1-Benchmarking Telecommunication and Access in Information Society

SIBIS
IST2000-26276
Statistical Indicators Benchmarking the Information Society

Benchmarking Telecommunication and Access
in the Information Society

by

March 2003
Benchmarking Telecommunications and Access in the Information Society
2
This publication has been prepared by Technopolis Ltd., Brighton (UK) in the context of the
IST-26276-SIBIS project (SIBIS Statistical Indicators Benchmarking the Information
Society) in co-operation with the other partners in the project.
All publications of the SIBIS project including this report are available in electronic format
on the Internet at:
www.sibis-eu.org

Neither the European Commission nor any person acting on behalf of the Commission is
responsible for the use which might be made of the following information. The views
expressed are those of the authors and do not necessarily reflect those of the European
Commission. Nothing in this report implies or expresses a warranty of any kind. Results from
this report should only be used as guidelines as part of an overall strategy.

For further information, contact:
empirica GmbH
Oxfordstr. 2
D-53111 Bonn
E-mail: sibis@empirica.com

A great deal of additional information on the European Union is available on the Internet. It
can be accessed though the Europa server: http://europa.eu.int

SIBIS project and European Communities, 2003
Reproduction is authorised provided the source is acknowledged.

Printed in Germany.

3
TABLE OF CONTENTS
FOREWORD...............................................................................................................................4
PREFACE....................................................................................................................................5
1 Executive summary..............................................................................................................7
1.1 Introduction to the topic ................................................................................................. 7
1.2 Results of the data ........................................................................................................ 8
1.3 Description of future work ............................................................................................ 13
1.4 Conclusions ................................................................................................................ 13
2 Introduction ........................................................................................................................16
2.1 Topic area definition .................................................................................................... 16
2.2 The approach.............................................................................................................. 18
2.3 Overview of the report ................................................................................................. 19
3 Identification of the indicator framework and hierarchy.....................................................21
3.1 Traditional indicators for telecommunications and access .............................................. 21
3.2 New SIBIS indicators selection..................................................................................... 22
4 Analysis of Data.................................................................................................................26
4.1 Analysis of indicators about citizens and society............................................................ 26
5 The use of compound indices............................................................................................43
5.1 The data..................................................................................................................... 44
5.2 Snapshot indicators and country groupings factor and cluster analysis......................... 46
6 Further developments: remaining gaps in the statistical coverage of the topic................52
6.1 The existing gap on regional level data indicators for T&A.............................................. 52
6.2 Some methodological problems when collecting information on T&A with telephone survey
methodologies ............................................................................................................ 58
7 Conclusions........................................................................................................................60
8 ANNEX 1 References .....................................................................................................62
9 ANNEX 2 - Methodology of the survey..............................................................................63
9.1 General Population Survey (GPS) ................................................................................ 63
9.2 Decision Maker Survey (DMS) ..................................................................................... 70
9.3 Questionnaires ............................................................................................................ 78
4
FOREWORD

For some years now statistical indicators on Information Society have been central in the
policy making process. This has been best demonstrated through the benchmarking exercise
of eEurope 2002 Action Plan, and its further inclusion as a key activity in eEurope 2005.
Having recognised this need and driven by the difficulties in obtaining reliable and
appropriate statistics, the IST programme supported a pan-European research effort during
Framework Programme 5. The prime objective has been to develop and make available
methodologies, tools and new statistical indicators which can help remedy the deficit in this
field.

It is in this context that the SIBIS project was launched (IST-26276, Statistical Indicators
Benchmarking the Information Society, www.sibis-eu.org). This document presents some of
the projects main findings so far.

There are at least two main reasons that make this document interesting. First, it is one of
the few original attempts to have a coherent and comprehensive approach in measuring the
Information Society. As such it is expected to stimulate further debate and research among
the professional statistical community, leading to an improved statistical competence in
Europe. Second, it provides a unique single source of data on real time which supports many
of the new IST research areas, at the launch of Framework Programme 6.

Building on the original SIBIS research, in particular on the results of the indicator surveys,
the project has produced 9 reports, selected from those addressed by e-Europe.

The SIBIS work attracts further interest since it also supports the e-Europe 2005 initiative.
SIBIS is carrying out an evaluation and a benchmarking of the eEurope 2005 initiative for the
15 EC Member States and the 10 Accession countries which will become available later in
2003. Both the reports and benchmarking results can be obtained from the SIBIS web site.

The publication of the SIBIS project results is a timely and direct contribution to benchmark
progress on key issues of the information society in general and the e-Europe initiative in
particular.

Thanassis Chrissafis

athanassios.chrissafis@cec.eu.int

DG INFSO-C6
5
PREFACE

This report is a main deliverable of the SIBIS project (Statistical Indicators Benchmarking the
Information Society), funded by the European Commission under the Information Society
Technology Programme (1998-2002). The overall goal of SIBIS is to develop and pilot
indicators for monitoring progress towards the Information Society, taking account of the e-
Europe action lines. SIBIS focuses on nine topics of interest, i.e. Telecommunications and
Access, Internet for R&D, Security and Trust, Education, Work and Skills, Social Inclusion, e-
Commerce, e-Government and e- Health.

Within the SIBIS project two surveys (a General Population Survey and a Decision Makers
Survey businesses), were conducted between March and May 2002 covering the nine
eEurope topics. This report describes the outcomes with respect to the topic of
Telecommunications and Access. The indicators presented are largely taken from a the
General Population Survey which covered consumers and, therefore, this topic should be
looked at in the context of informing interested parties on the access, usage and impact of
technologies on the general population, not on businesses.

The report is organised in eight chapters. The first three chapters give the reader an
overview of the main outcomes (Executive Summary), the context (Introduction) and the
indicators developed (Identification of the Indicator Framework and Hierarchy). The core of
the report is the analysis of indicators, provided in Chapter 4. The indicators are presented
as Access indicators, Usage indicators and Impact indicators of ICT technologies including
Internet, Broadband and Mobile Telephony. The indicators are analysed in a number of
ways, for example, by age groups, EU15 vs. US, and straightforward country comparisons,
giving a variety of interesting results. Finally, this chapter also includes innovative compound
indicators aimed at measuring impact of ICT across different clusters of countries and
emerging paths of users migrating to faster connections than dial-up.

The final chapters summarise the outcomes of the study as well as areas where future
research might be necessary.

The main audience should be policy makers, statistical offices at all levels (national, e.g.
CBS, Statistisches Bundesamt, Statistics Finland etc., and supranational, e.g. Eurostat,
OECD), industry leaders and researchers in the domain and those involved and interested in
benchmarking the domain throughout Europe and the world. These institutes should consider
the questions posed and the subsequent indicators developed by SIBIS a valuable input for
their yearly surveys. The project includes a series of workshops with such institutions in the
countries represented by the SIBIS consortium. The report is also of interest to the European
Commission (in particular DG INFSO) and to government official and regulator bodies and
ICT Observatories.

For each of the nine topics a separate SIBIS report (WP2) was issued in 2001. The WP2
report was aimed at setting the scene on the topic, defining the gaps in the statistical
coverage and suggesting innovative indicators to be developed through the subsequent
survey. The current report, although an independent document, is an interim report. The
final version for all SIBIS reports will be issued in April 2003.

SIBIS is led by Empirica (Bonn, Germany), and includes the following project partners:
RAND Europe (Leiden, The Netherlands), Technopolis Ltd. (Brighton, UK), Databank
Consulting (Milan, Italy), Danish Technological Institute (Taastrup, Denmark), Work
6
Research Centre Ltd. (Dublin, Ireland), Fachhochschule Solothurn Nordwestschweitz (Olten,
Switzerland).

Technopolis is an International research and consulting organisation focusing on science,
technology and innovation policy. Technopolis works for regional and national government
bodies, international organisations as well as private companies. We are based in the UK,
Sweden, Netherlands, France and Austria. Our work spans most scientific disciplines,
technologies and industrial sectors. Our strong emphasis is on implementation and providing
advice about 'how to do it better'. Depending on the needs of individual projects, we work
alone or together with other international policy and technology research organisations.

This report has been peer-reviewed in accordance with our quality assurance standards and
may, therefore, be represented as a Technopolis product. For more information about
Technopolis or this document, please visit our web site (www.technopolis.co.uk), or contact:

Rebecca Alden (Senior Consultant)
rebecca.alden@technopolis-group.com

Maria del Mar Negreiro (Consultant)
maria.del.mar.negreiro@technopolis-group.com
7
1 Executive summary
1.1 Introduction to the topic

The topic of telecommunications and access is a physical cornerstone of the information
society and is both wide ranging and horizontal in nature. In many ways the topic can be
considered as the fundamental enabler it allows the other eEurope domains to happen.

Telecommunications and access does not stop with infrastructure. There are already great
changes and innovations taking place in accessing Internet technologies, mobile telephony,
multimedia and other emerging technologies, all of which contribute to the current state-of-
play in Europe. Although in many cases the Internet is becoming pervasive, access and
usage patterns do vary according to socio-economic and geographic factors. Mobile
telephony is also an exceptionally fast-growing sector.

For the purposes of indicator development in SIBIS the term 'telecommunications' has been
interpreted very broadly to include all the networks (cable, mobile, Internet, as well as copper
wire) over which all types of information (voice, data, sound, image) is carried. So, although
the main focus is on telephony networks, computer networks, the Internet, cable (TV as well
as telephony), and wireless forms of transmission are also included. Overall, a more
accurate descriptor in these circumstances would be communications networks.

Access is another loose defined topic. It can be defined formally as the ability to retrieve
data, graphics, sound, text etc whether on-line. Translated into the context of eEurope, this
topic covers the wide range of devices by which users access electronic information e.g.
computers, telephones, multimedia kiosks, televisions and other hybrid devices.

In defining the statistical boundaries of the SIBIS study for the topic telecommunications and
access, the work undertaken has taken into consideration the fact that fixed
telecommunications networks have been in existence for over 100 years and there is a long
history of indicators associated with measuring progress in this area. The main areas of
measurement have traditionally been in measuring the size and growth of the market and
different technologies as well as being used as an aid to predicting revenues, profits,
universality and potentiality. The process of evolving measurement and indicators was
simplified by the fact that telecommunications in most countries was developed by a state
monopoly. With just one provider it was easier to measure output.

Newer networks have not been subject to such long-term scrutiny. Most of the underpinning
technologies for these networks have only been developed in the last 25 years and also
identifying the potential user base has become more difficult. No longer are telephone lines
supplied to just two groups - business and residential customers. Today an array of network
products and services are targeted at multiple niche user groups who will consume more
than one service. Measurement of this is further complicated by the increase in the speed of
adoption. Technologies can be adopted and disposed of in less than a decade. Witness the
usage of WAP which will be replaced by the advanced second and third generation mobile
telephony standards. Any tracking measurement over a medium to long term must make
allowance for the possibility of technology redundancy by seeking to measure user
experience rather than the technology.

This investigation and indicator development focuses on issues of access and usage, which
at their most basic can be described in a binary form. Firstly, users either have access to a
8
particular service or not. Secondly, there are also a potentially limited number of ways in
which services can be used. Although some issues of impact have been explored through
SIBIS, the study does not expand to examine the technical quality or perceived value of such
services. For example, it does not attempt to analyse the percentage of time a connection to
the Internet is working at 100% of its capacity or whether a user considers one ISP's content
to give better value for money than another. This is a potentially very interesting extension to
the work currently being tackled under this project and certainly an important area to
examine. However, due to constraints in piloting indicators, it was considered more pressing
to establish indicators that inform the baseline at this early stage.

More sophisticated and elaborate indicators such as compound indicators are also rarely
available due to the problems associated with combining methodologies. Similarly, although
basic indicators exist for newer technologies, (and these are becoming more common), in
many instances they are also often not comparable, nor yet ready to meet the challenge of
emerging topics of interest. Examples of the latter include the ability to robustly measure
VOIP (Voice over Internet Protocol). This is because of the rapid cycle of technology
adoption.

The same issues apply to access mechanisms. Although telephone (fixed, mobile) and
television ownership rates are well known, there is less information on the extent to which
newer forms of access mechanism are available or used. New channels include digital TV,
Internet-enabled phones, and interactive TV. Emerging channels will be dominated by the
next generation of mobile products.

Finally, this topic area does not cover content; neither the provision nor the use of it.
Although tremendously interesting, it is an entire domain in itself, and would increase the
necessary research to facilitate indicator development by about 100%. Nor does this topic
include ICT market size, productivity issues and matters related to quality of service.

1.2 Results of the data

A large number of innovative telecommunications and access indicators were identified
within the work of the SIBIS project and a proportion of those were piloted in the SIBIS
survey in order to test, to a certain extent, their usefulness and validity. A small number of
the piloted indicators are not innovative but provide a filtering role in order to ask questions
on more interesting and sophisticated indicators. The report also presents some of the more
basic indicators in order to provide a better context for the latter. Also as mentioned above,
the limited size of the survey meant that a number of innovative indicators identified as part
of the SIBIS work could not be piloted. Some of these are nevertheless discussed in the
section on gaps. The indicators are presented as Access indicators, Usage indicators and
Impact indicators. The indicators are analysed in a number of ways, for example, by age
groups, EU15 vs. US and straightforward country comparisons giving a variety of interesting
results.

1.2.1 Access indicators

Share of Internet users with broadband access
This indicator is a baseline indicator and a fundamental measure of the impact broadband
access has had on the population of Europe. The figures presented in the report show the
share of at-home Internet users in Europe who have a broadband connection at home,
9
including cable modem, DSL, and alternative broadband technologies such as wireless. The
results of the survey indicate that while there is significant variation across countries, the
overall level of penetration of broadband is low.

An average of about 7% of at-home users respondents have such a connection, While
behind the US, with just over 16%
1
, the average does include some Member States that
have very high levels of broadband. For example, Belgium which benefits from a dense
urbanised geography where access to telecoms networks such as cable is relatively
ubiquitous.

In other Member States, the midband technology of ISDN is popular. In particular, Germany
has an ISDN penetration of 17% among the online population. This may be due to an
aggressive marketing of ISDN services to consumer customers of Deutsche Telekom since
the late 1990s. In contrast, the UK has a rate of only approximately 2% as ISDN was
primarily marketed as a business product and had high associated installation costs. The
lesson from midband is that consumer can and do migrate to better access technologies.

It is also worth contrasting its penetration level with the measure of those without any Internet
access. Across the EU15 nations, 55.7% of the European population have no Internet
access at home at all. The Netherlands had the lowest number of people not connected to
the Internet at 27%, while Greece had the highest number of people not connected with
about 85%. Respondents' access methods were also analysed on the basis of their age.

While much is expected of broadband, it still has a long adoption cycle ahead. Thus
narrowband is still the predominant bandwidth for Internet connections across Europe.
However, given that dial-up uptake is stagnating in the US, while this does not necessarily
mean death of dial-up access, it highlights the high levels of broadband uptake that the US is
currently registering in comparison to the EU 15. Dial up subscribers are less prevalent in the
US than last year
2
. Similarly in the UK , where broadband at home penetration rates has
increased two fold since August last year, narrowband users are also stagnating.

Take-up of Internet consumer market narrowband (metered/unmetered) and
broadband
3

Aug 02 Nov 02 Feb03
UK homes connected to the Internet

42% 42% 45%
using narrowband
84% 80% 79%
using narrowband
(unmetered)
38% 35% 38%
using narrowband
(metered|)
29% 28% 25%
using narrowband
(unsure whether metered or unmetered)
18% 16% 6%
using broadband
7% 9% 13%

1
Current US penetration BB rates at home stands at over 20% according to ITU data, February 2003
2
Morgan Stanley , March 2003
3
Oftel, Monitoring Internet studies, March, 2003.
10
As governments encourage the roll-out of broadband because of its associated benefits this
indicator will be an important surrogate measure for the impact of policy.

Migrators to a faster connection than dial-up
Understanding the migration path from dial-up connection to broadband access is important
if measures to encourage its adoption are to be better formulated. The US experience
indicates that long experienced narrowband users migrate to it because of the better quality
of access and the faster and persistent connections associated with broadband. Likewise,
the closer the narrowband and broadband gap is in relation to availability and low connection
prices, the more likely users are to migrate from narrowband. This measure is important for
informing policy that encourages broadband usage.

Access to the Internet At Home
This indicator supplies useful data about the context and nature of Internet usage. At-home
usage is seen as the natural locus for effective use of being online. The combination of
privacy and consistent access from the same PC means that users are more willing to make
use of applications such as e-commerce, interacting with organisations that hold personal
information about them, including government and health providers for example. While at-
home usage is generally established as a norm, the at-home location is not as endemic as
might be expected. In contrast to the 50% plus of respondents have such access in Northern
nations there are many Internet users in Europe who do not have this privilege.

Figure1: Internet usage by location
Internet usage by location
(in % of population)
0
10
20
30
40
50
60
70
80
90
100
D
KS
N
L
F
I
N
U
K
I
R
LDAL
E
U
-
1
5
B
I
EF
E
L
P
C
H
U
S
A
occasional usage (less than once a month)
usage only somewhere else
usage only at work
usage only at home
usage at home and at work

Base: respondents with Internet access at home who have used the Internet at least once in the last 4 weeks
(N=4,631). EU15 results weighted by EU15 population (N=3,838).
Source: SIBIS 2002, GPS

Internet Drop-Outs
This indicator attempts to measure the level of users dropping out of using the Internet at
home. As such, it can be used to detect possible failures in Internet provision. A high figure
could indicate structural problems within the Internet sector, such as too high access costs or
a lack of appropriate applications and content. The highest reported drop out rate was
Sweden at 13.7%. This is interesting, as Sweden is a mature Internet nation. Users have had
time to travel along the full user life cycle and have been exposed to a variety of alternative
11
access contexts. There could be a variety of reasons for this which would be interesting to
explore in detail. For example, it could also be because employees are now accessing the
Internet from the workplace and have therefore dropped their at-home connections.

1.2.2 Usage indicators

Figure 2: Online tenure
Experience of online usage
0
10
20
30
40
50
60
70
80
90
100
D
KS
F
I
N
N
L
U
KA
I
R
LLB
E
U
-
1
5
D
I
PEF
E
L
C
H
U
S
2 years and more 1 year - 2 years 6-12 months < 6 months Don't know non-Internet users

Question: When did you use the Internet for the first time?
Base: Respondents who accessed the Internet at least once in the last 12 months (N=6,905), EU15 results
weighted by EU 15 population (N=5,828)Source: SIBIS 2002, GPS Column percentage.

Internet users according to on-line tenure
This indicator measures the Internet tenure of Internet users, or the length of time since the
first Internet use. It is clear from the EU15 average of 61% of Internet users having two or
more year's tenure, that exposure to, and experience of the Internet encourages broadband
adoption since it correlates with the countries that have a wider penetration of broadband
connections. This relation has been further explored in the migrators from dial-up to a faster
connection compound indicator.

Degree of e-mail networking intensity
E-mail is often considered the 'killer app', a popular reason for people to go online. While
volume measures of e-mails sent generate very large numbers, understanding the nature of
a user's email network and their intensity of e-mail usage provides an indicator that has a
higher utility. Respondents were asked to indicate their degree of networking among friends
and relatives and the degree of usage of e-mail networks with friends and relatives. E-mail
penetration among friends and family was highest in Scandinavian countries where over half
of the respondents friends and families have an e -mail address. The US also had a higher
rate of email penetration than the EU average.

Alternative devices for accessing the Internet
Assessing the usage levels of different methods of accessing the Internet is an important
indicator as currently there is a renaissance in this area. While access via TV and dedicated
12
devices were experimented with in the mid-Nineties, the newer methods are more robust and
potentially more appealing to consumers. TV-set orientated devices such as digital TV (DTV)
set-top boxes and games consoles today have network links. GSM networks now allow
palmtop computers and WAP phones to access the Internet. Rather than replace PC-based
Internet access these methods and devices could result in bi-modal usage by consumers.
Though it is also important to explore whether different usages are intrinsic to the use of
different technologies. These measurements are particularly useful to look at by age group
as younger groups are the predominant bimodal users.
Despite the wide spread marketing of DTV in some markets it is not being heavily used to go
online. Only 9.9% of those respondents (who have used an alternative Internet access
method, indicated that they used DTV for it.

Degree of mobile networking intensity: Degree of mobile ownership networking
among friends and relatives
A natural corollary of the higher degree of mobile phone ownership in Europe is a high
networking intensity. Over 60% of respondents relatives and friends owned a mobile phone.
Again, this indicator is a basis for exploring the usage of mobile phone services. The
secondary indicator confirms that communication applications, such as SMS, are by far the
most important driver of phone usage, especially among those under 25.

Figure 3: Mobile data use according to age groups
Mobile usage according to age groups in EU-15
(in % of each age group)
0
10
20
30
40
50
60
70
80
up to 24 25 to 49 50 to 64 65 and more
mobile Internet user only SMS user neither SMS nor mobile Internet (incl. don't know) no mobile phone

Source: SIBIS 2002, GPS,
Base: all respondents weighted by EU15-population (N=10,306)

Over 80% of under 25s use SMS, and their usage (see fig.11b) includes not only
communication purposes, but also services such as sports, news, ringtones, downloads, and
purchasing the likes of tickets is also relatively stronger than among those in older groups.

1.2.3 Impact indicators

While it is very feasible to count the instances of occurrence associated with
Telecommunications and Access technologies it is important to understand the impact they
13
have on citizens too. However, impact indicators have been more difficult to operationalise.
Survey questions seeking responses about impacts are fielded only to current Internet users.
Therefore, they fail the plausibility test. The technology must have had an impact as it is
being used. However, some impact indicators have emerged that are useful. In particular,
consumer perceptions of broadband. The majority of broadband users perceive it as better
than expected

1.3 Description of future work

No survey is perfect and no bundle of indicators complete. There are topics which constitute
large subjects in themselves and are the focus of other existing studies. These include new
IPv6 and e-content.

There are indicators that because of the survey methodology are problematic to gather data
for. It is difficult to interview those under 16 by telephone for example. It is also difficult to
measure quality of service as users may not be exposed to multiple services that would allow
them to make comparative qualitative judgements, most have just one ISP or mobile network
provider

Likewise some very innovative indicators could not be tracked on a Pan-European level
anymore since there were not enough representative samples and they would not be
transferable at this stage to major European surveys such as Eurostat.

