a discourse community
in a Shakespeare classroom. Language Awareness, 16(1), 4156.
Foster, P. (1998). A classroom perspective on the negotiation of meaning. Applied Linguistics, 19(1),
123.
Hauck, M., & Youngs, B. (2008). Telecollaboration in multimodal environments: The impact of task
design and learner interaction. Computer Assisted Language Learning, 21(2), 87124.
Hazari, S., North, A., & Moreland, D. (2009). Investigating pedagogical value of wiki technology.
Journal of Education Systems Education, 20(2), 187198.
Hernandez, N., Hoeksema, A., Kelm, H., Jefferies, J., Lawrence, K., Lee, S., Miller, P. (2008).
Collaborative writing in the classroom: A method to practice quality work. Retrieved from
178
http://www.edb.utexas.edu/cscl/2008/topicpapers/s2paper.pdf
Hess, K. K (2011). Local assessment toolkit: Exploring cognitive rigor. The National Center for the
Improvement of Educational Assessment. Retrieved from
http://www.nciea.org/publication_PDFs/CRM_ELA_KH11.pdf
Hirvela, A. (1999). Collaborative writing: Instruction and communities of readers and writers. TESOL
Journal, 8(2), 712.
Hyland, K. (2003). Second language writing. New York, NY: Cambridge University Press.
Keith, M. (2006). Wikis and student writing. Teacher Librarian, 34(2), 7072.
Learning & Technology, 13(1), 7995.
Kessler, G. & Bikowski, D. (2010). Developing collaborative autonomous learning abilities in computer
mediated language learning: Attention to meaning among students in wiki space. Computer Assisted
Language Learning, 23(1), 4158. Retrieved from http://llt.msu.edu/vol13num1/kessler.pdf
Kessler, G., Bikowski, D., Boggs J. (2012). Collaborative writing among second language learners in
academic web-based projects. Language Learning & Technology, 16(1), 91109.
http://llt.msu.edu/issues/february2012/kesslerbikowskiboggs.pdf
Kim, Y. (2008). The contribution of collaborative and individual tasks to the acquisition of L2
vocabulary. The Modern Language Journal, 92, 114130.
Kost, C. R. (2011). Investigating writing strategies and revision behavior in collaborative wiki projects.
CALICO Journal 28(3), 606620.
Lamb, A. & Johnson, L. (2007). An information skills workout: Wikis and collaborative writing. Teacher
Librarian, 34(5), 5759.
Lamb, B. (2004). Wide open spaces: Wikis, ready or not. EDUCAUSE Review, 39(5), 3648.
Lee, L. (2002). Enhancing learners communication skills through synchronous electronic interaction and
task-based instruction. Foreign Language Annals, 35(1), 1623.
Lee, L. (2010). Exploring wiki-mediated collaborative writing: A case-study in an elementary Spanish
course. CALICO Journal, 27(2), 260276.
Lund, A. (2008). Wikis: A collective approach to language production. ReCall, 20(1), 3554.
Mak, B. & Coniam, D. (2008). Using wikis to enhance and develop writing skills among secondary
school students in Hong Kong. System, 36, 437455.
Nunan, D. (1992) Collaborative language learning and teaching. Cambridge, UK: Cambridge University
Press.
Pellettieri, J. (2000). Negotiation in cyberspace: The role of chatting in the development of grammatical
competence. In M. Warschauer & R. Kern (Eds.), Network-based language teaching: Concepts and
practice (pp. 5986). Cambridge, UK: Cambridge University Press.
Peterson, M. (2008). Non-native speaker interaction management strategies in a network-based virtual
environment. Journal of Interactive Learning Research, 19(1), 91117.
Pica, T., Kanagy, R., & Falodun, J. (1993) Choosing and using communicative tasks for second language
instruction. In G. Crookes & S. Gass (Eds), Tasks and language learning: Integrating theory and
practice. Vol 1 (pp. 934). Clevedon, UK: Multilingual Matters.
179
Skehan, P. (1998). A cognitive approach to language learning. Oxford, UK: Oxford University Press.
Skehan, P., Foster, P. (2001). Cognition and tasks. In P. Robinson (Ed.) Cognition and second language
instruction (pp. 183205). Cambridge, UK: Cambridge University Press.
Storch, N. (2002). Patterns of interaction in ESL pair work. Language Learning, 52, 119158.
Storch, N. (2005). Collaborative writing: Product, process, and students reflections. Journal of Second
Language Writing, 14, 153173.
Storch, N. (2009). The nature of pair interaction. Learners interaction in an ESL class: Its nature and
impact on grammatical development. Saarbrcken, Germany: VDM Verlag.
Storch, N., & Wigglesworth, G. (2007). Writing tasks: Comparing individual and collaborative writing. In
M. del Pilar Garca_Mayo (Ed.) Investigating tasks in formal language learning, (pp. 157177). London,
UK: Multilingual Matters.
Swain, M. (2000). The output hypothesis and beyond: Mediating acquisition through collaborative
dialogue. In J. P. Lantolf (Ed.), Sociocultural theory and second language learning (pp. 97114). Oxford,
UK: Oxford University Press.
immersion students working together. Modern Language Journal, 82(3), 320337.
Teachers Guide to the Common European Framework (n.d). Retrieved from
http://www.pearsonlongman.com/ae/cef/cefguide.pdf
Van Lier, L. (1996). Interaction in the language curriculum: Awareness, autonomy, and authenticity.
London, UK: Longman.
Vygotsky, L. S. (1978). Mind in society. Cambridge, MA: Harvard University Press.
Wigglesworth, G., & Storch, N. (2009). Pair versus individual writing: Effects on fluency, complexity
and accuracy. Language Testing, 26, 445466.
180