The usage indicators piloted give only a partial picture of usage. Co-ordinating them with
surveys concerning e-content (or with panel measurement data methodologies) would close
a gap and allow a fuller picture to emerge of what consumers are doing online. Also there are
problems constructing indicators on impact from the SIBIS dataset as only responses from
current users are collected.

A more difficult gap to fill is the lack of regional level data. While SIBIS indicators can be
classified by nation, cutting the data by region is more difficult. As much EU assistance is
delivered at regional level and many regional programmes contain an Information Society
priority it is increasingly important to measure access and usage at this level.

The current turmoil associated with third generation mobile telecoms networks should not
dilute efforts to fill a gap associated with them. The development of 3G will eventually lead to
new data delivery mechanisms across wide areas. They will be an important infrastructure
for the Information Society in years to come, especially as a means of overcoming the digital
divide. The usage of mobile services indicator currently gives only a partial insight into this
topic, but could easily be expanded when a representative sample of 3G users can be
assembled. Studying the mobile networks is desirable, but given the constraints on how
many questions can be fielded to respondents it was decided to concentrate on questions
relating to the telecommunications and access networks that are currently more widely
adopted.

1.4 Conclusions

Analysing how Europe is doing in the area of telecommunications and access is complex. It
is clear from the data that one size does not fit all in attempting to measure the access,
usage and impact of telecommunications and access across the Member States and the
14
other participating countries. From an analysis of the data from the General Population
Survey (GPS) and from the compound indicators a number of differences have been
identified across the participating European Countries and between the Europe Union and
the US.

Benchmarking data is often presented separately for what are in fact related indicators. To
add value to such approaches, the results of the SIBIS surveys have been analysed and
presented in a series of integrated "snapshots" that enable easy understanding of inter-
related indicators and patterns across Europe. Such snapshots have been prepared for
mobile telephony, Internet usage and migration to higher bandwidth connections by home
users. They show that trends are not always linear and that both individual countries and
clusters of countries can show differing development and migration paths depending on
national circumstances.

The Internet snapshot indicates three Internet clusters, with some overlap with and some
divergence from the mobile clusters. For the Internet snapshot Greece, Portugal, Italy and
Spain join France as laggards, and the Netherlands, Denmark and Switzerland join the
forerunners. As might be expected, there is a clear positive association between PC usage,
home access to the Internet and recent/regular Internet usage

The mobile snapshot shows the expected strong positive association between individual
mobile phone ownership and extent of ownership amongst one's family and friends. At the
time of the SIBIS survey, France and the US clearly lagged behind on both counts
4
. Although
there is some tendency for countries with greater mobile penetration to have more usage of
SMS by mobile phone owners there is enough divergence to suggest that other factors also
play a role. Some countries in the intermediate cluster (e.g. Belgium and Switzerland), have
SMS usage levels as high as those portrayed in cluster 1. Generally the percentage of SMS
users is over two times higher in the EU-15 than it is in the US. Moreover, in some countries
such as Finland mobile owners are four times more likely to be SMS users than their US
counterparts.

The migration from dial-up to a faster connection snapshot suggests that for most countries
there is a "traditional" migratory path. More regular usage of the Internet is associated with
an increased proportion of users with longer (2 or more years) usage, and a trend towards an
increased proportion who have migrated to a faster connection in the forerunner countries.
They seek a better online experience, e.g. quicker downloads or always on connections.
The clustering of the US, Sweden, Finland and Denmark illustrates this in the cluster of
bubbles portrayed above. The more experienced users are, the more likely they are to
upgrade to higher bandwidth and, therefore, to being realistically able to use all the benefits
and services within the Information Society.
As a general rule, there are clear signs that the propensity of Internet users to do various
things online increases with increasing levels of Internet penetration in a country increasing
tenure and increasing take up of broadband technologies.

In addition a number of methodological lessons have also been learnt:

It remains important to go on counting instances
It remains extremely important, in the light of these results, to go on counting instances in
some of the European countries involved in the survey. However, the next step of
measurement is where SIBIS adds value as there are numerous pan-European surveys
already in existence which are counting. SIBIS has the opportunity to suggest sets of

4
Although the mobile phone market is very dynamic and France has seen strong growth in mobile ownership
since the SIBIS survey levels are still low in comparison to other countries.
15
indicators which help provide a picture of the how and why?. In spite of this need, it must be
noted that identifying usage patterns and impact is much more difficult. The evolution of the
usage behaviours also means that the classifications have to be reassessed regularly to
make sure that they are not out of date. As user numbers increase, so does the depth of
content available to them. This is an exciting area of development, which can give great
insight into the use of technologies in Europe.

A number of flexible indicators could be chosen and used to test the levels of
penetration of technologies in each country before piloting sophisticated indicators
The differences across the Member States and Switzerland are high and vary for each
technology area. This means that the current approach of asking exactly the same set of
questions in each country leads to some country data, particularly in the cases of usage and
impact indicators, being statistically insignificant. A way of creating surveys, that are
comparable, but do not ask obsolete questions, is to devise a system of standard flexible
indicators which could be drawn on to create a set which best work in a country specific
situation. A set of penetration indicators could also be identified which would be asked
across the board in order to judge when some countries start to have significant samples of a
chosen technology in use so that more sophisticated indicators could be introduced.

Measure also experience not only technology
As mentioned above, the shortening of technology product life cycles makes any tracking
measurement problematic. This problem is compounded by the fact that user definitions and
perceptions of technology vary across countries. Therefore, over the medium and long term,
measuring experience, measuring consumers satisfaction levels, insulates indicators from
changing technology and its varying nomenclature. It is a complementary more qualitative
factor that has to be taken into account.
16
2 Introduction
2.1 Topic area definition

Problem description

The topic of telecommunications and access is a physical cornerstone of the information
society. It covers both the physical networks over which information is carried, as well as the
products and services used to access those networks. It is also horizontal as it cuts across
many of the other priority action lines of the eEurope initiative. In many ways the topic can be
considered as the fundamental enabler it allows the other eEurope domains to happen.

Focusing on the physical networks as a whole, achieving a reality of eventual universal
access and actual use of potential converging technologies and content is an extremely
complex topic. It relies on a number of technological, environmental and socio-economic
conditions in Europe to aid its development. It is currently being hampered by a number of
factors including the narrowband legacy, lack of appropriate content and competition
5
to
name, but a few. In moving towards universal access, broadband is probably the most
important enabling technological development of the current time, and it is, therefore,
imperative to measure who has access to it, and what it is being used for.

Also, although convergence is almost universally accepted as eventually certain to happen,
this meta-process is still in its infancy. Its nascent nature is still difficult to predict so the
impact on businesses and consumers on it is also hard to measure. There are also the
issues of the digital divide, which encompasses problems associated with, the availability of
telecommunications infrastructure, associated costs, linguistics, culture and other socio-
economic conditions.

The digital divide, universal access and the convergence of technologies are the large
hurdles that must be overcome before Europe will truly be a digital society.
Telecommunications and access does not stop with infrastructure, there are already great
changes and innovations taking place in accessing Internet technologies, mobile telephony,
multimedia and other emerging technologies, all of which contribute to the current state-of-
play in Europe. Although in many cases the Internet is becoming pervasive, access and
usage patterns do vary according to socio-economic and geographic factors. Mobile
telephony is also an exceptionally fast-growing sector.

Understanding their meta-processes, the hurdles and how technologies are adopted and
used requires measurement. While some indicators do exist to allow measurement there are
important gaps in this existing data. The SIBIS telecommunications and access indicators
should go someway towards helping understand the factors and actions that will influence
current progress towards an eEurope.

5
Even though it is two years since Member States were required by EU law to open their local exchanges to
enable competi ng operators to provide DSL latest figures from February 2003 the European Competitive
Telecommunications Association (ECTA) revels that still less than 5% of the nearly 200 million telephone lines
in the EU are equipped for DSL. Moreover, over 80%current DSL exchanges are incumbents ones
17
2.1.1 Framework for assessing the area

For the purposes of this study the term 'telecommunications' has been interpreted very
broadly to include all the networks (cable, mobile, Internet, as well as copper wire), over
which all types information (voice, data, sound, image) is carried. So, although the main
focus is on telephony networks, computer networks, the Internet, cable (TV as well as
telephony), wireless forms of transmission are also included. Overall, a more accurate
descriptor in these circumstances would be communication networks.

Access is another loose defined topic. It can be defined formally as the ability to retrieve
data, graphics, sound, text etc whether on-line or off-line. Translated into the context of
eEurope this topic covers the wide range of devices by which users access electronic
information e.g. computers, telephones, multimedia kiosks, televisions and other hybrid
devices.

In terms of defining the statistical boundaries within which this study is conducted, fixed
telecommunications networks have been in existence for over 100 years, so there has been
plenty of time for statisticians, users and the industry to have developed indicators. These
indicators typically measure the size and growth of the market and different technologies as
well as being used as an aid to predict revenues, profits, universality and potentiality. The
process of evolving measurement and indicators was simplified by the fact that
telecommunications in most countries was developed by a state monopoly. With just one
provider it was easier to measure output.

However, newer networks competitive fibre-optic networks, wireless, the Internet and
cable, have not been subjected to such long-term scrutiny. Most of the underpinning
technologies for these networks have only been developed in the last 25 years. The
networks are also provided by multiple organisations often cooperating and competing with
each other at the same time. Monitoring output becomes a much more difficult process.

Identifying the potential user base has also become more difficult. No longer are telephone
lines supplied to just two groups - business and residential customers. Today an array of
network products and services are targeted at multiple niche users groups who will consume
more than one service.

As the physical communication infrastructure underpinning these new networks is expensive
to develop and deploy it requires large amounts of capital to be invested. The demands of
the capital market have accelerated the need for 'up-to-date' data about market coverage
and penetration of these newer networks. While previously, the telecommunications industry
was used to gather, analyse and publish data on an annual basis, the financiers of the new
networks were seeking data on a quarterly basis in order to comply with financial reporting
requirements of capital markets, in particular the US. Thus while the data collection cycle
was compressed, the amount of data produced increased.

The growth imperative associated with the Internet boom also led to measures being abused.
An Internet service provider (ISP) can count the number of users it has in different ways: the
number of users; the number of accounts; the number of active accounts in the last three
months; and the number of accounts including those where the user has requested
cancellation, but which have not been processed. Indeed, some governments also used such
definitional gymnastics to define an 'online citizen' as someone living within three kilometres
of a public Internet access point. Indeed, although basic indicators for public switched
18
telephone networks (PSTNs) are widely available, commonly collected data from different
sources can still be conflicting.

More sophisticated and elaborate indicators such as compound indicators are rarely
available due to the problems associated with combining methodologies. Similarly, although
basic indicators exist for newer technologies, (and these are becoming more common) in
many instances they are also often not comparable, nor yet ready to meet the challenge of
emerging topics of interest. This is because of the rapid cycle of technology adoption. New
network and access technologies are introduced and are measured before the market has
had time to accept or reject it as valid technology. For example, VOIP is a technology that
was first mooted in the mid-1990s, but even today it is a still a poorly-adopted technology.
However, the lack of a large enough user base, that is statistically representative, does not
deter some from modelling and projecting the user behaviour and comparing it with other
access technologies. Such comparisons are not reliable bases.

The same issues apply to access mechanisms. Although telephone (fixed, mobile) and
television ownership rates are well known, there is less information on the extent to which
newer forms of access mechanism are available or used. New channels include digital TV,
Internet-enabled phones, and interactive TV. Emerging channels will include the new
generation of 3G products.

2.2 The approach

As there is already a lot of statistical data in the public domain, the study takes a dual
approach. The SIBIS project has researched, among others
6
, the topic of
telecommunications and access and identified a selection of current knowledge gaps that
cannot be plugged on a pan-European levels with public domain sources. A General
Population Survey, covering all 15 Member States, Switzerland and the US commissioned by
SIBIS in April 2002 was used to pilot test a number of the new indicators in the general
population and collect baseline data to fill these gaps
7
. It was not possible to pilot all the
identified indicators in the survey, but a large number have nevertheless been included. The
indicators that have been collected through the SIBIS survey include a range of availability,
access, usage and impact indicators. This data can be custom analysed and thus, elicit
primary data that is not easily available elsewhere. In addition, the study looks at combining
existing indicators and measures that are in the public domain in order to examine the
usefulness of new compound indicators
8
. Constructing compound indicators can, however,
be methodologically risky. For example, it is quite likely that the base indicators have not
been calculated on the same basis, or using the same sample. There is also an issue of
timeframe in such a fast moving areas where combining current data with data produced
even 6 months ago may not be useful. However, compound indicators can be very
rewarding, useful and interesting. While they may not irrefutably demonstrate a cause and
effect relationship, they can be indicative of something, or a propensity to do something
and can often provoke further questions; the answers to which can be particularly
illuminating.

6
EGovernment, eHealth, eEducation, eCommerce, Work Employment and Skills, Internet for R&D, Social
Inclusion, Trust and Security
7
This GPS included questions from all 9 topic areas.
8
This may also include new SIBIS indicators
19
2.2.1 Identification of the stakeholders and their interactions
In order to give a framework to this study it is important to identify who might potentially be
interested in these new indicators and also use them in some way in existing surveys. One of
the main aims of SIBIS is to be used as a test bed for innovative indicator development that
can be picked up by official bodies. The project is not only aiming to provide results of a pan-
European survey, but pilot these new indicators to test their validity so that they can be used
elsewhere. The project is providing a complementary resource, which can be exploited by
relevant stakeholders across Europe.

The main groups of stakeholders involved in indicator generation and use in the field of
telecommunications and access are:

Table 1: Stakeholders and Indicators

Indicator Generation Indicator Use
Statistical Agencies National and European,
ICT Observatories,-National and European
Industry Telecommunication operators,
equipment manufacturers
Policy bodies Telecommunication Regulators
Analysts and Consultants Policy bodies
Industry Analysts and Consultants
Regulators Users, consumers, watchdogs and consumer
groups
Data Publishers IDC, Private enterprises and public institutions
Measurement companies (such as RIPE or
Netwizards)
Private sector, public institutions, and individuals
(citizens).

There is a lot of formal interaction between certain groups especially between regulators
and industry (statutory obligations) and regulators and policy-makers (a key channel for
policy making). Given the economic strength of the telecoms sector and the perceived
European lead in 3G telecoms, there can also be strong relationships between analysts,
consultants, data publishers and the industry. In many instances, information transfer is quite
open and transparent. Indeed, some measurement and analyst firms are respected as
reliable independent data providers that have inform policy within both public regulators and
private corporations over a number of decades.

2.3 Overview of the report

This report presents the research on telecommunications and access indicators carried out
by the SIBIS project, and the results of testing many of these indicators in the SIBIS General
Population Survey (GPS) carried out in April 2002 in the EU 15 Member States, Switzerland
and USA.

Chapter 3 defines the scope of the analysed topic and the main issues concerning statistical
measurements in this field. A description of main stakeholders and their interactions provides
the background of the research.

Chapter 4 presents the conceptual framework used to develop SIBIS indicators and
discusses the motivation of their development and their contribution to the understanding of
telecommunications and access development.

Chapter 5 and 6 presents the survey results and also presents some combined compound
indicators to look at the validity of combining different questions to illicit further results and
20
analyses. The analysis is conducted to determine whether the various indicators implicit in
the survey questions are closely correlated and can be combined to form compound
indicators.

Chapter 7 discusses some of the gaps still remaining in the topic of telecommunications and
access and some of the interesting indicators already identified that were not piloted in the
General Population Survey

Chapter 8 outlines the main conclusions resulting from research

21
3 Identification of the indicator framework and hierarchy

A comprehensive review of existing statistical indicators, and the problems associated with
collecting data has already been dealt with extensively prior to the production of this report in
WP1 and WP2
9
. This chapter deals solely with the choice of piloted telecommunications and
access statistical indicators, which are useful and relevant for citizens, business and the rest
of society.

3.1 Traditional indicators for telecommunications and access

Part of the work leading up to the piloting of indicators in the SIBIS General Population
Survey (GPS)
10
included the collection and description of over 500 relevant indicators which
were considered to be useful and/or frequently cited to measure the Information Society,
particularly relating to telecommunications and access.

Access, Use and Impact indicators are the main focus of the analysis undertaken through
the SIBIS GPS survey. This report also includes a small selection of existing indicators,
which come from external and Internationally well-recognised sources which have been
chosen to complement the work presented.

As stated previously in the project, there is no inclusion or subsequent analysis of indicators
of market size or of digital content, which constitutes a topic that should be studied in its own
right. Also some policy issues do not lend themselves particularly well to statistical analysis
and are, therefore, not included under the scope of SIBIS. These include, for example, the
speed of unbundling the local loop or discussions on radio spectrum frequency or policies
regarding the sale of 3G/wireless Internet licences. This is because whilst it is always
possible to arrive at a number of something, the real value lies in the contextual,
explanatory information.

3.1.1 Why access indicators are currently more popular than usage/impact
indicators

The Information Society relies fundamentally on device user bases. Unless PCs, PDAs,
phone or kiosks are purchased, installed and used, no network effect can occur. Traditionally
the IT manufacturing industry has relied on market indicators associated with shipments. It is
simpler to count boxes at a finite number of manufacturers than to identify the many
organisations and homes where multiple users may actually use them. On the software and
service side of IT a similar mentality has prevailed. Licences and accounts have been easier
to count than what people do with a product and service.

In the telecommunications and access universe this mentality has been adopted to measure
ISP accounts, number of phone lines and mobile phone handset sales. They are relatively
simple proxies for access. Some user surveys can be conducted to identify a multiple-user
figure and from this a measure of access can be made.

9
For more information, refer to the website www.sibis -eu.org
10
The SIBIS Decision Makers Survey took place in parallel but was more extensively used for other topic areas
such as eCommerce
22
Identifying usage patterns and impacts is much more difficult. Usage must first be observed
over a time period. It then needs to be classified. Then the classifications must be applied to
the observed data. This is a time-consuming process. The evolution of usage behaviour
means that classifications can become increasingly out of date or too broad. As user
numbers increase so does the depth of content available to them. Re-classification may be
required to sub-divide definitions. Comparing classifications over time can be problematic as
these changes must be back calculated to compare like with like. Sometimes this is just not
possible if a statistically representative sample is required.

3.2 New SIBIS indicators selection

Because the area of telecommunications and access is so diverse and so many indicators
already exist, it is difficult to highlight just one or two areas to investigate. Because of this it is
also tempting to create numerous compound indicators, to combine two or more items of
existing information, rather than creating new indicators to deepen knowledge and
understanding of the subject. However, as telecommunications and access is such a fast
changing area, if only the compound approach is adopted there is a risk of omitting key data,
such as the adoption rate of new technologies or new access mechanisms. This type of
information requires old-fashioned methodology (such as counting the instances), but applied
to new items. Therefore, the study pursues both approaches: the creation new innovative
indicators and the testing of compound indicators.
In order to follow a systematic approach the indicators are organised in a hierarchical way,
grouping them on different levels and relating them to sub-topics and issues. Thus, in order
to address the highlighted difficulties, the report suggests using three general areas or
domains for defining new indicators:

Access to new technologies
Usage of new technologies
Impact of new technologies

An earlier Topic Analysis report (WP2.1), provided a framework for the development of the
indicators by breaking down the topic into a structure of several sub-topics and issues related
to these sub-topics.

To further refine the presentation of these indicators, the following hierarchical trees were
developed with several levels of branches, which presents a picture of indicators that relate
to one (or maybe several) of these branches. It became clear during the project that with
regard to the costs of data gathering, an indicator is more efficient the higher it is located in
the hierarchy. But, on the other hand, indicators further down the hierarchy tend to be more
precise and may allow more flexibility in composing indices.

A tree has the following levels:

Topic level. Access, Usage, and Impact of new technologies.
Sub-topic level. The technologies piloted in SIBIS are mainly Internet (all connection
methods), Broadband, and Mobile telephony.
Constructs. These are terms for theoretical concepts which can not be measured
directly (e.g. satisfaction), but have to be operationalised via indicators that are capable
of measuring key aspects of the construct.
Construct dimensions. Sometimes, constructs need to be broken down into different
dimensions to make operationalisation easier (e.g., the notion of satisfaction might
contain four dimensions: cost, quality, reliability and speed)
23
Indicators. These are the indicators we might use to measure the construct dimension
Sub-Indicators. These are more detailed indicators which address a particular sub-
domain or aspect

Hence indicators which are being developing under the scope of SIBIS are grouped into:

Broadband/High Speed Networks this is probably the single most important enabling
technological development of the current time and it is, therefore, imperative to measure who
has access to it and what it is being used for. The indicators here include cost, availability,
take-up and consumption of broadband technologies, usage, quality of service provision and
regulation.

Internet although in many cases the Internet is becoming pervasive, access and usage
patterns do vary according to socio-economic-geographic factors. Proposed indicators
include how the Internet is being accessed (location, bandwidth and multiple platforms
users), purposes for which the Internet is used, quality of service provision.

Mobile as an exceptionally fast-growing sector, mobile telephony is important. Indicators
are shown for cost and awareness of cost, access, use and benefits of use, satisfaction with
mobile telephony, access to, and use of 3G telephony.

Overleaf is a diagram including all T&A questions piloted in the SIBIS surveys.

Figure 4: Overall Framework of Telecommunications and Access indicators piloted in the SIBIS
Survey

Telecommunications and
Access
Usage
A1 1, A1 2
Impact Access
Broadband Internet Mobi le Broadband Internet Mobile I nternet Broadb and Internet
A2 3, A2 6,
A2 7
A1 8, B1 -5 A30
A13
A3 , A4,
A6, A7, A 8,
A9, A10 ,
A15, D1
A1 , A5, A 6,
A1 1, A 14,
A19, A2 0
Not
i nclud ed

A1 A30 correspond to the consumers survey questions piloted in the GPS specifically for
telecommunications and access included in Work Package 3. Where others are mentioned
Mobile
24
(B1-5 and D1), these are questions piloted as part of other SIBIS topics are also relevant to
this topic
11
.

Below is a table including a selected list of T&A indicators, which was created by looking at
the most interesting indicators which were considered to be of primary importance for
tracking the development of telecommunications and access on a Pan-European level. Out
of the 18 indicators proposed below, 15 have been created using data directly piloted in
SIBIS GPS. Additionally, 1 indicator ('impact of broadband technologies'), has been analysed
using external data commissioned by Oftel in the UK.

Table 2: Indicators of relevance for SIBIS
Selected New Telecommunications and Access indicators relevant for the topic

No. Indicator Name Domain Sub-Domain
Cross-
Tabs
Piloted
in the
survey
1
Share of at home Internet users according to
type of bandwith (broadband, midband and
narrowband
SIBIS
Broadband By
country
and age
groups
X
2
Degree of Broadband technologies take up SIBIS GPS
Broadband By
country
X
3 Migrators to faster than dial -up connections Access Broadband By
country
X
4 Share of Internet users who access the
Internet from different locations -- (a) at home
(b) at the workplace (c) at an educational
institution (d) at free PIAP (e) at commercial
PIAP
Access Internet By
country
and age
groups
X
5 Intensity of online usage Access Internet By
country
X
Internet drop-outs - Share of persons who
used to have Internet access at home, and do
not have it anymore
Access Internet By
country
X
7 Degree of Internet access through mobile
phones
Access Mobile By
country
and age
groups
X
8 Users according to online tenure (share of
users according to length of time since first use
of the Internet)
Usage Internet By
country
and age
groups
X
9 Degree of e-mail networking intensity Usage Internet By
country
and age
groups
X
10 T&A Degree of multi -device online use.
(a) Digital TV
(b) PDA/ palmtop
(c) mobile phone
(d) others
Usage Internet By
country
and age
groups
X
11 Degree of mobile networking intensity: Degree
of mobile ownership networking among friends
and relatives

Usage Mobile By
country
X

11
There is potential overlap with all the other SIBIS topics. Telecommunications and access forms a baseline set
of indicators that can help with the analysis of numerous other indicators that are being developed.
25
No. Indicator Name Domain Sub-Domain
Cross-
Tabs
Piloted
in the
survey
12 Degree of SMS use according to different
purposes (communication, transactions,
downloads, news subscriptions)
Usage Mobile By
country
and age
groups
X
13 Use of mobile phones abroad Usage Mobile By
country
Not
piloted
14 Consumer perceptions of broadband

Impacts Broadband By
country
Not
piloted
15 Barriers to Internet usage Impacts Internet By
country
X
16 Compound Indicators: Internet Snapshot Impacts Internet Cluster
by
countries
X
17 Compound Indicators: Mobile Snapshot Impacts Mobile Cluster
by
countries
X
18 Compound Indicators: Migration Snapshot Impacts Broadband Cluster
by
countries
X

26
4 Analysis of Data

A large number of telecommunications and access indicators were piloted in the SIBIS
survey. Although these were still a subset of the number that was originally presented to go
forward in previous SIBIS work, they nevertheless constitute a large quantity of data to
analyse. This chapter presents the indicators developed in SIBIS and piloted in the GPS that
are considered to be of primary interest (presented in table 2). A number of the un-piloted
indicators will be presented in section 5 when looking at gaps and topics for future work,
and in the forthcoming SIBIS indicator handbook, where all relevant indicators piloted and
not piloted will be presented. Here we are concentrating our work on the indicators already
piloted.

4.1 Analysis of indicators about citizens and society
4.1.1 Access indicators

Degree of broadband technologies take-up
Although this is a relatively straightforward indicator, it is fundamental to measure the share
of Internet users who have access to broadband. Several other indicators stem from it and,
therefore, it was necessary to pilot it in the SIBIS survey. One of the principal features of
broadband in Europe is its diversity. Therefore, the question had to reflect the many
broadband access methods available across the participating countries. Hence respondents
to the SIBIS questionnaire were quizzed about the type of Internet connection they access at
home. Possible responses were; dial up modem, ISDN, satellite, cable modem, xDSL,
leased line, multiplex (T1/T3) or other. It is no surprise that dial up modems were the most
popular method of at-home connection. This is classified as narrowband as it less than
64Kbps. Further classification of the responses according to bandwidth is a contentious
process as definitions of bandwidth still vary.

The minimum bandwidth rate suggested for classification as broadband is >128Kbps, though
some classify >2Mbps as the threshold. The threshold is important because some definitions
of broadband include ISDN access that has an upper bandwidth of 128Kbps. In this
assessment, ISDN has been classified as midband and the definition of broadband as
>2Mbps has been adopted, in accordance with the practice of Oftel, the UK telecoms
regulator.

Broadband services are currently primarily provided through DSL and cable modem services.
Alternative technologies such as fixed wireless access, electric powerlines and satellites, are
anticipated, but have not been fully brought to market yet. This may be because of the
introduction of cheap-rate ADSL over the last year and a higher penetration of other
broadband technologies such as cable operators which currently have a large base of
subscribers in the UK, making it one of the most competitive markets in Europe in terms of
existing broadband infrastructures. Competitiveness in the broadband market in Europe is
still generally low. Most users upgrading to broadband are subscribing to incumbents deals.
Even though it is two years since Member States were required by EU law to open their local
exchanges to enable competing operators to provide DSL latest figures from February 2003
from the European Competitive Telecommunications Association (ECTA) revels that still less
than 5% of the nearly 200 million telephone lines in the EU are equipped for DSL. Moreover,
over 80% of current DSL exchanges are incumbents ones.
27

According to ITU
12
, one of the most interesting statistics used for comparison is the percent
of Internet subscribers that are broadband users. The results of a recent ITU study indicate
that broadband users almost always started by using a narrowband Internet connection.
Thus, Internet users can be considered to be the population from which broadband users are
drawn. The ratio of broadband users to total Internet users is the fraction of the relevant
population that has adopted the innovation of broadband.

Table 3: Share of Internet users according to type of connection at-home by country

BB
broadband
(e.g. DSL)
MB
only
"mid"band
(ISDN)
NB
Only
narrowband
(dial-up
modem)
only an
access type
not mentioned
DK No Access at
home

% % % % % %
AUSTRIA
11.06 6.48 17.96 1.33 10.26 52.91
BELGIUM
15.57 0.84 10.92 1.86 10.90 59.91
DENMARK
15.70 7.09 29.94 0.40 11.09 35.77
FINLAND
6.46 7.89 26.22 0.55 16.75 42.14
FRANCE
5.45 0.28 16.20 0.14 8.04 69.89
GERMANY
8.90 17.64 12.30 0.10 10.65 50.41
GREECE
1.53 1.76 8.38 0.17 4.97 83.19
IRELAND
1.87 2.97 25.77 1.29 20.69 47.41
ITALY
2.94 1.23 23.16 0.11 13.11 59.45
LUXEMBOURG
2.59 19.24 16.99 0.55 10.19 50.44
NETHERLANDS
15.12 13.20 34.30 0.10 9.69 27.59
PORTUGAL
3.31 1.17 10.60 0.38 5.97 78.57
SPAIN
5.76 0.48 13.81 0.21 10.00 69.72
SWEDEN
15.44 2.48 33.05 0.31 15.09 33.64
SWITZERLAND
9.53 10.17 19.72 1.04 19.84 39.70
UK
9.82 2.36 20.80 0.30 25.04 41.67
USA
16.57 0.10 34.56 0.35 11.88 36.54
EU15 7.50 5.68 17.96 0.26 12.73 55.87
BB= Broadband (includes cable, leased line, xDSL, T1 and T3, and Internet via Satellite technologies)
MB= Mid Band (ISDN)
NB = Narrowband
Other access type not mention (e.g. mobile)
Base: all respondents (N=11,832), weighted; EU15 results weighted by EU15 population (N=10,306).

The above figures illustrate the level of broadband enabled penetration among at-home
users in European Member States, Switzerland and the US. According to the data, in the
European Member States (EU15), approximately 7% of at-home users connections have
broadband. A new report by Forrester Research confirms SIBIS figures. According to this
study, by the end of 2002, the number of Western Europeans with broadband Internet
access at home had climbed to six million, or seven percent. This is a low figure if compared
to penetration of BB technologies in countries outside the EU, such as Japan or the US. In
2001, eMarketer research, estimated that broadband was adopted by less than 6% of
Japanese households, but it found that by the end of 2002 broadband was in 15.5% of

12
ITU Internet Reports 2002: Internet for a Mobile Generation.
28
Japanese households
13
which is just shy of the US at 16% last spring. While behind the US,
which has around 16%
14
, there are individual countries in the EU that seem to have equally
high levels of broadband penetration. For example, Belgium which benefits from a dense
urbanised geography where access to telecoms networks such as cable is relatively easy.
Similarly, Austria is high and is dominated by the Vienna urban area which is also highly
cabled. In other Member States the midband technology of ISDN is popular. Germany for
example has an ISDN penetration level of over 17% among the at-home Internet
connections. This may be due to an aggressive marketing of ISDN services to consumer
customers of Deutsche Telekom since the late 1990s. In contrast, the UK has an ISDN
penetration of around 2%. This may be because ISDN was primarily marketed as a business
product and had high associated installation costs. The lesson from midband is that
consumer can and do migrate to better access technologies.

Table 4: Percentage share of respondents by access type at home according to age group

Age groups (according to Eurostat classification)
broadband (e.g. DSL) Frequency 203.24 421.15 120.04 28.70
Column% 12.31 9.17 5.43 1.56
only "mid"band (ISDN) Frequency 91.43 378.66 100.10 11.09
Column% 5.54 8.24 4.53 0.60
only narrowband (dial-up
modem)
Frequency 392.72 1079.90 300.56 74.41
Column% 23.79 23.51 13.61 4.05
only an access type not
mentioned
Frequency 3.85 13.88 6.30 2.86
Column% 0.23 0.30 0.29 0.16
DK Frequency 231.52 627.70 342.67 107.96
Column% 14.03 13.67 15.51 5.87
no access at home Frequency 727.83 2071.70 1339.29 1614.30
Column% 44.09 45.11 60.63 87.77

When looking at age group comparisons, it is important to note that the majority of SIBIS
respondents older than 50 years old do not have an Internet connection at home. This is
particularly true for respondents older than 56, where 87% respondents do not access it from
home. On the contrary, the majority of SIBIS respondents younger than 50 years old have an
Internet connection at home.

13
Ministry of Public Management Home Affairs, Post and Telecommunications, Japan, January 2003
14
Currently and according to ITU (December 2002) the number of USA households connected via BB at home
surpasses 20%
29
Figure 5: Type of Internet connections at home by age groups
At home internet connections according to age groups in EU-15
0
5
10
15
20
25
broadband (e.g. DSL) only "mid"band (ISDN) only narrowband (dial-up modem)
only an access type not mentioned DK

Hence looking at Internet connections at home according to age groups, SIBIS results have
portrayed that people between 24 and 49 years old show higher adoptions rates of both high
speed and slow connections at-home. This correlates well with other indicators. For example,
the young are also more likely to migrate from low-speed to higher speed connections and
are traditionally described as early technology adopters. It is also the young who tend to be
more interested in downloading digital media.
Migrators to fast-speed connections
The analysis of the number of at-home Internet users who have migrated from slow dial-up
connections to a faster one (be that a broadband or midband connection), is of interest for a
variety of reasons. It could indicate a certain level of sophistication in usage or perhaps
indicate which countries have cheaper higher-speed connections. It is a good indicator to
combine with other indicators including the cost of different connections, pricing packages,
usage patterns and the availability of plug-and-go connections.

30
Table 5: Percentage of migrators from a slower connection to a higher speed connection by
country

Country Migrator (%) Non-migrator (%)
AUSTRIA 41.6 57.1
BELGIUM 48.4 46.0
DENMARK 82.0 17.1
FINLAND 63.3 36.7
FRANCE 54.9 43.0
GERMANY 43.7 55.6
GREECE 27.2 72.8
IRELAND 39.8 60.2
ITALY 55.4 32.0
LUXEMBOURG 44.9 51.2
NETHERLANDS 64.8 34.2
PORTUGAL 68.8 27.2
SPAIN 75.3 24.7
SWEDEN 83.9 16.1
SWITZERLAND 44.4 55.6
UK 43.1 56.1
EU15 51.1 47.3
USA 79.3 19.9

Question: At home, did you have a connection before which was slower than your current one?
Base: respondents with Internet fast connections at home (midband and broadband), who used to subscribe
before to a slower service (N=1,621), weighted. EU15 results weighted by EU15 population (N=1,376).
Source: SIBIS 2002, GPS. Column percentage.

The figures suggest that migration to a faster connection than dial-up is highest in the US,
Sweden, Spain, Finland, the Netherlands and Denmark. The distribution of non-migrators,
i.e. those who have not in the past had a slower connection and, therefore, have most likely
signed up straight away to a midband or broadband connection, appears to be concentrated
in the UK, Switzerland, Luxembourg, Ireland, and Austria.

The next step in indicator development in this area would be to explore the satisfaction levels
of people with connections, why they migrate and what barriers are still present.

Access to the Internet from multiple places
This indicator measures how long respondents spend, and from where they access the
Internet. The nature of this indicator is to move beyond counting instances to measuring time
factors. The data collected from this indicator can be presented in a number of ways.

It is clear from the charts than in the most mature European Internet markets (DK, S, N, FIN,
UK and CH), as in the USA there is a large amount of multi-context users who access the
Internet from both at home and at work locations. Whereas in the other countries Internet
users access the Internet generally from only one location, whether at home or at work.

SIBIS also shows data that confirms most users access the Internet principally from home.
However, although the reference is to access the Internet at home, the data also shows that
there are established patterns of bimodal usage as many users appear to access the
Internet from more than one location. In the US as in Scandinavian countries, the UK and the
Netherlands there is a large proportion of these bimodal users who access the Internet both
at home and at work.

31
Figure 6: Internet users accessing the Internet from different locations
Internet usage by location
0
10
20
30
40
50
60
70
80
90
100
D
KS
N
L
F
I
N
U
K
I
R
LDAL
E
U
-
1
5
B
I
EF
E
L
P
C
H
U
S
A
occasional usage (less than once a month)
usage only somewhere else
usage only at work
usage only at home
usage at home and at work

Question: Have you accessed the Internet in the last four weeks from
Base: respondents with Internet access at home who have us ed the Internet at least once in the last 4 weeks
(N=4,631). EU15 results weighted by EU15 population (N=3,838)

The indicator also looks at whether at-home users access the Internet from other locations
outside of the home and work, and, if so, how much time was spent accessing elsewhere in
a typical week. The choices were:

1 At-school
2 In public places with free internet access
3 In public places with paid Internet access
4 Another place

Response rates were low and, therefore, it is difficult to draw accurate conclusions for each
category. Such nomadic access is usually associated with life events such as studentship,
vacations or long term travel abroad, i.e. periods of life when there is no long-term residence.
Here it is important to highlight the fact that telephone survey methodologies do not include
responses from people younger than 16 years old, and, therefore, at-school responses were
low. This might have affected responses for In public places with free Internet access and
In public places with paid Internet access.

Different access locations can lead to different on-line activities and there are potentially
many disadvantages when compared to at-home access. In addition to being able to access
information at-home at any time, a person may be less likely to access, for example,
personal health or financial information from a library or other public facility.

A recent study in the USA
15
shows that ICT use by individuals is strongly affected by use in
the workplace. Not only are individuals who work in certain occupational areas more likely to
use a computer and the Internet, but exposure in the workplace makes an individual
substantially more likely to use a computer and the Internet. For example, 78% of

15
US Department of Commerce, 2002
32
households in which at least one family member uses a computer at work have a computer
at home. In contrast, only 36% of households in which no family member uses a computer at
work have a computer at home.

Figure 7: Intensity of online usage in different locations
Intensity of online usage
0
10
20
30
40
50
60
70
80
90
100
D
K
U
KS
N
L
DB
E
U
-
1
5AL
F
I
NE
I
R
L
I
PF
E
L
C
H
U
S
High (over 6 hrs/week) Medium (between 1 and 5 hrs/week) Low (less than 1 hrs/week)
occ. user (not asked) non-Internet user

Question: How much time do you spend in a typical week on using the Internet at home?
Base: respondents with Internet access at home who have used the Internet at least once in the last 4 weeks
(N=4,631). EU15 results weighted by EU15 population (N=3,838).

With regards to the intensity of online usage, the time spent on-line shows differing patterns
across all the European countries surveyed. Scandinavian countries and the UK show a
similar pattern to the US where the majority of users spent on-line for up to 5 hours per week.
On the contrary, the EU15 average, and notably in Southern countries, particularly in Spain,
the majority of users do spend lesser time using the Internet.

The USA shows a high level of at-home use, but have the advantage of free local calls,
encouraging more frequent connection to the Internet.

In the United States, use in all types of locations is rising. Similarly, a potential emerging area
for measuring an alternative Internet location is Wireless LANS which are being increasingly
deployed for providing Internet access in specific locations. Commercial Wi-Fi 'hot' spots are
springing up across a number of European countries at locations such as railway stations,
airports, business parks and coffee shops. There are already several thousand hot spots
across the US and numbers are expected to grow rapidly in Europe. As wireless 'hot' spots in
Europe become more prevalent, it would be interesting to ask about hotspots levels of use in
public places such as airports, restaurants, etc where access has been provided. This needs
to be distinguished from other free PIAPS.

33
Figure 8: Broadband access and intensity of online use
Broadband access and intensity of online usage
A
B
DK
FIN
F
D
EL
IRL
I
L
NL
P
E
S
CH
UK
US
EU-15
0
5
10
15
20
0 5 10 15 20 25 30 35
Intensive users (6 hrs and more per week)

Source: Technopolis analysis of SIBIS GPS 2002

The figure portrays a clear association between having a broadband at home connection at
home and intensity of Internet use. SIBIS shows how in countries with higher broadband
penetration rates at home, users tend to spend longer online per session when compared
with narrowband users. This indicates not only that broadband encourages users to use the
Internet more, but also that broadband technologies allow them to get a greater and more
effective online experience than narrowband users. Those who move to higher bandwidth
seek a better online experience, for example, through quicker downloads or always on
connection that enable better exploitation of the potential benefits of the Information Society
Internet Drop Outs
This indicator looks at the percentage of people in the EU Member States, Switzerland and
the US who used to have Internet access at home and have now stopped their connections.
There could be a number of reasons for this, for example, cost, Internet fatigue, faster
access at work. It could also be the case that the respondent falls in a transition phase and
has cancelled a narrowband connection, and is waiting for a faster one to be installed. This is
not a definite drop-put, but a migrator waiting to get a faster connection. Hence what is
needed in future work is some indication of why people drop out (cost, lack of interest, using
elsewhere) and whether the change is temporary or not.

In any case, if this indicator returns high levels it is worth investigating. The highest number
of Internet dropouts is found in the USA with 20.2% of respondents saying they used to have
a connection at home but do not anymore. In Europe, the countries with the highest drop out
rates are the UK, Sweden, and Denmark, all early adopting nations where users have had
time to experience a full cycle of Internet use.

34
Table 6: Percentage of Internet dropouts by country

Country Yes No
AUSTRIA 8.1 91.9
BELGIUM 6.6 93.4
DENMARK 9.5 90.5
FINLAND 6.3 93.7
FRANCE 3.3 96.7
GERMANY 3.9 96.1
GREECE 3.8 96.2
IRELAND 6.9 93.1
ITALY 2.8 97.2
LUXEMBOURG 3.8 96.2
NETHERLANDS 7.8 92.2
PORTUGAL 3.4 96.5
SPAIN 4.7 94.7
SWEDEN 13.7 86.3
SWITZERLAND 6.8 93.2
UK 12.2 87.5
EU15 5.09 94.7
USA 20.2 79.8

Question: Did you once have Internet access in your home?
Base: Share of respondents with formerly Internet access at home (N=6,266), weighted. EU15 results weighted
by EU15 population (N=5,635).

35
4.1.2 Usage indicators

Users according to on-line tenure
This indicator measures share of users according to length of time; tenure, since first use of
the Internet. The SIBIS survey classifies tenure into three bands; less than six months since
first Internet usage; more than six months, but less than two years; and more than two years.

According to the SIBIS results, on-line tenure is highest in the US, Scandinavian countries,
Germany and UK, and is generally lower in the Mediterranean countries. These Northern
European countries have more experience of usage and also more mature Internet markets.
There are also other issues of culture and language. Therefore, it can be expected that
Northern European countries have higher rates for Internet penetration rates, e-commerce
transactions and e-skills aptitudes. It would be rewarding to compare broadband subscribers
with on-line tenure, as well as on-line tenure to different age groups, regions and gender
issues. Such indicators could be very useful for tracking the digital divide. Following these
guidelines SIBIS has included among the Telecoms and Access compound indicators one
exploring the correlation between online tenure, migrators to a higher-speed connection, and
Internet usage among the at-home population.

Thus this indicator portrayed how on-line tenure is an important indicator because depending
on how experienced users are with the Internet, the more likely they are to use the Internet
effectively and widely. It is also more likely that the more experienced users are, the more
willing they are to migrate to a broadband connection.

Table 7: On-line tenure by country

First use of Internet
Country < 6 months
ago
6-12 months
ago
1-2 years ago 2 years +
ago
AUSTRIA 10.4 6. 4 20.6 62.3
BELGIUM 8.1 8.7 22.3 60.6
DENMARK 1.4 2.9 12.0 83.5
FINLAND 1.4 4.2 7.9 86.5
FRANCE 15.4 12.5 27.9 43.8
GERMANY 5.8 11.9 35.4 46.6
GREECE 9.6 16.6 25.6 47.2
IRELAND 5.3 10.2 28.0 55.7
ITALY 7.2 9.9 24.5 57.6
LUXEMBOURG 18.0 2.7 20.4 59.0
NETHERLANDS 3.3 8.2 25.1 63.4
PORTUGAL 9.2 7.0 17.4 64.4
SPAIN 8.0 12.0 33.3 46.3
SWEDEN 1.8 3.9 11.5 82.8
SWITZERLAND 9.6 8.4 22.3 56.9
UK 4.7 12.5 22.8 59.2
EU15 7.0 10.8 26.9 54.8
USA 3.3 3.4 14.4 78.8
Base: Respondents who accessed the Internet at least once in the last 12 months (N=6,905), EU15 results
weighted by EU 15 population (N=5,828)
Source: SIBIS 2002, GPS Column percentage.