http://llt.msu.edu/issues/2013reviewerlist.pdf
2013 LLT Reviewers
2013 LLT REVIEWERS

The editorial staff would like to recognize the reviewers of submitted LLT manuscripts during 2013. This
journal could not function without these scholars academic service, and we would like to thank them
warmly for their contributions.
Farhan AbuSeileek
Jozef Colpaert
Luke Harding
Suhall Ahmed
Frederik Cornillie-Fee
Debra Hardison
Yuka Akiyama
Elena Cotos
Richard Harrison
Khalid Al-Seghayer
Troy Cox
Mirjam Hauck
Vahideh Alipour
Scott Crossley
Volker Hegelheimer
Enza Antenos
Catia Cucchiarini
Francesca Helm
Nike Arnold
Euline Cutrim Schmid
Rebecca Hinks
Saeed Ayiewbey
James Neil Coburn
Bert Hodges
Ruth Ban
Mark Darhower
Christopher Holden
Daro Luis Banegas
Tracey Derwing
Joseph E. Hopkins
Clara Bauler
Piet Desmet
Phil Hubbard
Nancy Bell
Melinda Dooly
Ari Huhta
Dawn Bikowski
Laura Ducate
Jessican Ivy
Robert Blake
Jonathan deHann
Jess Izquierdo
Alex Boulton
Miriam Eisenstein Ebsworth
Juha Jalkanen
Klaus Bradl
Joy Egbert
Gerriet Janssen
Jack Lester Burston
Charles Elster
Min Jung Jee
Mnica Cardenas-Claros
Gulcan Ercetin
Song Jiang
Walcir Cardoso
Michael Filsecker
Nobuhiro Kamiya
Angela Chambers
Jons Fouz Gonzlez
Jung Jin Kang
Kuo-En Chang
Carolin Fuchs
Orlando Kelm
Yulin Chen
Senta Goertler
Claire Kennedy
Hsueh-Chih Chen
Adrienne Gonzales
Richard Kern
Hsin-I Chen
Marta Gonzlez-Lloret
Greg Kessler
Chin-Wen Chien
Karen Gregory
Daesang Kim
Chengzhi Chu
Stefan Gries
Claudia Kost
Dorothy Chun
Nicolas Gromik
Mustafa Kurt
Joseph Collentine
Laura Gurzynski-Weiss
Leena Kuure
Karina Collentine
Paul Andrew Gruba
Chun Lai
Penelope Collins
Regina Hampel
Eva Lam
181
2013 LLT Reviewers
Batia Laufer
Ana Oskoz
Julie Sykes
Joshua Fahey Lawrence
Luisa Panichi
Skipp Symes
Yoonhee N. Lee
Gi-Pyo Park
Maija Tammelin
Ronald Leow
Scott Payne
Joo Antonio Telles
John Levis
Lynn Pearson
Celia Thompson
Tim Lewis
Diane Pecorari
Kimberly L. Tohill
Mimi Li
Jill Pellettieri
Constanza Tolosa
Meei-Ling Liaw
Manuel Peralbo-Uzquiano
Heidi Vaarala
Tsun-Ju Lin
Mark Peterson
Alireza Vafaeepour
Huifen Lin
Maribel Poza
Larry Vandergrift
Chin-Hsi Lin
Nora Presson
Robert Vanderplank
Hsien-Chin Liou
Peter Prince
Nina Vyatkina
Min Liu
Eva Prionas
Gillian Lord
Ravi Purushotma
Rmi Adam van