Currently, the measurement used for early adopters users in SIBIS is over 2 years, a
measure that is already taking in a large proportion of the respondents. In the US for
example, the majority of users (78%) have two years or more experience. For future work, it
36
would be interesting to change this indicator and class early adopters as those who have had
perhaps 5 years or more experience of using the Internet. This would give more leverage to
explore and distinguish the usage patterns between the experienced and the non-
experienced.

Figure 9: On-line tenure according to country
Experience of online usage
0
10
20
30
40
50
60
70
80
90
100
D
KS
F
I
N
N
L
U
KA
I
R
LLB
E
U
-
1
5
D
I
PEF
E
L
C
H
U
S
2 years and more 1 year - 2 years 6-12 months < 6 months Don't know non-Internet users

Base: all Internet users (incl. users, who did not use it in the last 4 weeks), weighted (N=6,905). EU15 results
weighted by EU15 population (N=5,828).

Degree of e-mail networking intensity
For this indicator, respondents were asked to indicate their degree of networking among
friends and relatives and the degree of usage of e-mail networks with friends and relatives.
E-mail remains a basic tool for interacting on-line and the measurement of e-mail penetration
is important. It indicates a good level of Internet literacy and is fundamental to the completion
of many on-line processes from registration to transaction. It is already been tracked by Oftel
in the UK and by the Ministry of Labour in Finland. E-mail penetration among friends and
family was highest in Scandinavian countries and the US, where over half of the
respondents friends and family have an e-mail address.

According to the OECD 2002 ICT Outlook report, in most countries, regular home users use
the Internet most frequently for e-mail and information search. In Italy for example, the
respective response rates are 63% and 60%. In spite of these high rates it is important to
continue tracking this networking intensity among friends and family, especially considering
that the project will soon include survey results from a number of candidate countries.

It is interesting to note the US networking intensity among friends and family is twice as high
as the EU average. One element which SIBIS did not track and which could have been
interesting to explore is user usage of multiple e-mail addresses, i.e. how many of the
respondents have more than one active e-mail address in order to distinguish between at-
home and at-work e-mail networking and usage.

37
Cutting the data by different age groups reveals the majority of e-mail users are within the
younger age group 16-25. This figure would have probably been higher if the age group of
people younger than 16 years old had been included. Lower rates were recorded among
those over 50, but the total figures for e-mail network intensity show that while around 40% of
all SIBIS respondents communicated with friends or relatives through e-mail, over 50%
claimed not to use e-mail at all.

Analysis of PC-ownership and e-mail usage indicates that 15% of PC-owners users do not
communicate through e-mail.

Table 8: E-mail usage by age groups (EU15)

Age
E-Mail use in last 4
weeks
up to 24 25 to 49 50 to 64
65 and
more
Total
Usage
54 48.5 27.5 6.4 41.0
No usage (but PC-
Users)
24 18.2 13.1 6.1 15.3
No usage (PC-non-
users)
21.8 33.3 59.2 87.3 43.5
DK
1 0.1 2 - 0.1
Total
100 100 100 100 100
Question: Have you used or received any email messages for work or for private purposes during the last four
weeks?
Base: all respondents in EU15 (N=10,306); weighted by EU15 population

Figure 10. E-mail use according to different age groups:
E-mail users according to age groups in EU-15
0
10
20
30
40
50
60
70
80
90
100
E-mail users No usage (but PC-users) No usage (PC-non-users) don't know

Question: Have you used or received any email messages for work or for private purposes during the last four
weeks?
Base: all respondents in EU15 (N=10,306); weighted by EU15 population
38
Users accessing the Internet through other devices than a PC/ MAC
This question asked the respondents if in the last 4 weeks they had used any other devices
to access the Internet or on-line services, and if so, had they used:

Digital TV
PDA/Palmtop
WAP/GPRS
Games console
Other

It is important to study the degree to which different devices are being used for Internet
connections. In Europe, due to the current high penetration of mobile phones, the data
collected in SIBIS suggests emergent patterns of bimodal usage. However, the sample of
respondents presented below is too low to judge country patterns. This work is best analysed
as this stage through looking at the differences between the US and the EU average and
differences of usage among different age groups.

Figure 11: USA vs. EU15 other connections devices than a PC
Users accesing the Internet at home from alternative devices other than a PC
0
10
20
30
40
50
60
70
80
90
100
Other access
method: Digital
TV
Other access
method:
PDA/palmtop
Other access
method: mobile
phone w. WAP
Other access
method: game
console
Other access
method: other
Other access
method: DK
%
USA
EU15

Question: In the last four weeks did you access the Internet in any other way than via PC or Mac at least once?
Which devices did you use for that: Did you use ...(multiple response)?
Base: Internet-users who access the Internet also by other devices than PC or Mac, EU15 results weighted by
EU15 population (N=340). USA results weighted (N=55). Multiple responses permitted

In the US at-home users are consistently accessing the Internet through more platforms than
in the EU. This is particularly true for PDA/Palmtop, game consoles and digital television
access. Digital television access is twice as high in the US than in the EU. Likewise access
through PDA/Palmtop devices was four times higher. The results change for mobile phones
where the EU shows higher rates of mobile telephone Internet access than the US. The
other figure appears to be quite high. There could be any number of reasons for this,
including semantic confusion, for example, not counting a laptop as a PC.

39
If considering age groups SIBIS has noted that users younger than 25 include the highest
rates of early adopters and multi-device users. Particularly for devices such as mobile, digital
TV or game consoles. It has to be highlighted at this stage that because the data was
gathered using telephony surveys, which do not allow respondents younger than 16 years
old to be interviewed, early technology adopters could not be included in the response. Some
devices figures such as those for game console users are very likely to be under
represented.

Degree of mobile networking intensity: Degree of mobile ownership networking
among friends and relatives
This indicator measures the degree of ownership among friends and relatives. It is interesting
to highlight that over 60% of the respondents relatives and friends owned a mobile phone in
the EU, whereas in the US mobile intensity network among friends and family only reached
40%. Likewise, when looking at the different age groups, people younger than 25 years old
accounted for the highest penetration rates, since they stated that over 80% own a mobile
phone and over 75% of their family and friends are also mobile phone owners. It is also very
interesting to look at the degree of penetration of SMS use since this age group again
recorded the highest usage, with 70% of mobile owners below 25 years old using SMS.

Degree of SMS mobile data services use for communication, transactions, downloads,
news subscriptions

Figure12: Mobile phone owners according to different usage: EU 15, US and Switzerland, April
2002
Mobile usage and usage of SMS and Internet via mobile
0
10
20
30
40
50
60
70
80
90
100
B
D
KD
E
LEF
I
R
L
I
L
N
L
AP
F
I
NS
U
K
E
U
-
1
5
C
H
U
S
mobile Internet user only SMS user neither nor, but mobile phone owner

Base: all respondents (N=11,832), weighted; EU15 results weighted by EU15 population (N=10,306).

Given that mobile telephony penetration is generally high (more than 65 %) across all EU 15
countries, in order to explore mobile phone usage SIBIS data has been broken by different
age groups according to the degree of SMS usage in the last four weeks. Results have
portrayed that SMSs are generally popular across mobile phone owners younger than 50
years old. This is specially true for mobile phone owners younger than 25, since the majority
of them ( 80%) have used SMSs in the last four weeks.

40
Out of all of the usage options, the most popular use of SMS is for communication, and this
is a common pattern among all age groups. Looking at the other types of services that are
available via SMS, it would be interesting to further explore the use of SMS for downloads,
particularly by age groups. For example, are the young more likely to download ringtones
and logos than other age groups? It would also be interesting to look at the types of services
used in the area of WAP technology to see how different age groups are using the
technology, particularly as the under 25s are the main target user base for future multiple
connections methods.

Figure 13: Degree of SMS mobile data services used (communication, transactions,
downloads, news subscriptions)
SMSs use according to age groups in EU-15
(in % of mobile owners)
0
10
20
30
40
50
60
70
80
90
100
Up to 24 25-49 50-64 65+ Total
Communicate with people Paying for: purchases, tickets or similar
Paying for: downloads, ringtones Receiving subscription services

Question: Have you, in the last four weeks, used SMS messages for?
Base: all mobile user who have used SMS in the last four weeks , weighted by EU15 population (N=4,443)
Source: SIBIS 2002, GPS;

When looking at different mobile usage patterns according to age groups, it is important to
highlight that SIBIS respondents younger than 25 years old are the most active SMS users.
In addition, they access the Internet though their mobile phones to a larger degree than other
age groups, though this figure still remains below 10%. Subsequently, it is not a surprise that
young mobile users are generally also mobile data users, whether using only SMS or also
Internet data services. Whereas in other age groups there is a higher propensity of mobile
users to use only voice calls and not mobile data (SMS or mobile Internet ). Moreover, a big
proportion of respondents older than 50 years do not own a mobile phone for personal use.
Hence, for future indicators tracking the proportion of bimodal users accessing the Internet
through both mobile phones and computers, it seems that age will play a role when defining
future mobile Internet user profiles, being among the main explanatory demographic factors.

41
Figure 14: Mobile phones use according to age groups
Mobile usage according to age groups in EU-15
(in % of each age group)
0
10
20
30
40
50
60
70
80
mobile Internet user only SMS user neither SMS nor mobile Internet (incl. don't know) no mobile phone

Base: all respondents weighted by EU15-population (N=10,306),

4.1.3 Impact Indicators

Consumer perceptions of broadband
Broadband access to the Internet is often talked about in revered terms by consumers. Many
wish to use it, but the reality is, as witnessed by SIBIS's own indicator, that few have access
to it. However, it is important to be able to measure the perceived benefits of it and the
impact it has on consumers. A comparative benchmarking survey
16
in major nations to
understand consumer perceptions of broadband has shown that fewer than 20% consider
broadband to be 'worse than expected. In the UK and Italy over 40% of respondents
considered it to be better than expected.

Barriers to using the Internet
As Europe moves towards universal access to the Internet and investment in public Internet
access points and basic training schemes become more prevalent, it is necessary to compile
sets indicators such barriers to usage. If the Internet is available everywhere, why are some
people not using it? What are the barriers and inhibitors to adopt broadband technologies?,
and why are some people dropping out from it?.

From the data, it is clear that personal preference not to use the Internet is one of the largest
barriers. Nearly 59% of the survey respondents who did not use the Internet agreed with the
statement that 'it was not something for me' (a general sentiment that is a catch all which
could include opinions that also agree with other statements in the survey). A more specific
barrier was a lack of ICT skills. Other specific barriers were that the Internet is perceived as

16
International Benchmarking study commissioned by Oftel, UK, March 2002.
42
expensive and time wasting. It is interesting to note that a lack of suitable content was only
cited by around 22%. It would appear that the content gap has been closed.

Table 9: Barriers to Internet Usage: percentage of respondents agreeing or disagreeing with
statements describing why they do not use the Internet (include frequency columns)

Barriers to Internet usage
Agree completely Agree
somewhat
Do not
agree
DK TOTAL
It requires advanced
computer skills
30.21 27.58 25.30 16.91 100
It is not easy
enough to get
access to
15.32 20.39 36.83 27.46 100
It is too time
consuming
23.38 24.20 25.47 26.95 100
It is too expensive
to use
22.45 21.15 20.49 35.90 100
It lacks useful or
interesting
information
10.76 11.31 50.55 27.38 100
It is not something
for me
40.99 16.58 31.74 10.69 100
Please tell me for each statement whether you agree completely, agree somewhat or do not agree. The
Internet
Base: Respondents who have not accessed the Internet in the last 4 weeks, weighted by EU15 population
(N=5,321).

Figure 15: Barriers to Internet usage: percentage of respondents agreeing or disagreeing with
statements describing why they do not use the Internet
0 10 20 30 40 50 60 70 80 90 100
It r equires adv anced
computer skills
It is not easy enough to
get access to
I T i s too time consuming
I t i s too expensive to use
It lacks useful or
i nteresting information
It is not something for
me
Percentage
Agree completely Agr ee somewhat Do not agree Don't know

Please tell me for each statement about the Internet whether you agree completely, agree somewhat or do not
agree.
Base: Respondents who have not accessed the Internet in the last 4 weeks, weighted by EU15-population
(N=5,321).
Source: SIBIS 2002, GPS.
43
5 The use of compound indices

Consultation with the UK Office of National Statistics confirmed that it is acceptable to
construct compound indices, providing:

The data is from recognised and acknowledged sources
That is represents sample sizes/time periods which are sufficiently large/close in order for
the analysis to be meaningful

Firstly, one way of creating compound indices is to use the concept of eReadiness/digital
divide which is already the subject of several studies, including from the World Bank and the
OECD. These often concentrate on developing countries and access to basic ICT. Under the
terms of this study it could be instructive to look at patterns of access to emerging
technologies across the EU and whether there is a correlation with say wealth, educational or
employment status.

In addition, a second way of structuring compound indicators is to go back to the conclusions
of the research work for SIBIS. This initial work showed that there are already many
indicators measuring the existence of something, fewer measuring the use of that
technology and fewest measuring the impact or benefit of that technology. Thus, combining
them together can give us a snapshot index.

Thirdly, the way the individual indicators have been grouped in the previous section could
also provide compound indicators just by asking all the questions under one of the
headings.

Since, as stated above, e-readiness indices have already been developed elsewhere, a
second approach to compound indicators construction can also be undertaken. This is
through combining variables together which have shown to be relevant to provide a
snapshot index. Variables included in this analysis have been derived from the SIBIS GPS
data results, including innovative indicators regarding use and adoption of new technologies.
Following the methodological guidelines provided by the UK Statistical Office, compound
indicators have been constructed using exclusively SIBIS data. Originally, there was an
intention to develop indices combining SIBIS data with external sources, but either the data
provided was not as up to date as SIBIS, or were not sufficiently large in scope, covering
only a few countries. Although these indicators, at this stage tend to produce results which
are fairly obvious, they are statistically sound and this approach can eventually be extended
to combine telecommunications and access indicators with other SIBIS indicators developed
under the other topic areas (such as social inclusion, e-commerce, work, employment and
skills for example). Moreover, they will gain in importance once they are combined in the
second phase of the project on a EU25 Pan-European basis. No other available source right
now can provide such a broad snapshot across European countries.

In order to combine the variables and study their interactions, a multivariate factor analysis of
the data was carried out. The principal advantage of this approach is that it does not rely
solely on subjective judgement (it does not involve weighting of variables), but it is primarily
dependent on groupings of statistically similar entities. This confers a degree of objective
validity to the procedure which is at least beneficial as a starting point, although there is
certainly a place for using judgement as to the utility of the results obtained, and for use of
manual modifications as appropriate.

44
The purpose of this methodology is to:

Improve understanding of the relationships between various aspects of ICT access and
usage
Determine whether the various indicators implicit in the survey questions are closely
correlated and can be combined to form compound indicators
To assess whether the survey responses contain information which can be used to
explain variations in ICT usage in terms of, for example, differences in individuals social
and economic attributes
To test for statistically significant differences between countries, and whether there is
evidence for distinct clusters of nations with similar ICT characteristics within each
cluster, but significant differences between clusters.

5.1 The data

Use has been made of the full SIBIS survey data set, containing over 11,000 records of the
responses of individuals from the 15 member states, Switzerland and the USA.

Numerous modifications were done to the coding of the data to make it suitable for analysis.
In addition, some questions were unsuitable for inclusion on the grounds that they were only
put to a biased sample of respondents. For example, questions relating to the perceived
degree of ease/difficulty of use of the Internet were only asked if the interviewee had
indicated that they were a user. In addition, records with important omissions, such as dont
knows, were omitted.

These modifications left a data set comprising of some 30 variables and nearly 5,000
records, very adequate for EU-wide multivariate factor analysis. The variables related mainly
to PC, e-mail and Internet access and usage, use of mobile telephones, and socio-economic
and demographic status.

5.1.1 Compound Indicators Internet Snapshot

The first compound to explore comprises Internet access and usage (for both narrowband
and broadband connections). The following four original variables are the main components
of this factor:

Table 10: Internet usage snapshot

Indicator
Source
Description
SIBIS GPS Percentage using a PC in the previous 4 weeks
SIBIS GPS Percentage with access to the Internet at home
SIBIS GPS Percentage using the Internet in the previous 4 weeks

45
5.1.2 Mobile snapshot

A second compound indicator to explore comprises mobile telephone ownership and what
might be termed basic patterns of usage and communication that do not involve the Mobile
Internet.

The following three original variables are the main components of this factor:

Table 11: Mobile usage snapshot

Indicator Source Description
SIBIS GPS
Ownership of a mobile telephone
SIBIS GPS
Having a high proportion of family/friends owning
mobiles
SIBIS GPS
High use of SMS messages for communication with
other people.

5.1.3 Migration to faster Internet Access snapshot (T&A 18)

A third compound indicator comprises at home active users with more than two years
experience since first use of the Internet who used to have a slow at-home Internet
connection before and have now migrated to a faster connection than dial-up.

The following three original variables are the main components of this compound:

Table 12: Migration to faster Internet Access snapshot

Indicator Source Description
SIBIS GPS
Internet At-Home users in the last four weeks
SIBIS GPS
Having more than two years online tenure since first use
of the Internet
SIBIS GPS
Having migrated to a faster connection than dial-up

5.1.4 Multiple regression as a means of examining linkages between ICT usage and
other standard variable

Additionally, and in the interest of examining the extent to which ICT usage can be
associated with an individuals social and economic circumstances, different demographic
variables, such as Terminal education age, Occupational status, and age group, have
been selected as independent variables to conduct a Multiple Regression statistical data
analysis. Hence, these three variables will be used in an attempt to explain plausible
variations in different dependent variables related to access, use and impact of Internet and
mobile technologies, such as:

use of a PC in the previous 4 weeks
sending or receiving an e-mail in the previous 4 weeks
Time spent using the Internet at home
46
Time spent using the Internet at work
ownership of a mobile telephone
Use of mobile to access the Internet

5.2 Snapshot indicators and country groupings factor and cluster
analysis

The resulting dataset was subject to factor analysis, designed to look for a small number of
higher-level factors consisting of combinations of strongly correlated groups of variables
from the original dataset. In this way the original set of variables may be simplified and the
results may suggest a usable set of compound indicators made up of combinations of raw
measures.

The principal advantage of this approach is that it does not rely solely on subjective
judgement, but is primarily dependent on groupings of statistically similar entities. This
confers a degree of objective validity to the procedure which is at least beneficial as a
starting point, although there is certainly a place for using judgement as to the utility of the
results obtained, and for use of manual modifications as appropriate.

Having identified a set of factors in this way, a cluster analysis was performed on each to
identify groups of countries with similar characteristics with respect to that factor. In general,
cluster analysis is used to group members of a population (of, for example, animals,
companies or countries) into classes with similar attributes, and hence to define useful
groupings. In the current application, we seek to identify groups of countries where the
pattern of (for example) IT usage (one of the identified factors, see below), is similar within
each group. This is done by looking, for each pair of countries, at statistical differences (or
distances) between the average values of the variables comprising the new factor, and
grouping those countries where the differences are relatively small.

A cluster analysis was then carried out to examine how countries in the survey compared
with respect to this factor. As shown below, three very clear and distinct clusters of countries
emerged, which might be classified as high, medium and low PC impact.

Factor 1: Internet usage snapshot

The Internet snapshot explores Internet access and usage (for both narrowband and
broadband connections). The data shows that access to and use of PCs and the Internet is
typically twice as prevalent in the cluster 1 defined countries as in the cluster 3 defined
countries. The three variables which form this snapshot (percentage using a PC in the
previous 4 weeks, percentage with access to the Internet at home. Percentage using the
Internet in the last 4 weeks) appear to offer a useful compound indicator of penetration level.
Thus, Cluster 1, which includes highly mature Internet markets such as the US and
Scandinavian nations, presents a more dynamic trend than the other two clusters of less
developed Internet markets. Additional analysis indicates that most respondents who have
used a PC in the last four weeks also have an at-home Internet connection. They have also
accessed the Internet in the last four weeks and most of them also have Internet at home.

In Cluster 2, and especially in Cluster 3, PC usage is at a higher level than Internet usage
within the last four weeks. It is no surprise that it follows that the measure of Internet at
home connections is even lower than that. This factor implies that in Mediterranean countries
many PC users do not access the Internet in the same time period. It also suggests that
many people who access the Internet do not have an Internet connection at home.
47

This composite indicator is useful in that it highlights some usage patterns. Interpreted as an
index it could be transferred to the SIBIS+ candidate countries to track the degree of Internet
impact.

Figure 16: Internet usage snapshot
0
10
20
30
40
50
60
70
80
0 10 20 30 40 50 60 70 80
Has used a PC in last 4 weeks (%)
Has used Internet in last 4 weeks (%)
EL
P
I
F
E
B
D
A
IR
L
UK
FI
N
DK
CH
US
NL
S
Cluster 3
Low Impact
Cluster 2
Medium Impact
Cluster 1
High Impact
Note: Luxembourg is hidden
behind Germany
L
EU-15

Key: The size of the 'has used Internet in last 4 weeks' circles represents the percentage of the
sample not the volume
Note: Luxembourg is hidden behind Germany.
Source: Technopolis analysis of SIBIS 2002, GPS

Membership of the three clusters is as follows.

Cluster 1: high impact Cluster 2: medium impact Cluster 3 low impact
Denmark Austria France
Finland Belgium Greece
Netherlands Germany Italy
Sweden Ireland Portugal
Switzerland Luxembourg Spain
US
UK

Factor 2: The Mobile Telephony snapshot
A second group of clusters to emerge comprises mobile telephone ownership and what
might be termed basic patterns of usage, which do not involve the Internet.

The most important contributory variables in this case are:

A high proportion of family/friends owning mobiles
48
High use of SMS for communication with other people.

A cluster analysis on these constituents produced the following:

Figure 17: Mobile Telephony Snapshot
50
60
70
80
90
100
50 60 70 80 90
Mobile phone ownership (%)
M
o
r
e

t
h
a
n

7
5
%

o
f

f
r
i
e
n
d
s

a
n
d

f
a
m
i
l
y
a
l
s
o

o
w
n

m
o
b
i
l
e

p
h
o
n
e
s

(
%
)
High SMS usage (%)
US
20
F
42
EL
53
E
58
P
56
B
70
CH
69
DK
48
D
58
I
61
UK
63
NL
38
IRL
72
FIN
79
L
53
A
66
S
53
Cluster 2
Cluster 3
Cluster 1
EU-15
57


In this case, membership of clusters falls as follows:

Cluster 1 Cluster 2 Cluster 3
Denmark France Austria
Greece USA Belgium
Italy Finland
Luxembourg Germany
Netherlands Ireland
Portugal Switzerland
Spain UK
Sweden

Regarding ownership of a mobile by respondents and the prevalence of a high proportion of
their family and friends owning mobiles, there is little to distinguish Clusters 1 and 3, as can
be seen in the figure above. However, the prevalence of mobile ownership in Cluster 2
comprising France and the USA is clearly much lower.
The main variable distinguishing members of the three clusters is SMS communication with
other people. It is over twice as high in Cluster 3 countries as in Cluster 2 countries.
Hence, it is clear that although mobile penetration is currently quite high in most Western
countries, differences in usage patterns between countries are occurring, as well as
divergences in the use of data mobile services for communication.
Another potentially interesting factor analysis for future work would combine the central
themes of Internet and mobile telephony usage. Had the data sets been large enough to be
statistically significant the following variables could have been combined:

49
Proportion of those using the Internet who have accessed it in a way other than via a PC
Proportion of those owning a mobile with WAP or 2.5G capability who have used it for
viewing WAP pages or web pages, or for e-mail, in the previous four weeks
Proportion of those with such a mobile who used it for making purchases via the Internet
Proportion of those with a mobile who have used SMS messages for receiving financial
information, sports results or other subscription services.

This factor could perhaps be described as representing sophisticated usage of a mobile,
and as such may form a potentially useful compound indicator.

Care in interpretation would be required on account of the highly filtered or conditional
nature of some of the questions in this group. For these reasons, specification of the
countries contained in each cluster is likely to mislead rather than inform at this stage.
However, it is an important indicator to include in future work.

Factor 3: Migration to a faster Internet connection snapshot
Traditionally two factors have influenced when Internet users migrate to a faster connection.
Based on experiences of the US and Scandinavian markets, it has been noted that once a
majority of a total population has Internet access there is a migration of users with tenure,
commonly defined as those with two years or more Internet experience, to faster
connections. They seek a better online experience e.g. quicker downloads, always on
connections. The clustering of the US, Sweden, Finland and Denmark illustrates this in the
figure below. Not only do they have relatively high percentages of regular and tenured
Internet users, but the size of the group with faster connections than dial-up is large too.
Above 60% of SIBIS active at-home Internet users (those who have accessed the Internet at
least once in the last four weeks). In comparison, there is a clear second cluster of countries
where this migration level is lower. The exception here seems to be the Netherlands which
registers similar migration levels, as can be seen in the bubble size, despite having lower
amount of users in the last four weeks with more than two years tenure than the first cluster
of countries.

Figure 18: Migration to faster access
20
30
40
50
60
70
80
90
100
20 30 40 50 60 70 80 90 100
Population who have accessed the Internet in last four weeks (%)
P
o
p
u
l
a
t
i
o
n

w
h
o

h
a
v
e

h
a
d

a
c
c
e
s
s
t
o

t
h
e

I
n
t
e
r
n
e
t

f
o
r

2

y
e
a
r
s
+
Users who have migrated to a faster than dial-up connection (%)
EL
5
P
14
E
15
F
10
I
6
B
20
EU-15
15
D
23
L
20
A
15
IRL
4
CH
14
UK
9
NL
25
US
21
S
23
FIN
16
DK
29
Typical migration path
Note: Ireland is hidden behind
Luxembourg


50
The more experienced users are, the more likely they are to upgrade to higher bandwidth
and, therefore, to being realistically able to use all the benefits and services within the
Information Society.

Effects of circumstances on IT usage multiple regression
Multiple regression enables the testing of relationships between a dependent variable and a
set of independent variables thought likely to influence it. The intention is to explain
variations in the dependent variable in terms of variations in the independent variables the
statistical robustness of the overall regression can be assessed, as can the strength of the
influence of each individual independent variable.

In this application, the interest is in exploring the extent to which variations in questionnaire
results on aspects of telecommunications and access between individuals can be explained
by differences in social and economic circumstances. The following were chosen as
independent variables due to their presence in the survey data:

Terminal education age
Occupational status
Household Income

A fourth socio-economic variable, Social Grade, was not used as it is defined in terms of
education and occupation, and would lead to severe multi-colinearity problems.

So far, these three variables have been used in an attempt to explain variations in the
following six dependent variables in turn:

Use of a PC in the previous four weeks
Sending or receiving an e-mail in the previous four weeks
Time spent using the Internet at home
Time spent using the Internet at work
Use of mobile to access the Internet

Some summary results are provided below.

Table 13: Multiple regression results

Dep. Variable Terminal
Education Age
Occupational
Status
Household income
Coefficient t Coefficient t Coefficient t
Use of PC
1.22 18.7 0.02 16.6 0.06 9.33
Use of e-mail
0.13 43.4 0.03 19.2 0.05 7.76
Time on Internet
at home
0.21 10.6 0.04 8.81 0.13 7.0
Time on Internet
at work
0.28 14.1 0.10 22.2 0.09 5.1
Ownership of
mobile
0.02 2.8 .004 2.9 0.05 8.82
Internet access
from mobile
-.002 -0.56 0.02 2.7 -0 -0.09

For the first four regressions, coverage use of a PC e-mail and Internet, all the variables
(judging by the t-statistics) are significantly different from zero, and all act in the same
direction more education, higher occupational status and high household income all imply
51
greater use of each of the facilities. Associations with the ownership of a mobile, however,
are much weaker, and barely significant. Association with the use of a mobile to access the
Internet is totally insignificant.
52
6 Further developments: remaining gaps in the statistical
coverage of the topic

This chapter describes on the one hand the 'gaps', i.e. those areas within the topic where
official statistics currently fail to provide adequate measures for the issues discussed
previously in SIBIS. On the other hand, it also comments on some methodological questions
which could be complemented and discussed in further work to be conducted after SIBIS. It
generally shows:

Where and why the SIBIS project has not developed indicators in certain areas
And how and why these indicators should be collected in future work

Therefore the main aim of this chapter is to both fill some existing gaps and introduce some
suggestions for future work on the topic area. Some of the indicators which could not be
piloted in the GPS are nevertheless explained and specified in the forthcoming Indicator
handbook, so it would be possible for relevant stakeholder to collect them in future work.
Therefore, some of these considerations will be further discussed, and illustrated with
existing and recent complementary data from external sources. Given this, it seems very
relevant to highlight future work that should be done to cover existing gaps in
telecommunications and access indicators, particularly those that can assist in mapping the
digital divide.

While it has only taken approximately five years for Internet penetration of households to rise
from 5% to 50% it may take longer to achieve universal access. Evidence from Scandinavia
and the US indicates that the penetration levels start to plateau at around 65% of the
population. Prior to this point it is imperative to understand and measure the digital divide.
Also it is becoming increasingly important to improve understanding of existing devices in
order to follow progress on how they can narrow the digital divide.

It is also important to understand which demographic or social groups are under-using or not
using new technologies. As universal services such as education, social support, and
medicine are increasingly being delivered, at least in part, online, it is highly probable that a
new type of social exclusion will occur. Therefore, identifying availability of technologies by
region, uptake, access device, and usage patterns and analysing them by income, education
and social measures is a way forward. But qualitative regional surveys about barriers to
technology access will also be required if solutions are to be developed.

6.1 The existing gap on regional level data indicators for T&A.

There is a data black hole at sub-national levels within Europe. This is due to two reasons:
1. Definitions of regions vary. 2. Cost of surveying.

Variations in regional definitions are problematic for the collection of indicator data at a sub-
national level. Disparities occur at a number of levels.

While Eurostat's definitions and hierarchy of regions is well established it does not always
concur with regions as defined in policy. Many of the European regional development and
social fund programmes are being targeted at sub-regional levels. So while a programme
may be called 'East of Scotland' its policies are targeted at a ward level across multiple
Eurostat regions in the East of Scotland. Such postcoding is effective at targeting the delivery
53
of development aid but it creates immense difficulties for measuring the impact of the
Information Society within the communities. While it is said that telecommunications is a
borderless technology, the regional impact of networks is important to understand. European
Union structural funds are being spent on improving regional access and use of these
networks. Such spending is justified on the basis of the impact on citizens and enterprises in
those regions. Therefore, it must be open to measurement in order to assess the impact. The
importance of understanding the impact of networks on regions is highlighted by preliminary
research by DG Regional Policies that indicates that Information Society actions within
Structural Funds programmes account for an estimated expenditure of EUR 15,849 billion.

Official definitions of regions, such as those used by Eurostat, can and do vary from national
definitions of regions and, therefore, existing measurements of the availability of
technologies. Many measures of media penetration and usage are based on amalgamations
of regions or on the geographic coverage of transmission stations. Such broadcast areas are
of particular interest to the commercial sector so have been extensively used to map new
technology penetration.

Custom surveying of a population within such specific regions is possible, but it is expensive.
Plus, the demand for statistically representative sampling can be frustrated simply because a
region has a small population and a small uptake of a technology. Until a technology is
widely adopted the user base will be too small to accurately measure in order to state that it
is in an early-adopter phase.

Collecting profile information at postcode level is a possible solution. Aggregating the
information allows the creation of regions to match other definitions but this is a data- and
time consuming approach that is best deployed over a long period before representative
results can be analysed.

3G - An imperative to track
Third generation mobile phone technologies, (3G) always grab headlines, for good reasons
and for bad. In the current economic climate the headlines are associated with the revision of
rollout plans across Europe, as 3G license-holders try to conserve cash.

These are signs of a new realism entering the mobile sector. The news that availability of a
full 3G mobile telecommunications network will not be available in Sweden for another four
years, could signify delay of one of the key elements of the information society roll out in the
EU. Sweden was expected to be one of the first European countries to make 3G services
available throughout the country and had opted providently for a joint collaborative plan that
suited the country best rather than auction to the highest bidder.

However, one of the four license holders in the country, Orange, has asked for a three year
delay in the original schedule that would have seen more than 99 per cent of the population
covered by the end of 2003. The new date by which Sweden should have the proposed 3G
coverage is the end of 2006.

Additionally, there has also been a request from Orange to reduce the level of coverage from
8.86 million people in Sweden to 8.3 million. The move is attributed to the economic
difficulties that have been experienced by telecommunications operators, who are now
looking to reduce spending in the 3G area. As well as the request made to the Swedish
telecoms authority for postponement, Orange has also announced that it will cut back
spending at its German venture, Mobilcom. The Spanish telecommunications company
Telefonica, has also said that it is pulling out of spending any money on non-Spanish 3G
developments.

54
Table 14: European 3G licences rollout schedules

EU Countries Licences Average cost of
licence(m)
Coverage obligations (share of adult
population)
Germany 6 8,468 25% by Dec 2003; 50% by 2005
UK 5 7,193 80% by end 2007
France 2 619 25% by Jul 2003;80% by 2009
Netherlands 5 537 60% by end 2007
Belgium 3 150 30% by Jul 2003; 85% by 2006
Italy 5 134 Regional capitals within 30 months
Spain 4 134 Cities > 250,000 citizens by Aug 2002
Denmark 4 128 30% by end 2004; 80% by end 2008
Austria 6 117 25% by Dec 2003; 50% by Dec 2005
Portugal 4 100 20% by end 2002; 60% by end 2007
Switzerland 4 34 20% by end 2002; 50% by end 2004
Norway 2 25 80% by end 2004
Sweden 4 nil* 99.98%by end 2004
Finland 4 nil* None
Source: Goldman Sachs, August 2002

However, one of the four license holders in the country, Orange, has asked for a three year
delay in the original schedule that would have seen more than 99 per cent of the population
covered by the end of 2003. The new date by which Sweden should have the proposed 3G
coverage is the end of 2006.

Additionally, there has also been a request from Orange to reduce the level of coverage from
8.86 million people in Sweden to 8.3 million. The move is attributed to the economic
difficulties that have been experienced by telecommunications operators, who are now
looking to reduce spending in the 3G area. As well as the request made to the Swedish
telecoms authority for postponement, Orange has also announced that it will cut back
spending at its German venture, Mobilcom. The Spanish telecommunications company
Telefonica, has also said that it is pulling out of spending any money on non-Spanish 3G
developments.

Although only one of four license holders in Sweden, its move could still also affect the
others, as the plans for establishing the 3G infrastructure in rural areas in the country is a
collaborative efforts between all four license holders. Thus this could have an effect in
Sweden as in other EU countries, of increasing the digital divide for 3G technologies.

Thus, should those seeking to develop indicators for the Information Society also delay the
measurement of these technologies and their impact? Existing technology is old. GSM's
underlying software platform was designed nearly 15 years ago. Its replacement while
delayed will be inevitable. Applications based on 3G will change consumer behaviour in a
similar manner to the way broadband Internet access changed narrowband Internet users.
3G allows a persistent connection to be established between a users device and a
networked information source. This 'always-on' functionality allows a deeper level of
interaction to be designed. For example, it allows the tracking of location and the provision of
location-sensitive information. If a suitable price regime is established it could also
encourage widespread ongoing use of digital content rather than sporadic connection to
sources. For all the above reasons it is important to track 3G.

Although the shift to 3G is a radical one, in most cases entailing the construction of new
networks, the transition from 2G to 2.5G services is likely to be accompanied by a more
significant conceptual shift than that from 2.5G to 3G, introducing the concept of always-on
55
mobile communications. This is already beginning to bring the Internet and mobile
communications onto a more common ground. The shift to 3G mobile will take this concept
even further, but will essentially involve convergence and interoperability. Although mobile
data services are already available on 2G platforms through WAP, i-mode and short
messaging service (SMS), it is through the advent of 2.5 and 3G that users will begin to fully
reap the benefits of the mobile Internet, through high-speed communications and multimedia
applications. What we have learned from the success and failures of 2G technologies is that
person-to-person messaging, simple interfaces and timely content delivery will be the key to
future service development and revenue generation. A mere simulation of the fixed-line
Internet experience will not compel users to take up mobile data services. The development
of an adequate payment system for mobile devices is also crucial: for always-on
applications, per-minute billing may have to give way to volume-based billing. As it has
happened with Internet subscription packages, there will be a need to bring flat-rate
connection packages for allowing consumers to fully exploit 3G services and content
possibilities.

The combination of mobile and Internet technologiesfor instance in the form of SMS
messagingis already transforming the way people interact and the way business is done.
Some 24 billion SMS messages were sent worldwide in the first quarter of 2002. Messaging
services have also brought information technology closer to groups that have traditionally
had limited access to it, such as children and the deaf community. High-speed data services
combined with additional functionality, such as location technologies and improved security,
will further enhance the user experience. On a technical level, the viability of future 3G
services will rely on continued efforts towards the interoperability of radio interfaces, the
evolution to an IP-based core network and the harmonisation of formats for content delivery.
At the service level, convergence between the fixed and mobile Internet is already
happening, through services such as mobile instant messaging and fixed-line SMS. This
interoperability will eventually encompass complementary and alternative network
technologies, such as wireless LANs, short-range connectivity technologies, fixed broadband
networks, etc. Regulators and industry players alike need to realise that there are a number
of different options for providing mobile Internet services, and that 3G services must be
considered in their global context.

In the short term, the key imperative for tracking 3G is the issue of the digital divide. While
absolute penetration levels will remain low for a number of years the geographic penetration
will vary significantly across Europe and within nations. 3G operators will target rollout in
populated areas where there is likely to be demand for roaming access to digital networks,
for example, in urban centres and travel locations such as airports and stations. The digital
divide will be starkly illustrated by this rollout pattern. In particular, it will heavily emphasise
the digital divide between rural and urban areas. It is ironic that while rural areas have been
traditionally denied access to broadband networks, because of the limitation of physical cable
networks, they will be denied the benefits of wireless technologies, which could overcome
those limitations. The emergence of digital deserts needs to be mapped. A practical
example occurs in Spain. Telefonica is restricting its 3G rollout to cities with more than
250,000 people. Thus, for instance Extremadura, a region with a capital city of 100,000
people, will be totally excluded from 3G benefits.

Need to track the deployment of alternative Broadband technologies where
infrastructures have not arrived
Generally, when looking at the penetration of different BB technologies, it is possible to
highlight how there is a divergence in the penetration and level of competition of BB
technologies in different countries. In fact by 2005, approximately 8.6 million potential
residential and SOHO users in Europe will still not have access to high-speed cable modem
or Digital Subscriber Line (DSL) services, according to analysis by McKinsey & Company.
56
Since there is a lack of infrastructures in deprived and isolated regions across the EU, it is
important to track what policy measures can be implemented to solve these issues. One of
the best European markets in terms of consumer's choice is the UK, where DSL and cable
technologies are widely available. Moreover, for rural areas where broadband land
infrastructures have not developed the government is starting to roll our broadband wireless
policies. Traditionally there have been some linked problems in the provision of BB by
satellite, since it was a technology not affordable for residential consumers and business
compared to existing competitive BB technologies such cable or DSL. There was a belief that
if ADSL roll-out is not achieved in rural areas, no competitive access arrangements will be
available for the foreseeable future.

However, this perspective has been challenged and can be questioned though different case
studies in regional areas which have shown how satellite can be initially expensive (as it
happened with ADSL in the beginning) but can be affordable as there is increasing take-up
and use from consumer/business groups. For instance some recent wireless successful
cases studies in 2002 include Northern Ireland and Wales.

6.1.1 The Imperative to measure subscription and connection rates

Past evidence indicates the importance of monitoring the cost of access to digital networks
and content. The popular adoption of the Internet in both Europe and the US did not pass the
10% threshold until acceptable price points and payment plans were in place. In the US, the
1996 move to flat-rate Internet access, so called 'eat-as-much-as-you-like' tariff structures,
spurred rapid adoption of the Internet. AOL's move from metered to flat rate pricing allowed it
to pass the 2 million users milepost. In Europe, adoption was limited by metered access to
not only online services, but also to the actual telephony network. Not until 2000, when flat
rate local call charges were popularly used by ISPs, did adoption start to accelerate
significantly.

Therefore, it can be expected that the existence of metered pricing structures associated with
new technologies, be it WAP, 3G or public access points, will hold back adoption of these.

Within the SIBIS GPS a question regarding cost of using Internet has not been included
since it has been noted by our experts that telephone surveys are not the best methodology
for doing that. Industry interviews would be best way of getting that information, but it is
highly complicated to do this due to the large number of ISPs and different types of
unmetered and metered types of offers across the EU. This is maybe one of the most difficult
areas to benchmark currently in Europe due to the broad diversity of subscription connection
packages to the internet in the different EU countries for both narrowband and broadband
technologies.

It is important to track how many experience narrowband users upgrade to broadband. As it
is necessary to track how Internet at home penetration rates evolve according to
narrowband/metered, narrowband/unmetered and broadband households. A comprehensive
trend portraying an example of this set of indicators is included below,

57
Table 15: Take-up of Internet consumer market narrowband (metered/unmetered) and
broadband
17

Aug 02 Nov 02 Feb03
UK homes connected to the Internet 42% 42% 45%
using narrowband
84% 80% 79%
using narrowband
(unmetered)
38% 35% 38%
using narrowband
(metered|)
29% 28% 25%
using narrowband
(unsure whether metered or
unmetered)
18% 16% 6%
using broadband
7% 9% 13%

Among the established eEurope priorities, broadband adoption is included at the top.
Broadband indicators relating to penetration and costs are very important, as broadband is
seen as gateway for the full development of e-commerce and other online services. For
those Internet users who just go online to access their e-mail and do a bit of web surfing, the
relationship between the cost of dial-up and broadband is important. For the millions of
Internet users in this category there is little economic incentive, as of late, to trade up to
broadband and pay significantly more.

However, a very positive aspect to highlight here is that Oftels
18
study, which included data
on consumer perceptions of broadband, portrayed how the majority of users who migrate to
broadband cannot contemplate returning to narrowband. The majority of users upgrading to
broadband are satisfied with the quality of service they receive in comparison to narrowband.

6.1.2 Further problems when tracking multi-platform and multi-context usage

Developments in broadband, and increasing penetration of alternative access devices, will
bring increasing difficulties when it comes to tracking usage across platforms and contexts.
As new networks are created, the open environment of the Internet must become a lessor
part of the digital universe. Encouraging owners and operators of the new private networks to
allow access for metering, and thus measuring, is vital.

Because PC-based online interaction with digital content is an established universe of
information possibilities it will remain the dominant digital medium. Therefore, measurement
of how alternative access technologies are used in conjunction with it is important. Finding
out what users do on their mobiles, PDAs and through their set-top boxes must be placed in
the context of the online PC environment, rather than being perceived a 'battle of the
platforms'.

17
Oftel, Monitoring Internet studies, March, 2003.
18
Survey on consumer perception studies of broadband, 2002
58
Multi-context usage is more difficult to measure. Until there is majority adoption of digital
wallets, IDs or other user authentication technologies, this will remain a difficult to measure
as authoritatively identifying the user will be much more difficult than identifying a particular
access device.

6.2 Some methodological problems when collecting information on
T&A with telephone survey methodologies
6.2.1 RDD telephone survey

Random digit dial telephone surveys use an application that randomly dials telephone
numbers to ensure that the sample of respondents is truly random and or statistically
representative. This method does, however, have problems. Market research industry codes
of conduct forbid surveyors from interviewing children under 16 without consent of their
parents. This can skew the responses to questions related to children's topics. In the topic in
question, young users are a relevant group for of those interested in ICT to track. In the case
of SIBIS, many users of games consoles will be under 16 so while they may be using such
devices to access the Internet they are unlikely to be surveyed about this if contacted by
telephone. Likewise, the multi-context access (question GPS A9) could not track users
accessing the Internet from primary and secondary education institutions. Those younger
than 16 years old were not monitored in this study, despite many of them being early and fast
adopters of new technologies. Another point to be made is that if children under the age of
16 could have been included in some way, it would have been possible to track some of the
e-Learning action plans objectives of promoting European Youth in a Digital Age. This is
something which must be consider when developing indicators in future work. Alternative
methodologies to telephone surveys must be applied in order to track children and teenagers
use of ICTs.