Compernolle
Andreas Lund
Paul Rama
Peter Macintyre
Mariana Aguilar Ramrez
David Malinowski
Jim Ranalli
Rob Martinsen
Leila Ranta
Richard McCallum
Hayo Reinders
Levi McNeil
Jonathon Reinhardt
Kara McBride
Fernando Rosell-Aguilar
Anna Mikhaylova
Susanne Rott
Nicole Mills
Gulnara Sadykova
Hazel Morten
Shannon Sauro
Karthyn Murphy-Judy
Jean Schultz
Adreas Mller-Hartmann
Klaus Schwienhorst
Suthagar Narasuman
Yue Sheng
Tatsuya Nakata
Patrick Snellings
Hilary Nesi
Bill Snyder
David Neville
Susana Sotillo
Katherine Nielson
Glenn Stockwell
Grantham OBrien
Robert Summers
Robert ODowd
Yu-Chih Sun
David Oakley
Sean Sutherland
Elvis Wagner
Min-Feng Wang
Yuping Wang
Shudong Wang
Liang Wang
Paige Ware
Chantelle Warner
Cynthia White
Shona Whyte
Ciara Wigham
Lee Wilberschied
Lawrence Williams
Magdalena Wolska
Stefanie Wulff
Richard Xiao
Yang Xiao
HuiLing Xu
Il Yaln
Jackie Xiu Yan
Chunsheng Yang
182
2013 LLT Reviewers
Ai Chun Yen
Holly Ulmer York
Makoto Yoshii
Bonnie Lynne Youngs
Victoria Zenotz Doctor
Qi Zhang
Jinjing Zhao
Dongping Zheng
Bin Zou
Katerina Zourou
183

Language Learning &amp; Technology 18

Enviado por

Dados do documento

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Language Learning &amp; Technology 18

Enviado por

Direitos autorais:

Formatos disponíveis

Volume 18 Number 1

Fostering Foreign Language Learning

Using Peer Computer-Mediated Corrective

Towards Transparent Computing:

Contact: Editors or Managing Editor

Ecological Affordance and Anxiety in an Oral

2013 Reviewer Acknowledgements

Language Learning with Technology: Ideas for

Contact: Editors or Managing Editor

About Language Learning & Technology

Copyright 2014 Language Learning & Technology, ISSN 1094-3501.

Sponsors, Board, and Editorial Staff

ADVISORY AND EDITORIAL BOARDS

Michigan State University

University of Hawaii at Mnoa

Portland State University

Copyright 2014 Language Learning & Technology, ISSN 1094-3501.

Editorial Board (continued)

Southern Methodist University

University of California, Santa Barbara

University of California, Irvine

Simon Fraser University

University of California, Berkeley

The Open University

University of South Carolina

State University of New York, Albany

University of Hawaii at Mnoa

Web Production Editor

Michigan State University

Book & Multimedia Review Editor

Southern Methodist University

Action Research Column Editor

Emerging Technologies Editor

Virginia Commonwealth University

Social Media Director

Michigan State University

Universidad de los Andes

Universidad de los Andes

University of California, Santa Barbara

University of Hawaii at Mnoa

Reginald Gentry, Jr.

Kobe Womens University

University of Hawaii at Mnoa

University of Hawaii at Mnoa

Sookmyung Womens University

University of Hawaii at Mnoa

Universidad de los Andes

Copyright 2014 Language Learning & Technology, ISSN 1094-3501.

Language Learning & Technology

February 2014, Volume 18, Number 1

Copyright 2014, ISSN 1094-3501

Towards Transparent Computing

Language Learning & Technology

Towards Transparent Computing

Towards Transparent Computing

Language Learning & Technology

Towards Transparent Computing

Language Learning & Technology

Towards Transparent Computing

Language Learning & Technology

Towards Transparent Computing

Language Learning & Technology

Towards Transparent Computing

HTML 5 Official specs from W3C

Language Learning & Technology 18

Language Learning & Technology 18