For instance, specialised children's panels can be assembled in order to solicit the opinions
of children.

6.2.2 Some final methodological questions related to e-content gaps: Service and
applications usage

Although SIBIS is not looking at developing e-content indicators, it is important to remark at
this stage some indicators which could be added to easily track some indicators on e-
content.

While it is possible to identify the top sites in terms of traffic, visitors to a site, simple metrics
tell only part of the story. Time spent per visits is important, as is the number of return visits.
Consumers have the capacity to prioritise and favour just a handful of sites. Such sites, that
are subject to regular visitation and deep usage, do not have exclusive mindshare but their
consistent usage makes them useful indicators of usage trends. Understanding which ones
are patronised by particular groups is important. Usage intensity can be calculated by
relating the time spent per visit to the frequency of return visit. However, it is extremely
important that the nature of a site is understood when interpreting usage intensity data. A tax
return site that is receiving frequent short visits may be fulfilling an education role by reaching
lots of people. But, if the online tax return form within it takes 30 minutes to complete and
only 5% of visitors are spending 30 minutes plus then that service is not being used properly.
59

Gathering such metrics can be done using panels that deploy metering applications to
participating users' PCs, but the restricted sample size, measured in the thousands, means
that only very popular sites can be analysed. At-work usage metrics cannot be studied and
compared consistently across Europe yet as only a few such panels exist in certain markets.
If a site has less than 5% of total traffic in a sector the resulting data may not be reliable. To
analyse less popular sites in this ways requires the aggregation of a much larger sample.
Using surveys that place the meter at ISP servers can allow this to happen and allows
reliability to extend to data gather from sites with 1% traffic level.

With regards to browsers and applications, SIBIS only concentrates on piloting email
services usage in the GPS. Besides email, in future work it would be useful to track other
popular applications that are widely used such as peer-to-peer applications or instant
messengers. Likewise, tracking killer applications for emerging technologies, such as
broadband or WAP is important since its absence it is one of the commonly mentioned
inhibitors for not adapting to those technologies. Hence, it seems particularly important to
track users favourite content services in relation to broadband penetration and mobile-PC
bimodal usage developments. For instance, the EITO study currently includes an indicator
that records the three most widely used Internet applications.

60
7 Conclusions

Telecommunications and access is a topic area which underpins the advancement of Europe
in the Information Society. The study itself helps to confirm a number of existing assumptions
and explore differences across the EU and the US in access, usage and impact as well as
giving an opportunity to test a set of new indicators on a pan-European level. The work also
contributes to measuring the progress in adopting the New Regulatory Framework in
Electronic communications, networks and services
19
by tracking the penetration of
broadband across Europe.

From an analysis of the data from the GPS and from the compound indicators a number of
differences have been identified across the participating European Countries and between
Europe and the US.

Analysing how Europe is doing in the area of telecommunications and access is complex. It
is clear from the data that one size does not fit all in attempting to measure the access,
usage and impact across the Member States and the other participant countries. These
country comparisons will be explored further once the second phase of SIBIS, SIBIS+, which
will include NAS countries is implemented, giving EU25 data.

The 'Migrators to faster connections than dial-up snapshot' has portrayed how two factors
have influenced when Internet users migrate to a faster connection. Based on the
experiences of the US and Scandinavian markets, it has been noted that in the majority of
active at-home online population, there is a migration of users with tenure, commonly defined
as those with two years or more Internet experience, to faster connections. They seek a
better online experience e.g. quicker downloads, always on connections.

As well as the compound indicators, across the other indicators tested in the SIBIS survey,
there is evidence of national differences and great disparities across Europe. There are
some large differences in Internet usage between Northern European Countries and
Mediterranean countries but similar adoption rates, and much higher mobile penetration and
mobile data usage, particularly SMS.

A number of methodological lessons have also been learned as a consequence of this study.

It remains important to go on counting instances
It remains extremely important, in the light of these results, to go on counting instances in
some of the European countries involved in the survey. However, the next step of
measurement is where SIBIS adds value as there are numerous pan-European surveys
already in existence which are counting, SIBIS has the opportunity to suggest sets of
indicators which help provide a picture of the how and why? In spite of this need, it must be
noted that identifying usage patterns and impacts is much more difficult. Usage must first be
observed over a time period and then needs to be classified. These classifications then need
to be reapplied to the observed data. The evolution of the usage behaviours also means that
the classifications have to be reassessed regularly to make sure that they are not out of date.
As user numbers increase, so does the depth of content available to them.

This is an exciting area of development, which can give great insight into the use of
technologies in Europe, especially given that the second phase of SIBIS will be looking at

19
COM(2000)239
61
these indicators over the NAS countries, providing one of the first comprehensive pictures of
the state of telecomms and access indicators in the EU25.

A number of flexible indicators could be chosen and used to test the levels of
penetration of technologies in each country before piloting sophisticated indicators
The differences across the Member States and Switzerland are high and vary for each
technology area. This means that the current approach of asking exactly the same set of
questions in each country leads to some country data, particularly in the cases of usage and
impact indicators, being statistically insignificant. A way of creating surveys that are
comparable, but do not ask obsolete questions, is to devise a system of standard flexible
indicators which could be drawn on to create a set which best work in a country specific
situation. A set of penetration indicators could also be identified which would be asked
across the board in order to judge when some countries start to have significant samples of a
chosen technology in use so that more sophisticated indicators could be introduced. In fact,
one of the main conclusion embodied in SIBIS is that it is more important given the current
technological disparities in penetrations and prices in Europe to measure experience rather
than technology.

As highlighted in SIBIS research, the shortening of technology product life cycles makes any
tracking measurement problematic. This problem is compounded by users definitions and
perceptions of technology that vary across countries. Therefore, over the medium and long
term, measuring experience, measuring consumers satisfaction levels seems mandatory on
a pan-European level.

There is much more work that could be done. There are still many gaps that have been
identified in this report and also there is much that can still be done with the results that have
already come out of the survey. Some future work in SIBIS+ will look at comparing some of
the indicators proposed in this report on an EU25 basis. This will make SIBIS one of the most
comprehensive resources in the development of IS indicators and benchmarking
comparisons in Europe.

62
8 ANNEX 1 References

EITO - European Information Technologies Observatory (2001-2002).

European Commission (2000), Measuring Information Society. A Eurobarometer survey
carried out by INRA (Europe).

European Commission (2001), The Digital Divide - A Research Perspective. Report to the
G8 Digital Opportunities Task Force.

Eurostat (2001), Telecoms Outlook. 2002 European Commission, Brussels.

Goldman Sachs, August 2002, Mobile notices, European 3G rollouts

ITU-Internet for a mobile Generation, 2002

Nielsen/Netratings Broadband Study Q2 2002

OECD (2001), Understanding the Digital Divide. OECD, Paris.

OECD (2002), ICT outlook, OECD, Paris.

Oftel -Consumer perceptions of broadband services and International Benchmarking, 2002

Ovum, Consulting Insight: Broadband Advisor: 'Delivering on the EU Commission broadband
vision', April 2003,

63
9 ANNEX 2 - Methodology of the survey
9.1 General Population Survey (GPS)
9.1.1 Outline of the study

The survey was conducted in April-May 2002 in all 15 EU Member States plus Switzerland
and the USA, using computer-aided telephone interviews. The survey was co-ordinated and
executed by INRA, Germany. The population for this study was all persons aged 15 and
over living in private households in the respective countries and speaking the respective
national language(s). Subjects discussed included ownership and use of ICT equipment,
use of the Internet and e-commerce activities, competence in the use of new media,
questions on health and the Internet, the Internet and security concerns, e-government,
telework, mobile work and other new ways of working, as well as further education and
satisfaction with working conditions. 11,832 interviews were successfully completed. The
average interview length per country varied between 10 and 20 minutes.

9.1.2 Methodology

Subject of study Topics of this survey were statements on interviewees ownership and use
of ICT equipment, use of the Internet and e-commerce activities,
competence in the use of new media, questions on health and the Internet,
the Internet and security concerns, e-government, telework, mobile work
forms, as well as further education and satisfaction with working conditions.
Study concept The study was conceived and executed as a cross national study. The co-
ordination was carried out by INRA Deutschland GmbH, Mlln, on behalf of
the client. The study consisted of two parts, a trial in Germany with a
subsequent main survey in all participating countries.
Overall responsibility
and co-ordination
INRA Deutschland GmbH, Mlln
Countries and
executing institutes
Belgium: INRA Belgium
Kroonlaan 159-165 Avenue de la Couronne
1050 Brussels
Denmark: Gallup A/S
Sundkrogsgade 10
2100 Copenhagen
Germany: INRA Germany GmbH
Papenkamp 2-6
23879 Mlln
Finland: Taloustutkimus Oy
Lemuntie 9
00510 Helsinki
France: BVA
B.P. 59
78222 Viroflay Cedex
Greece: MEMRB K.E.M.E
24 Ippodamou St.
11635 Athens
Great Britain: BMRB International
Saunders House, 53 The Mall, Ealing
London W5 3TE
Ireland: Lansdowne Market Research Ltd.
49 St. Stephens Green
Dublin 2
Italy: INRA Demoskopea
64
Via Salaria, 290; Via Rubicone 41
00199 Roma
Luxembourg: ILReS. S.A.
46, Rue du Cimetire
1338 Luxembourg / Bonnevoie
Netherlands: NIPO
Grote Bickersstraat 74
1013 ks Amsterdam
Austria: Spectra
Brucknerstr. 3-4/5
4020 Linz
Portugal: METRIS
Av. Eng. Arantes e Oliviera, No. 3-2
1900 Lisboa
Sweden: GfK Sverige AB
Box 401
22100 Lund
Switzerland: Link Institut
Spannortstrasse 7/9
6000 Luzern
Spain: INRA Espaa S.A.
Calle Alberto Aguilera 7-5
28015 Madrid
USA: I.C.R
605 West Street
Media, Pennsylvania 19063-2620
Survey methodology The study was carried out as a telephone survey (Computer Assisted
Telephone Interview C.A.T.I) in all countries.
Population The population for this study is all persons aged 15 and over living in
private households in the respective countries and speaking the respective
national language(s).
Switzerland: Here the survey was carried out in both the German and
French speaking parts of Switzerland.
USA: The population includes English speaking people in the 48
continental federal states of the USA (excluding Alaska and Hawaii).
Finland: Finnish speaking population.
Random sampling and
selection process
Belgium: 3-stage selection process based on the INFO BEL telephone
directory. Addresses in 10-fold translation, random sampling of
households, selection of the target person via a birthday key.
Denmark: Geographically stratified sample based on telephone
directories. Telephone numbers are generated so as to also include
unlisted numbers. The selection of the target person results from a
birthday key.
Germany: Within the ADM telephone sampling system a representative,
multistage random sample is drawn for each survey area. The selection
data is based on the batch of all registered fixed network telephone
numbers. Master numbers are formed by detaching the two final digits.
Through the generation of new final digits from 00 to 99 number blocks are
produced which contain listed as well as non listed numbers. As far as
possible, business numbers were removed from this sampling frame. All
telephone numbers were given an area code number, either the original or
according to the known distribution in the number block. In this way
numbers can be sorted regionally, thus increasing the precision of the
sample. By means of the relation between sample size and distribution of
households per regional cell an allocation table is produced which
determines the number of samples to be drawn for each cell. Assuming a
random starting point, all areas are processed with a fixed step width
through set stages until the number to be selected from which areas for
each cell has been determined. Subsequently the numbers are drawn
randomly in a second selection stage. The selection of the target person
results from a birthday key.
Finland: Geographically and socio-demographicaly stratified random
65
sample based on information from official statistics. The selection of target
households takes place at random. The selection of target persons results
from a birthday key.
France: Geographically and socio-demographically stratified random
sample of 8000 starter addresses based on France Telecom directories.
The selection of target persons results from a birthday key.
Greece: Multistage stratified random sampling. The geographical
stratification takes place on the basis of NUTS 1, NUTS 2 and location
size. The size of each unit is determined on the basis of official statistics.
The selection of the target person results from a birthday key.
Great Britain: The sample is based on a draw data-file. Through the
generation of new final digits a sample frame is established which contains
listed and unlisted as well as so far non-existent numbers. The selection of
target households takes place at random. The selection of the target
person is via quota.
Ireland: Geographically sorted random sample based on the Eircom
telephone directory. Additional telephone numbers are also generated in
order to include unlisted numbers. The selection of the target person
results from a birthday key.
Italy: Geographically and socio-demographically stratified random sample.
705 sample points result from the stratification. Additional telephone
numbers are generated in order to also include unlisted numbers. The
selection of the household results from Random Digit Dialling and the
selection of the target person results from a birthday key.
Luxembourg: The sample is based on a draw data-file. Through the
generation of new final digits from 00 to 99 a sample frame is established
which contains listed and unlisted as well as so far non-existent numbers.
The selection of target households takes place at random. The selection of
the target person results from a birthday key.
Netherlands: Geographically stratified random sample. The geographical
sorting is based on postcode areas. Target person selection takes place
through an algorithm, which selects the interviewee on the basis of age
and gender of people living in the household.
Austria: Geographically stratified random sample. The selection of the
target household takes place through RDD (Random Digit Dialling). The
selection of the target person results from a birthday key.
Portugal: Geographically and socio-demographically stratified random
sample. 200 sample points result from the stratification. The selection of
households takes place via Random Digit Dialling, the selection of target
persons via a birthday key.
Sweden: Geographically and socio-demographically stratified random
sample. 200 sample points result from the stratification. The selection of
households takes place via Random Digit Dialling, the selection of target
persons via a birthday key.
Switzerland: Geographically stratified random sample based on
postcodes. Each postcode represents a sample cell. The selection of
households takes place via Random Digit Dialling and the selection of the
target person via an algorithm which selects the interviewee at random on
the basis of a list of household members.
Spain: Geographically stratified random sample based on NUTS2 areas.
148 randomly selected sample points result from the sorting. Within these
sample points addresses of target households are randomly drawn.
Selection of target persons results from a birthday key. After about two
thirds of the fieldwork the screening was targeted towards male members
of the household due to a disproportionate number of female interviewees.
USA: Geographically stratified random sample based on the MSG-
Genesys sampling process. The selection of households takes place via
Random Digit Dialling and the selection of the target persons via a birthday
66
key. After the 758th interview the screening was targeted towards male
members of the household due to a disproportionate number of female
interviewees.
Survey period The interviews were carried out in the following period:
04.03. -18.05.2002
Interviews undertaken Total: 11,832
Belgium 16.0 min Luxembourg 16.2 min
Denmark 18.1 min Netherlands 18.4 min
Germany 17.5 min Austria 15.8 min
Finland 17.3 min Portugal 12.1 min
France 12.0 min Sweden 20.2 min
Greece 10.2 min Switzerland 19.0 min
Great Britain 18.0 min Spain 12.5 min
Ireland 17.7 min USA 18.3 min
Average interview
length
Italy 14.0 min
Interviewers used Total: 632
Additional comments
to the data set
Belgium: In order to improve the sample, an additional 85 interviews were
carried out in some cells.
Finland: In order to improve the sample, an additional 169 interviews were
carried out in some cells.
Netherlands: In order to improve the sample, an additional 30 interviews
were carried out in some cells.
Switzerland: In Switzerland respondents were not asked to deduct tax
from income (Z19), as that is not the norm there.
Data supply One labelled SPSS data set of the main survey of all interviews.

67
Field report and outcomes
B DK D FIN F EL UK IRL I L NL AT P S CH E USA
Method C.A.T.I.
1 gross sample (utilised addresses) 4506 3154 9999 2621 7300 5022 11392 3890 12006 8764 3640 4669 1403 5177 2327 6494 18162
1.1. non-contacts thereof: 311 242 1701 40 3401 2346 139 1111 4436 5023 803 193 91 455 638 1239 4192
1.1.1 unobtainable 0 235 1202 0 2342 2077 123 654 4436 3748 522 124 43 113 638 644 3656
1.1.2 engaged 3 7 436 0 57 206 1 316 0 705 164 8 32 55 0 5 536
1.1.3 answer phone, fax, modem 308 0 63 40 1002 63 15 141 0 570 117 61 16 287 0 590 0
1.1.4 other 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1.2 sample neutral non-response thereof: 1874 1917 4492 984 511 1022 5088 1051 2659 1316 805 2322 410 2808 322 1095 8789
1.2.1 invalid telephone numbers 955 1516 3760 97 60 529 4308 498 1657 790 652 858 334 2297 230 398 5725
1.2.2 not in the population 472 202 41 782 374 176 119 405 364 0 153 1248 47 16 0 164 478
1.2.3 business numbers 300 82 285 12 27 220 437 0 340 455 0 75 15 193 0 434 1331
1.2.4 other 147 117 406 93 50 97 224 148 298 71 0 141 14 302 92 99 1255
2 net sample thereof: 2321 995 3806 1597 3388 1654 6165 1728 4911 2425 2032 2154 902 1914 1367 4160 5181
2.1 refusal 1470 468 2451 912 2231 747 5012 1134 3592 1000 1248 1609 364 1246 529 2255 3198
2.2 termination 114 0 87 0 30 0 80 11 201 0 0 1 6 19 0 115 143
2.3 target person contacted but interview
impossible thereof:
152 26 267 16 127 402 73 83 118 925 254 44 32 146 316 775 836
2.3.1 possible appointment outside field time 0 23 14 1 23 9 26 14 106 763 208 7 6 30 80 321 156
2.3.2 appointments to continue interview outside
field time
152 0 200 0 104 295 47 65 12 17 11 34 18 24 194 179 669
2.3.3 other 0 3 53 15 0 98 0 4 0 145 35 3 8 92 42 275 11
2.4 complete interviews 585 501 1001 669 1000 505 1000 500 1000 500 530 500 500 503 522 1015 1004
3 exhaustion rate (%) (2.4/(2.1+2.2+2.4)) 27.0% 51.7% 28.3% 42.3% 30.7% 40.3% 16.4% 30.4% 20.9% 33.3% 29.8% 23.7% 57.5% 28.5% 49.7% 30.0% 23.1%

68
9.1.3 Weighting
1. Transformation from household sample to person sample:
As only one person per household is interviewed, the described sample procedure provides
a household sample, i.e. each household of the base population has the same likelihood of
being in the sample but not each person. With the weighting stage of the transformation the
equal likelihood of households is replaced mathematically by the equal likelihood of the
individuals. To this end, each data set is multiplied by the amount of people in the household
aged 15 or over. This number is subsequently divided by the average household size in
order to obtain the actual case number.

2. Adjustment of unweighted sample structure to the official statistic:
Because random samples are not evenly distributed across all population strata, the
distribution of unweighted samples regularly and systematically deviate from the population
distribution from official statistics. Through the mathematical weighting the sample
distribution is adjusted to the official statistics. The national weighting factor (P10) which
results from the iterative weighting was included in the data material. To this end the
following criteria are used in the respective countries.

Austria: age, gender, region; Belgium: age, gender, region, locality size; Denmark: age,
gender, region; Germany: age, gender, region, locality size; Greece: age, gender, locality
size; Finland: age, gender, region; France: age, gender, region, locality size; Ireland: age,
gender, region; Italy: age, gender, region, locality size; Luxembourg: age, gender, region,
locality size; Netherlands: age, gender, region; Portugal: age, gender, region, locality size;
Sweden: age, gender, region; Switzerland: age, gender, region; Spain: age, gender, region,
locality size; UK: age, gender, region; USA: age, gender, region, locality size.

3. Adjustment of weighted sample structure to the EU15-member states population:
This weighting factor was necessary to calculate total figures according to the whole
population of the European Union member states. Furthermore it is useful to compare the
EU with the US. Population sizes of each member state are weighted to reduce the
distortion based on the sample sizes in each country. The different country-specific
weighting factors are the following:

Austria 0.44 Italy 1.63
Belgium 0.48 Luxembourg 0.02
Denmark 0.29 Netherlands 0.80
Germany 2.29 Portugal 0.55
Greece 0.59 Spain 1.09
Finland 0.21 Sweden 0.48
France 1.56 United Kingdom 1.57
Ireland 0.20
Switzerland,
USA
none

69
9.1.4 Sample characteristics and effect of weighting

Total EU15

u
n
-
w
e
i
g
h
t
e
d

w
e
i
g
h
t
e
d

%

u
n
-
w
e
i
g
h
t
e
d

%

w
e
i
g
h
t
e
d

u
n
-
w
e
i
g
h
t
e
d

w
e
i
g
h
t
e
d

%

u
n
-
w
e
i
g
h
t
e
d

%

w
e
i
g
h
t
e
d

Total sample 11832 11832 100.0 100.0 11832 10306 100.0 100.0
Country
Austria 500 500 4.2 4.2 - - - -
Belgium 585 585 4.9 4.9 - - - -
Denmark 501 501 4.2 4.2 - - - -
Finland 669 669 5.7 5.7 - - - -
France 1000 1000 8.5 8.5 - - - -
Germany 1001 1001 8.5 8.5 - - - -
Greece 505 505 4.3 4.3 - - - -
Ireland 500 500 4.2 4.2 - - - -
Italy 1000 1000 8.5 8.5 - - - -
Luxembourg 500 500 4.2 4.2 - - - -
Netherlands 530 530 4.5 4.5 - - - -
Portugal 500 500 4.2 4.2 - - - -
Spain 1015 1015 8.6 8.6 - - - -
Sweden 500 500 4.2 4.2 - - - -
Switzerland 522 522 4.4 4.4 - - - -
UK 1000 1000 8.5 8.5 - - - -
USA 1004 1004 8.5 8.5 - - - -
EU15 - - - - 10306 10306 87.1 100.0
Age groups
up to 24 1964 2019 16.6 17.1 1731 1651 16.8 16.0
25 to 49 5511 5309 46.6 44.9 4817 4593 46.7 44.6
50 to 64 2515 2495 21.3 21.1 2191 2209 21.3 21.4
65 and more 1833 2000 15.5 16.9 1558 1839 15.1 17.8
dont know 9 9 0.1 0.1 9 14 0.1 0.1
Terminal education age
up to 13 695 717 5.9 6.1 693 728 6.7 7.1
14 715 742 6.0 6.3 701 881 6.8 8.5
15 to16 1794 1750 15.2 14.8 1641 1820 15.9 17.7
17 to 20 3587 3515 30.3 29.7 2997 2937 29.1 28.5
21 and more 3266 3275 27.6 27.7 2743 2495 26.5 24.2
still studying 1687 1751 14.3 14.8 1463 1372 14.2 13.3
dont know 88 81 0.7 0.7 77 73 0.7 0.7
Household type
one person household 2006 1611 17.0 13.6 1682 1408 16.3 13.7
household with kids aged under 6 1723 1754 14.6 14.8 1451 1440 14.1 14.0
household with kids aged 6+ 2970 3152 25.1 26.6 2653 2655 25.7 25.8
two person household without
kids
5063 5240 42.8 44.3 4467 4768 43.3 46.3
no answer on household size 70 75 0.6 0.6 53 35 0.5 0.3
Household income (according to national household income quartiles by Eurobarometer)
First quartile (lowest income) 1774 1580 15.0 13.4 1548 1299 15.0 12.6
Second quartile 2132 2084 18.0 17.6 1878 1764 18.2 17.1
Third quartile 2536 2521 21.4 21.3 2214 2087 21.5 20.3
Fourth quartile (highest income) 2968 3102 25.1 26.2 2502 2725 24.3 26.4
dont know 1214 1295 10.3 10.9 993 995 9.6 9.7
refusal 1208 1249 10.2 10.6 1171 1436 11.4 13.9
Employment status
paid employment 4966 4853 42.0 41.0 4291 4133 41.6 40.1
self-employed 935 941 7.9 8.0 809 799 7.8 7.8
70
Total EU15

u
n
-
w
e
i
g
h
t
e
d

w
e
i
g
h
t
e
d

%

u
n
-
w
e
i
g
h
t
e
d

%

w
e
i
g
h
t
e
d

u
n
-
w
e
i
g
h
t
e
d

w
e
i
g
h
t
e
d

%

u
n
-
w
e
i
g
h
t
e
d

%

w
e
i
g
h
t
e
d

unemployed/ temporarily not
working
701 683 5.9 5.8 621 631 6.0 6.1
in education 1687 1751 14.3 14.8 1463 1372 14.2 13.3
retired or other not working 3441 3510 29.1 29.7 3034 3292 29.4 31.9
dont know 102 94 0.9 0.8 88 80 0.9 0.8
Social grade (ESOMAR classification)
unskilled manual workers and
other less well educated workers/
employees
1332 1318 11.3 11.1 1238 1323 12.0 12.8
skilled workers and non-manual
employees
1525 1445 12.9 12.2 1316 1287 12.8 12.5
well educated non-manual and
skilled workers
1434 1402 12.1 11.8 1254 1121 12.2 10.9
managers and professionals 1577 1586 13.3 13.4 1265 1167 12.3 11.3
not specified 5964 6081 50.4 51.4 5233 5408 50.8 52.5

9.2 Decision Maker Survey (DMS)
9.2.1 Outline of the study

The survey was conducted in March-May 2002 in seven EU Member States using computer-
aided telephone interviews. The survey was co-ordinated and executed by INRA, Germany.
The population for this study is defined as all establishments belonging to four aggregated
industry sectors in the seven Member States. The interview was conducted with IT
responsible persons in companies across all sectors of the economy. Subjects discussed
included ownership and use of ICT equipment, use of the Internet and e-commerce and e-
business activities, e-business security, e-government, web-site accessibility and ICT in
research and development. 3,139 interviews were successfully completed. The average
interview length per country varied between 14 and 18 minutes.
9.2.2 Methodology

Subject of study
Basic ICT: use of ICT and e-business technologies
e-commerce
e-business security
e-government
web-site accessibility
research and development
establishment demography
Study concept
The study was conceived and executed as a cross national study. The co-
ordination was carried out by INRA Deutschland GmbH. The study
consisted of 2 parts, a trial in Germany with a subsequent main survey in
all participating countries.
Overall
responsibility and
co-ordination
INRA Deutschland GmbH, Mlln
Countries and
executing institutes
Germany: INRA Germany GmbH
Papenkamp 2-6
23879 Mlln
71
Finland: Taloustutkimus Oy
Lemuntie 9
00510 Helsinki
France: BVA
101 avenue du General Leclerc
78222 Viroflay Cedex
Greece: MEMRB K.E.M.E
24 Ippodamou St.
11635 Athens
Survey methodology The study was carried out as a telephone survey (Computer Assisted
Telephone Interview C.A.T.I) in all countries.
Population
The population for this study are establishments (in each respective
country) in the four sectors:
Manufacturing, Construction, Primary Sector
Distribution, Catering, Transport & Communication
Financial & Business Services
Public administration, education, health, other personal and social
services
Target person at the establishment was the person who is responsible for
or significantly involved in decisions in the area of IT/ DP. In larger
establishments/ organisations the head or another executive of the IT/ DP
department. In smaller establishments/ organisations also the owner/
proprietor or managing director/ board member.
Random sampling
and selection
process
General: The sample was set up according to given industry and size
class quota. Accordingly a strat ified random sample was drawn from the
universe, allowing for the relevant industries within the four aggregated
sectors. Drawing the sample was organised locally by the national
executing institutes.
Germany: The sample was drawn from the Heins und Part ner Business
Pool. Heins und Partner have created a high quality business pool based
on the available address inventories consisting of about 3.4 m data sets
that have undergone comprehensive validation. For every enterprise
comprehensive additional information is available, including corporate
structure and branch office structure (220,000 branch offices) and is
continuously being updated. The sample was drawn from the
establishment file, which results from the transformation of enterprises into
establishments and appending branch offices to the headquarters.
Finland: The sample was taken from the so called "Blue Book -
Salesleads database" which is edited by Helsinki Media Company Oy
(Sanoma Magasines Finland). This data base contains of about 170,000
data sets and is being updated every two months.
France: The sample was drawn from the INSEE Siren file (the national
office of statistics). INSEE, as a public organisation, is responsible for
gathering all economic and social data in France. These data sets are
being updated every two months.
Greece: The sample was drawn from the address inventory of ICAP
(major establishment data base for Greece and member of the European
Association of Directory and database Publishers). The data base is being
updated every 18 months and also contains public sector addresses.
Additionally, public sector addresses were taken from the national
telephone inventory.
72
Great Britain: The sample was drawn from "BTs Business Database".
This is a representative data base of all establishments in the UK having a
telephone number (including addresses by BT, Mercury, cable and about
92 further telecom carriers). The data base consists of about 1.6 m
addresses and is being updated every two months.
Italy: The sample was drawn from Dun & Bradstreet's data base. This
data base is considered to be the most reliable source for Italy.
Spain: The sample was drawn from Schober's data base. This data base
is the most voluminous record as regards number of establishments for
Spain.
Survey period The interviews were carried out in the following period:
21.03. -15.05.2002
Interviews
undertaken
Total: 3,139
Germany 16.0 min Great Britain 16.2 min
Finland 16.4 min Italy 18.2 min
France 14.1 min Spain 16.4 min
Average interview
length
Greece 15.1 min
Interviewers used Total: 212
Quality control All interviewers were instructed about the questionnaire before the
beginning of field work. Field work was permanently controlled by
supervisors. Because of computer aided realisation of interviews
systematic errors of data gathering can be excluded. Furthermore the
proper realisation of interviews was monitored according to institute
standards. Following is the share of interviews monitored (by telephone):
Germany: 10%
Finland: 5%
France: 10%
Greece: 20%
Great Britain: 8%
Italy: 10%
Spain: 30%
Additional
comments to the
data set
Question A8: Turnover indication in national currencies were translated in
Euro except for UK.
Some indications seem to be very high, others very low. National institutes
have re-examined and verified statements by calling the respondents again
and reassured turnover answers were as stated.
Problems
encountered
In all countries fulfilling the quota for the largest establishments was difficult
(mainly 500+ / partly also 200-499 employees). In these establishments on
the one hand the availability of target persons is significantly lower, on the
other hand are these target persons "over-researched (which in part
results in a general interdiction to take part in surveys).
Due to this in France it was necessary to adapt the quota in order to
achieve the number interviews aimed at (i.e. interviews - as far as possible
- carried out in establishments of the next smaller size class).
Data supply
One labelled SPSS-data set of the main survey of all interviews.

73
Field report and outcomes

D FIN F EL UK I E
1 Sample (gross), i.e. number dialled at least once 4917 1923 8061 1728 8726 10846 8489
1.1 Telephone number does not exist 787 47 598 43 416 1160 808
1.2 Not an establishment (i.e. private household, etc.) 46 15 0 2 0 0 235
1.3 Fax machine/ Modem 81 0 152 31 0 0 519
1.4 Quota completed, therefore address not used 0 849 1599 2 2659 848 1397
1.5 No target person in establishment 858 226 1261 35 1766 822 2043
1.6 Language problems 0 15 0 0 0 0 10
1.7 SUM (1.1+1.2+1.3+1.4+1.5+1.6) 1753 1152 3610 113 4841 2830 5012
2 Net sample (1 minus 1.7) 3164 771 4451 1615 3885 8016 3477
2.1
Nobody picks up phone (and max. contacts not yet
exhausted)
325 2 326 229 32 804 18
2.2 Line busy, engaged 45 0 31 235 2 1852 9
2.3 Answering machine 111 4 82 15 0 0 482
2.4
Contact person refuses (i.e. refusal at reception,
switchboard)
436 228 912 38 1354 1056 1022
2.5 Target person refuses 1044 204 1569 107 1672 1410 896
2.6 no appointment during fieldwork period possible 33 14 356 36 176 680 203
2.7 open appointment 604 4 642 644 52 1668 111
2.8 target person is ill/ cannot follow the interview 1 3 18 0 0 0 18
2.9 Interview abandoned 53 1 14 4 97 34 102
2.10 Interview error, cannot be used 0 5 0 6 0 0 109
2.11 SUM (2.1+2.2+2.3+2.4+2.5+2.6+2.7+2.8+2.9+2.10) 2652 465 3950 1314 3385 7504 2970
2.12 SUCCESSFUL INTERVIEWS 512 306 501 301 500 512 507
3
Completion Rate
(2.12 / (2.11+2.12)), in %
16.18 39.69 11.25 18.63 12.87 6.38 14.58

Target and actual numbers of interviews
F D I E UK FIN EL
Quota Group required - achieved - required - achieved -
1 - 9 30 33 30 34 33 32 18 18 17
10 - 49 35 36 36 37 35 35 21 21 22
50 - 199 35 38 37 40 35 35 21 21 25
200 - 499 40 44 41 43 41 40 24 28 22
500+ 15 9 14 13 15 15 9 9 6
I
Manufacturing,
construction, primary
sector
Sum 155 160 158 167 159 157 93 97 92
1 - 9 45 50 47 45 46 45 27 28 27
10 - 49 40 42 41 41 43 40 24 24 25
50 - 199 30 28 31 26 30 30 18 18 18
200 - 499 15 19 15 16 15 15 9 5 9
500+ 10 5 10 8 10 10 6 5 6
II
Distribution, catering,
transport and
communication
Sum 140 144 144 136 144 140 84 80 85
1 - 9 30 32 30 34 30 30 18 16 17
10 - 49 20 19 21 23 21 20 12 14 11
50 - 199 10 13 10 17 10 10 6 6 8
200 - 499 10 13 10 6 10 10 6 7 6
500+ 10 8 9 4 7 8 6 6 6
III
Financial and business
services
Sum 80 85 80 84 78 78 48 49 48
1 - 9 20 20 24 19 20 20 12 13 13
10 - 49 25 29 25 26 25 25 15 16 16
50 - 199 30 22 30 34 30 30 18 18 18
200 - 499 35 32 35 31 35 35 21 23 20
500+ 15 9 16 15 16 15 9 10 9
IV
Public administration,
education, health, other
personal and social
services
Sum 125 112 130 125 126 125 75 80 76

Total 500 501 512 512 507 500 300 306 301

9.2.3 Weighting

For the SIBIS DMS a sample stratified by sector/ size cells was used which ensured that in
each sector, establishments from all size classes were sampled. In order to be able to raise
figures to national level, some form of weighting is required which adequately reflects the
structure and distribution of establishments (or related variables) in the universe of the
74
respective country (and, by implication, EU15). All presentation of SIBIS results indicates
clearly which of these weighting schemes was used.

9.2.3.1 Original weight

Within each country, the interviews were split according to a quota plan which guaranteed
that the sample is not dominated by micro and small companies. The quotas roughly reflect
the distribution of employment over sector and establishment size bands in the EU, and
derive from research into establishment sampling frames undertaken for previous studies by
Infratest and GfK in the course of ECaTT. They represent best estimates, but do not take
account of country differences.

The quota scheme looks as follows:

empirica SUGGESTED QUOTAS: Sectors (aggregated) X Size
1- 9 10 - 49 50 - 199 200 - 499 500+ Total
% of
total abs
% of
total
% of
total
% of
total
% of
total
% of
total
Quota I 6% 30 7% 35 7% 35 8% 40 3% 15 31% 155
1 Mining, Energy
2 Manufacturing
3 Construction
Quota II 9% 45 8% 40 6% 30 3% 15 2% 10 28% 140
4 Distribution
5 Hotels, Restaurants
6 Transport, Communication
Quota III 6% 30 4% 20 2% 10 2% 10 2% 10 16% 80
7 Banking, Insurance
8 Business Services
Quota IV 4% 20 5% 25 6% 30 7% 35 3% 15 25% 125
9 Public Administration
10 Education
11 Health and Social Work
12
Other personal or social
services
Total 25% 125 24% 120 21% 105 20% 100 10% 50 100% 500
Financial & Business
Services
includes:
Public administration,
education, health, other
personal & social services
includes:
Manufacturing,
Construction, Primary
Sector,
includes:
Distribution, Catering,
Transport & Communication
includes:

(The absolute numbers refer to countries with n=500)

Weighting was used in cases where the quotas could not be reached exactly in line with this
quota plan (mostly due to the limited absolute number of establishments in the two biggest
size classes). Note that because of the use of a single quota plan for all countries, country
differences in the distribution of employment over establishment size bands which occur in
reality are not reflected in the data. This is due the lack of available data on the distribution
of employment across establishments size bands in almost all EU Member States, and
constitutes a considerable problem. This weight is, therefore, not used for presenting SIBIS
results.

9.2.3.2 Weighting by employment

The data available on the distribution of employment over establishment size bands is very
limited for most EU Member States. SIBIS used data from a variety of sources, including

BT database (United Kingdom)
ISTAT Industry and Services Intermediate Census latest available, 1996 (Italy)
75
National Statistical Service of Greece - latest available, 1995 (Greece)
SIREN (France)
Tilstokeskus Official Statistics (Finland)
Heins + Partner B-Pool (Germany)
Schober Business Pool (Spain)

and adjusted using data from the DG Enterprise/ Eurostat SME Database (latest available,
1997), to estimate the establishment/ employment structure for each country in the sample.
The table below shows the resulting establishment size structure per country.

Country
D E EL F FIN I UK EU7
1 to 9 23% 23% 59% 17% 13% 38% 14% 23%
10 - 49 19% 28% 16% 22% 16% 22% 31% 24%
50 - 199 21% 21% 8% 21% 19% 14% 26% 20%
200 - 499 13% 9% 6% 14% 16% 7% 13% 12%
E
s
t
a
b
l
i
s
h
m
e
n
t

s
i
z
e

b
a
n
d

500 and more 25% 18% 10% 25% 37% 19% 17% 21%
Total Column % 100% 100% 100% 100% 100% 100% 100% 100%

Using this weight, the weighted sample for each country therefore reflects employee
distribution between the five establishment size bands within that country. This means that a
data reference of, for example, 20% of all establishments in country A should be
understood to mean establishments accounting for 20% of all employees in country A.

9.2.3.3 Weighting by employment for EU7 averages

Additionally another weighting factor was created to calculate average figures for all
countries in the sample (which together represent roughly 82% percentage of total EU
employment). Each country is represented in this weight according to its share in the total
employment of the 7 EU countries in which the survey was conducted.

9.2.4 Sample characteristics and effect of weighting

Total
unweighted
weighted by
employment*
% unweighted
% weighted by
employment*
Total sample 3139 3139 100.0 100.0
Country
Finland 306 306 9.7 9.7
France 501 501 16.0 16.0
Germany 512 512 16.3 16.3
Greece 301 301 9.6 9.6
Italy 512 512 16.3 16.3
Spain 507 507 16.2 16.2
UK 500 500 15.9 15.9
Total
unweighted
weighted by
employment
(EU7)
% unweighted
% weighted by
employment
(EU7)
Number of staff at site
up to 9 803 713 25.6 22.7
76
Total
unweighted
weighted by
employment*
% unweighted
% weighted by
employment*
10 to 49 769 746 24.5 23.8
50 to 199 668 648 21.3 20.6
200 to 499 626 364 19.9 11.6
500 and more 273 668 8.7 21.3
Industry Sector
primary: manufacturing, energy,
mining, construction
990 989 31.5 31.5
secondary: distribution, catering,
communication and transport
873 878 27.8 28.0
third: financial and business
services
502 501 16.0 15.9
fourth: public administration,
health, education, other social/
personal
774 772 24.7 24.6
Type of organisation
headquarter of international
operating organisation
267 348 8.5 11.1
headquarter of organisation only
operating in country
607 536 19.3 17.1
division/ branch of international
256 290 8.2 9.2
division/ branc h of organisation
only operating in the respective
country
309 312 9.8 9.9
other type 76 40 2.4 1.3
only one establishment 1617 1608 51.5 51.2
dont know 7 6 0.2 0.2
77
9.3 Questionnaires
9.3.1 Questionnaire for the General Population Survey (GPS)

Structure of the questionnaire:

Module IN: Introduction and screening
Age
Educational attainment
Employment status
Occupation
Main working place

Module A: Basic ICT equipment access and use
Use of computer
Use of e-mail
Internet access and use
Methods of Internet access
Effects of Internet use
Barriers to using the Internet
Access to mobile phone
Mobile data services
Effects of mobile phone use

Module B: E-commerce and other uses of the Internet
Online activities
Barriers to buying online

Module D: Skills
Internet user experience and know-how

Module L: e-Health
Use of online health information
Perception regarding the trust placed in online health information provider
Rationale for health info search

Module J: Security
Security concerns
Reporting of security violations
Security-related awareness and behaviour

Module K: e-Government
Preference for e-Government services
e-Government experience
Barriers to e-Government

Module E: Telework
Home-based telework
Intensity of home-based teleworking
Duration of telework:
Financing of tele-workplace
78
Interest in telework:
Perceived feasibility
Effects of telework

Module F: Mobile work
Mobile work (Intensity):
Mobile telework

Module G: Tele-cooperation/Tele-collaboration
Co-operation with external contacts using ICTs
e-Lancing

Module H: Outcomes of work
Work-family balance
Job quality
Job satisfaction

Module C: Educational attainment and lifelong learning
Company-provided training
Training provided by other organisations
Self-directed learning
Modes of training (use of eLearning)

Module Z: Standard demography
Household size
Disability
Income

No
Branching
Question Answer categories

79
Module IN: Introduction and Screener questions GPS
INTRO TEXT
ALL
Hello my name is ... calling for ...
We are presently conducting a scientific survey for the European Union
in fifteen countries. I would like to talk to the person in your household,
that is at least 15 years old, and whose birthday is up next.
[INTERVIEWER: IF NECESSARY] To topic of this survey is the
internet and the work life.
[INTERVIEWER: IF NECESSARY] Your answers will be held strictly
confidential and will be used only for scientific purposes.
[INTERVIEWER: IF NECESSARY] Your participation is very important
to us, because you have been selected through a statistical procedure
that will result in a typical selection of people in [COUNTRY]
[PROMPT: The interview will last about 15 minutes]

IN1
ALL
Would you please tell me in which year you were born? |_1_|_9_|__|__|
[DK]
PROGRAMMING: IF respondent born after 1986 END INTERVIEW!

IN2
ALL
Have you finished your full-time education or are you still studying?

(1) finished education already
(2) Is still studying
(3) DK
IN3
IF IN2=1
At what age did you finish full-time education?
[PROMPT: HOW OLD WHERE YOU WHEN YOU STOPPED FULL-
TIME EDUCATION]
|__|__| years
[DK]
Transition X1
IF IN2=1
I would like to ask you a few questions regarding your employment
situation.

IN4
IF IN2=1
At present are you in paid work either as an employee, civil servant or
as self-employed?
(1) yes
(2) no
(3) DK
IN5a
IF IN4=1
Do you have one job or more than one job at present? (1) only one job
(2) more than one job
(3) DK
IN5b
IF IN5a=2,3
How many hours per week do you normally work, including paid
overtime and taking all your jobs together?
|__|__|__|
[DK]
Transition X2
IF IN5a=2
For answering the following questions, please consider only your main
job, i.e. the job you spend most of your working time on.

IN6
IF IN4=1
And are you ... [in your main job]
[INTERVIEWER: Read out answer categories]
(1) self-employed
(2) in paid employment (including civil servants)
(3) DK
IN7
IF IN4=2,3
And are you ...
(1) temporarily not working, e.g. because of
unemployment, paternal leave or illness
(2) retired
(3) not working, because you are responsible for
ordinary shopping and looking after the home.
(4) DK
IN8
IF IN6=1
What kind of work do you do? Are you a ...
(1) Professional (eg doctor, lawyer, accountant,
architect)
(2) Farmer, fisherman
(3) Business proprietor, owner of company/shop,
craftsmen, other self-employed person
(4) DK
No
Branching

80
IN9
IF IN8=3
How many employees do you have? |__|__|__|__|__|__|
[DK]
IN10
IF IN4=1
[In your main job,] Are you working full-time or part-time? (1) full-time
(2) part-time
(3) DK
IN11
IF IN4=1
How many hours per week do you normally work in your main job,
[PROGRAMMER: Skip the following if IN6=1] including paid overtime?
|__|__|__|
[DK]
[PROGRAMMER: INCLUDE CHECK WITH IN5B]
IN12
IF IN6=2,3
Are you employed ...
(1) on an unlimited permanent contract
(2) on a fixed term contract
(3) on a temporary employment agency contract
(4) on apprenticeship or other training scheme
(5) other
(6) DK
IN13
IF IN7=2,3,4
Would you like to be in paid work? (1) yes
(2) no
(3) DK
IN14
IF IN8=3 or IN6=2
What kind of work do you do? Are you ...
(1) working mainly at a desk
(2) not working at a desk, but travelling
(salesmen, driver, ...),
(3) not working at a desk, but in a service job
(retail shop, restaurant, ...)
(4) doing some other kind of work
(5) DK
IN15
IF IN6=2
What position do you hold?
(1) Employed professional (employed lawyer,
medical practitioner, accountant, architect etc.),
(2) Management
(3) Other non-manual employee
(4) Manual worker
(5) DK
IN16
IF IN15=2
And which of the following best describes your position?
(1) General management, director or top
management (managing directors, director
general, other director)
(2) Middle management, other management
(department head, junior manager, teacher,
technician)
(3) DK
IN17
IF IN15=4
And which of the following best describes your position?

(1) Supervisor
(2) Skilled manual worker
(3) Other (unskilled) manual worker, servant
(4) DK
IN18
IF IN15=2
How many employees you are responsible for? |__|__|__|__|__|__|
[DK]
No
Branching

81
IN19
IF IN6=2
For what kind of organisation do you work?
(1) a private firm or business or a limited
company
(2) in the public sector or in a charity, voluntary
organisation or trust
[PROMPT - DO NOT READ: (2) includes public
companies, local or central government, civil
service, armed forces, council, schools,
universities or other grant funded education
establishments, public authorities, charities,
voluntary organisations]
(3) DK
IN20
IF IN6=2
How many employees work in the company/organisation for which you
work?
(1) <10
(2) 10-49
(3) 50-249
(4) 250 and more
(5) DK
IN21
IF IN4=1
Do you work mainly ...
(1) in your own home
(2) in the same grounds or buildings as your
home
(3) in different places using home as a base (e.g.
travelling salesman, free insurance agent etc.)
(4) somewhere quite separate from home
(5) DK
Module A: Basic ICT equipment access and use GPS
Transition A
ALL
Now we would like to ask you a few questions about computers and
the Internet

A1
ALL
Have you used a PC, Mac or any other computer, for work or for private
purposes - in the last four weeks?
(1) yes
(2) no
(3) DK
A3
IF A1=1
Have you sent or received any e-mail messages, for work or for private
purposes, during the last four weeks?
(1) yes
(2) no
(3) DK
A4a
IF A3=1
How many of your friends and relatives have their own email address?
(1) all or almost all
(2) about three quarters
(3) about half
(4) about one quarter
(5) only few or no-one
(6) DK
A4b
IF A4a<5
And with how many of your friends and relatives do you communicate
regularly via email?
(3) about half
(6) DK
A5
ALL
Do you have access to the Internet in your home? (1) yes
(2) no
(3) DK
A6
IF A5=2
Did you once have Internet access in your home? (1) yes
(2) no
(3) DK
No
Branching

82
A7
ALL
Have you used the Internet at least oncein the last four weeks, at
home, at school or work or at any other place?
(1) yes
(2) no
(3) DK
A8
IF A7=2,3
Have you used it in the last 12 months at least once? (1) yes
(2) no
(3) DK
A9
FOR (a):
IF A7=1 and A5=1
FOR (b)-(f):
IF A7=1
How much timedo you spend in a typical week on using the Internet ...
[item]
[INTERVIEWER: Read out answer categories for the first 2 items]
(a) at home?
(b) at the workplace?
(c) at school, university or another educational institution?
(d) at a public place where Internet access is free?
(e) at an Internet caf or other place where you have to pay for access?
(f) at another place not mentioned yet
FOR EACH
(1) none
(2) less than 1 hour
(3) between 1 and 5 hours
(6) more than 20 hours
(7) DK
A10
IF A7=1 or A8=1
When did you use the Internet for the first time?
(1) < 6 months ago
(2) 6 - 12 months ago
(3) 1 year - 2 years ago
(4) 2 years + ago
(5) DK
A11a
IF A5=1
Do you know what technical method you use at home to connect to the
Internet?
(1) yes
(2) no
(3) NA
A11b
IF A11a=1,3
I will read to you a number of methods to access the Internet. Which of
these do you use at home?
[INTERVIEWER: Read out and code those that apply]
MULTIPLE ANSWERS
(1) Dial-up with modem
(2) Cable Modem
(3) Leased line
(4) xDSL
(5) ISDN
(6) T1 or T3 line [TRANSLATOR: Digital Multiplex
connection]
(7) Internet access via satellite
(8) Other not mentioned (e.g. mobile)
(9) DK
A12
IF A11b=2,3,4,5,6,7
At home, did you have a connection before which was slower than your
current one?
(1) yes
(2) no
(3) DK
A13
IF A12=1
Since moving to this faster type of connection, has the amount of time
you spend online per week decreased, increased or remained roughly
the same?
(1) Decreased
(2) Increased
(3) Remained roughly the same
(4) DK
A14
IF A7=1
In the last four weeks, have you accessed the Internet in any other way
than via PC or Mac, at least once?
(1) yes
(2) no
(3) DK
No
Branching

83
A15
IF A14=1
Which devices did you use for that: Did you use ...
[INTERVIEWER: Read out and code those that apply]
MULTIPLE ANSWERS
(1) Digital TV*,
(2) a PDA or palmtop,
(3) a mobile phone with WAP or 2.5G** capability,
(4) a game console
(5) other
(6) DK
[* TRANSLATOR: Make sure that you take local
brand names and colloquial terms into account]
** TRANSLATOR: Use term used in your country
(e.g. Germany: GPRS)]
A18
IF A7=2,3
Now I will read to you a list of statements about the Internet.
Please tell me for each statement whether you agree completely, agree
somewhat or do not agree.
The Internet ... [item]. Do you ...
(a) requires advanced computer skills,
(b) is not easy enough to get access to,
(c) is too time consuming,
(d) is too expensive to use,
(e) lacks useful or interesting information
(f) is not something for me
FOR EACH
(1) agree completely
(2) agree somewhat
(3) or do you not agree
(4) DK
A19
ALL
Do you have a mobile phone for your own personal use? (1) yes
(2) no
(3) DK
A20
ALL
How many of your friends and relatives have a mobile phone for their
personal use?
(3) about half
(6) DK
A23
IF A19=1 and A15~=3
Have you used your mobile phone to view webpages or WAP pages, or
to read your email, at least once in the last 4 weeks?
[TRANSLATORS: Confusion with SMS* to be avoided!]
(1) yes
(2) no
(3) DK
A26
IF A23=1
Have you used your mobile phone at least once in the last 12 months
to make any purchases in the Internet, to download online information
you are charged for or to make online payments?
(1) yes
(2) no
(3) DK
A27
IF A19=1
Have you, in the last four weeks, used SMS* messages for ...
(a) communication with other people?
(b) paying for purchases, admission tickets or something similar?
(c) paying for downloads such as ringing tones?
(d) receiving financial information, sport results or other subscription
services?
[* TRANSLATOR: Check if another term is more common in your
country]
FOR EACH
(1) yes
(2) no
(3) DK
No
Branching

84
A30
IF A19=1
(For (d) and (e)I: IF A19=1 and
(A8=1 or A7=1) and IN4=1)
Now, think about what your everyday life would be like if you didnt
have a mobile phone. Please tell me how much you agree that if you
didnt have a mobile phone (ITEM). Would you say that you

(a) you would often not be able to contact your friends and family, or be
reached by them
(b) you would be less exposed to dangerous electromagnetic radiation
(c) you would be more helpless in case of emergencies
(d) you would not receive some of the information you need for your job
(e) you would have less exchange with some of your business contacts
(f) you would have less fun
FOR EACH:
(2) agree somewhat
(3) do not agree
(4) DK
Module B: E-commerce and other uses of the Internet GPS
Transition B
IF A8=1 or A7=1
Now I would like to ask you a few questions about the Internet.
PROGRAMMING: B1 to B2: for each item in B1=1 ask directly B2, then
go to next item in B1

B1
IF A8=1 or A7=1
You can use the Internet for many purposes. Im going to read you a
list of things you can do online and ask you whether you have done this
online for your private purposes. For your private purposes, have you
used it in the last 12 months...
(a) to find information about a product or service
(b) to order a product or service
(c) to conduct online-banking or to buy financial products
(d) to search for any health-related information
(e) to look for a job
FOR EACH
(1) yes
(2) no
(3) DK
B2
IF B1=1 and A7=1
[FOR EACH B1 ITEM] Have you done so in the last four weeks? (1) yes
(2) no
(3) DK
B5
IF A7=1
(For (c) and (d): IF A7=1 and
IN4=1)
Many people in this country still do not have access to the Internet yet.
Now please imagine our country were without the Internet for one
month. What would it mean for your everyday life?
Please tell me how much you agree that if our country were without the
Internet for a month you would (ITEM). Would you say that you would

(a) be less well informed as a consumer
(b) feel socially excluded
(c) not receive some of the information you need for your job
(d) have less communication with some of your contacts at work / your
business contacts
(e) have less contact with some of your friends
(f) have less fun
FOR EACH:
(2) agree somewhat
(3) do not agree
(4) DK
No
Branching

85
Module D: Skills GPS
D1
IF A7=1 or A8=1
[Do not ask item (h) in UK, IRL,
USA]
I would like to ask you a few questions about your skills in using the
Internet. How confident would you feel... [item]
Please tell me whether you feel..
(a) using a search engine (such as Google or Yahoo) to find
information on the Internet [TRANSLATORS: List two most widely used
search engine brands in your country
20
]
(b) identifying the source of information provided on the Internet
(c) using e-mail to communicate with others
(d) using Internet chat-rooms to contact other people
(e) using the Internet to make telephone calls
(f) creating a personal web / Internet page
(g) downloading and installing software onto a computer
[PROGRAMMING: Do not ask item (h) in UK, IRL, USA]
(h) understanding the content of websites written in English
FOR EACH
(1) very confident
(2) fairly confident
(3) not confident
(4) Do not know what this means [DO NOT READ
OUT]
(5) DK
Module L: e-Health GPS
Transition L
IF B1(d)=1
You said before, that you have used the Internet to search for health-
related information:

L1
IF B1(d)=1
Have you been able to find health related information on the Internet? (1) yes
(2) no
(3) DK
L2
IF L1=1
Was the information suitable for your needs? (1) yes
(2) no
(3) DK
L3
IF L2=1
Websites with health related information are available in many
languages.
When you searched, did you find Websites in your mother tongue
sufficient or did you have to expand your search and consult sites in
other languages, or did you even have to rely solely on sites in other
languages?
(1) Websites in mother tongue were sufficient
(2) Had to expand my search and consult
websites in other languages too
(3) Had to rely solely on websites in other
languages
(4) DK
L4
IF B1(d)=1
And for what reasons did you search health-related information on the
Internet?
Did you search health-related information on the Internet to ...[item]
(a) seek a second opinion on your own, a family members, or a
friends medical diagnosis?
(b) be better informed on your general health?
(c) gather additional information since you care for an ill person or a
person with a disability?
FOR EACH
(1) yes
(2) no
(3) DK

20
For example, check http://www.jupitermmxi.com/europelanding.html
No
Branching

86
L5
IF B1(d)=1
How trustworthy would you consider each of the following providers of
health-related information:
[Item] : Are those ...
(a) Universities and other non-profit organisations active in the health
sector / the health field
(b) pharmaceutical companies
(c) private health insurance providers
(d) patient advocacy and self-help groups
(e) hospitals
(f) professional medical associations
FOR EACH
(1) very trustworthy
(2) fairly trustworthy
(3) not trustworthy
(4) DK
Module J: Security GPS
Transition J
IF A7=1
Now the topic is internet security.
J1
IF A7=1
How concerned are you about .[item]: Are you ...
(a) data security on the Internet, i.e. the loss or manipulation of your
data?
(b) privacy and confidentiality on the Internet, i.e. personal information
about you being misused by third parties?
FOR EACH
(1) very concerned
(2) somewhat concerned
(3) not concerned
(4) DK
J2
IF J1(a)=1,2 or J1(b)=1,2
Are these concerns stopping you from using the Internet to buy goods
or services online: often, sometimes, or never?
(1) often
(2) sometimes
(3) never
(4) DK
J3
IF A7=1
Would you report violations of your on-line security, privacy and
confidentiality to a third independent party, for example a public agency
created for this task?
(1) yes, very likely
(2) maybe
(3) no
(4) DK
J4
IF J3=1,2,3
Would it be easier for you to do so if you could do it anonymously? (1) yes
(2) no
(3) DK
J5
IF A7=1 & (B1(b)=1 or B1(c)=1)
How often are you aware of security features of websites when you use
the Internet to buy online: often, sometimes or never?
(1) often
(2) sometimes
(3) never
(4) DK
J6
IF A7=1 & (B1(b)=1 or B1(c)=1)
And how often do you take security features of websites into account
when deciding about whether to buy online: often, sometimes or never?
(1) often
(2) sometimes
(3) never
(4) DK
Module K: e-Government GPS
Transition K
IF A7=1
Now I would like to ask you a few questions about the contact to
government agencies through the Internet.

PROGRAMMING: K1 to K3: for each item in K1=1 ask directly K2, If
K2=1 ask directly K3, then go to next item in K1

No
Branching

87
K1
IF A7=1
Here is a list of activities that require citizens to get in touch with public
administration.
For each activity, please answer whether you would prefer to use the
Internet or prefer to use the traditional way, that is face-to-face, by
postal mail, fax or phone:
[INTERVIEWER: Repeat answer categories for the first 2 items]
(a) Tax declaration / filing your income tax return
(b) Use of job search services of public employment service
(c) Request for passport, driver's licence, birth certificates or other
personal documents
(d) Car registration
(e) Declaration to the police, e.g. in case of reporting theft
(f) Searches for books in public libraries
(g) Announcement of change of address
FOR EACH
(1) Internet
(2) traditional way
(3) do not use this service [DO NOT READ OUT]
(4) DK
K2
IF K1=1
FOR EACH
Is it possible to use the Internet for this in the area you live?
FOR EACH
(1) yes
(2) no
(3) DK
K3
IF K2=1
FOR EACH
Have you ever tried using the Internet for this?
FOR EACH
(1) yes
(2) no
(3) DK
K4
IF A7=1
For each of the following statements about online services of public
administration, please indicate whether you agree. Public services on
the Internet ...[item].
(a) are not useful enough
(b) are faster than the traditional way
(c) require that you install special equipment or software
(d) reduce the number of mistakes public authorities make
(e) do not seem as safe as using the traditional way
(f) make it possible to deal with the authorities at more convenient
times
(g) make it possible to deal with the authorities at more convenient
locations, e.g. from home or from the workplace
(h) are difficult to use
(2) agree somewhat
(3) do not agree
(4) DK
Module E: Telework GPS
Transition E
IF IN4=1 or IN13=1 or IN7=1
Now lets talk about another topic:
With the help of telephone, fax and computer, many types of work can
be done from home. If work results are transferred electronically, this is
sometimes called telework.

E1
IF IN4=1
Do you presently telework from home, for at least some of your working
time?
(1) yes
(2) no
(3) DK
E2
IF E1=2,3
Have you teleworked on a regular basis before, in the last five years? (1) yes
(2) no
(3) DK
E3
IF E2=1
Did you spend, on average, at least one full working day a week at
home when you were teleworking?
(1) yes
(2) no
(3) DK
No
Branching

88
E4
IF E1=1
Do you spend, on average, at least one full working day a week
teleworking from home?
(1) yes
(2) no
(3) DK
E5
IF E1=1
You indicated before that you work on average [PROGRAMMER: Insert
result from IN5b, if blank insert result from IN11] hours per week. How
many of these do you spend at home in a typical week?
|__|__|__|
[DK]
[PROGRAMMER: Insert check with IN5b or IN11]
E7
IF E1=1 and IN6=2
Has the equipment you use for teleworking at home been mainly, not
mainly but partly, or not at all been paid for by your employer?
(1) mainly paid for by employer
(2) not mainly, but partly paid for by employer
(3) not at all paid for by employer
(4) DK
E8
IF IN7=1 or IN13=1 or (E1=2,3 or
E4=2,3)
If it was offered to you, how interested would you be in ... [item]. Would
you be ...
(a) doing almost all your work teleworking at home
(b) telework where you did not spend all your working time, but at least
one full working day per week at home
(c) work in an office provided near your home which would allow you to
reduce commuting?
FOR EACH
(1) very interested
(2) somewhat interested
(3) not interested
(4) DK
E9a
IF E1=2,3 or E4=2,3
Would you say that your job is feasible for telework, under the
assumption that you spend at least one full working day per week at
home?
(1) yes
(2) no
(3) DK
E9b
IF E9a=2 and IN6=2
What are the main reasons why you consider your current job not to be
feasible for telework? Is it because ...
[INTERVIEWER: Read out answer categories and code all that apply]
MULTIPLE ANSWERS
(1) your company does not permit telework?
(2) your superior does not approve of telework?
(3) your job requires face-to-face contact with
customers, colleagues or other persons
(4) your job requires access to machines or other
things which cannot be accessed from home
(5) Other reasons (DO NOT READ OUT)
(6) DK
E10
IF E1=1
For what reasons did you start teleworking? Please indicate for each of
the following aspects how important it was for your decision to start
teleworking. [item] Was this ... for you.
(a) I needed a more peaceful working environment
(b) I want to participate more in family life
(c) I want to be closer to clients or customers
(d) I need to look after a child or an other person who needs care
(e) My company asked me to start teleworking
(f) I want to reduce commuting
(g) I wanted to have more flexibility in how to organise my work
(1) very important
(2) somewhat important
(3) not important
(4) DK
No
Branching

89
E11
IF E1=1
Most working people are not allowed to work from home. Please
consider you would not be allowed to telework from home, for whatever
reasons.
What would that mean for your ability to do your job? Would it mean
that you...[item]. Do you ...
(a) could not be in paid work at all
(b) could not do your job as well as with telework
(c) would have to look for another job which is located closer to your
home
(d) would have to reduce your working hours per week
FOR EACH:
(2) agree somewhat
(3) do not agree
(4) DK
Module F: Mobile work GPS
Transition F
IF IN4=1
Now lets talk about the topic of mobile working.
F1
IF IN4=1
In the last four weeks, have you spent any of your working time away
from your home and from your main place of work, e.g. on business
trips, in the field, travelling or on customers premises?
(1) yes
(2) no
(3) DK
F2
IF F1=1
You indicated before that you work on average [PROGRAMMER: Insert
result from IN5b, or if blank result from IN11] hours per week. How
many of these do you spend away from home and your main place of
work?
|__|__|__|
[DK]
[PROGRAMMER: Insert check with IN5b or IN11]
F3
IF F2>5
In the last four weeks, have you used online computer connections
when travelling? By this I mean have you accessed the Internet for
business purposes, or electronically transferred data to colleagues?
(1) yes
(2) no
(3) DK
F4
IF F3=1
For what purpose did you use these online connections? Have you
used these to ...
(a) access the Internet
(b) send or read e-mails
(c) connect to your company's internal computer system
FOR EACH:
(1) yes
(2) no
(3) DK
F5
IF F3=1
Where did you use an online computer connection? Have you used it in
the last four weeks at ...
(a) a hotel, conference site or similar location?
(b) another company's premises?
(c) an Internet caf or an other commercial teleservice center?
(d) or on the move, using a mobile device for data transfer?
FOR EACH:
(1) yes
(2) no
(3) DK
Module G: Tele-cooperation/Tele-collaboration GPS
Transition G
IF IN4=1 and (A1=1 or A7=1)
And how about the use of telecommunication technology at your work
place:

G1
IF IN4=1 and A1=1
When you communicate with external contacts, do you sometimes use
e-mail, video conference or electronic data transfer? [PROGRAMMER:
skip the following if IN6=1] By external persons we mean customers,
clients, suppliers, other business contacts, but also colleagues working
at other locations of the same company.
(1) yes
(2) no
(3) DK
G2
IF G1=1
In a typical week, how often do you ...[item] for these external contacts?
(a) use e-mail
(b) use video-conferencing
(c) use e-mail attachments or other electronic data transfer
FOR EACH
(1) 10 or more times a day,
(2) at least once aday,
(3) at least once a week
(4) less often than once a week
(5) never
(6) DK
No
Branching

90
G4
IF IN6=1 and A7=1
I would like to know about the role the Internet plays in your business.
Do you sometimes attract new business through the Internet or via e-
mail?
(1) yes
(2) no
(3) DK
G5
IF IN6=1 and A7=1
Do you sometimes deliver work results to your clients or customers
through the Internet or via e-mail?
(1) yes
(2) no
(3) DK
G6
IF G4=1 and G5=1
Does it sometimes happen that you communicate with clients or
customers exclusively by electronic means, i.e. via Internet, e-mail,
phone or fax and without meeting face-to-face?
(1) yes
(2) no
(3) DK
Module H: Outcomes of work GPS
Transition H
IF IN4=1
I would like to ask you a few more questions about your work.
H1
IF IN4=1
Please tell me for each of the following, how often you experience this.
How often do you .. [item]?
(a) Find your work stressful
(b) Come home from work exhausted
(c) Find your job prevents you from giving the time you want to your
partner or family
(d) Feel too tired after work to enjoy the things you would like to do at
home
(e) Find your partner/family gets fed up with the pressure of your job
FOR EACH
(1) often
(2) sometimes
(3) never
(4) does not apply [DO NOT READ]
(5) DK
H2
IF IN6=2,3
In your current work arrangement, do you agree with the following
statements about your job? [item] Do you ...
(a) I have a lot of say over what happens in my job
(b) I need to keep learning new things continuously
(c) I have concerns about whether my job is secure
(d) I have a high income
(e) I can adapt my starting & finishing times to my personal preferences
(f) I can adapt the number of weekly working hours to my personal
preferences
FOR EACH:
(1) strongly agree
(2) somewhat agree
(3) disagree
(4) DK
H3
IF IN4=1
On the whole, are you very satisfied, somewhat satisfied, neither
satisfied nor dissatisfied, somewhat dissatisfied or very dissatisfied with
your job / your main job?
(1) very satisfied
(2) somewhat satisfied
(3) neither satisfied nor dissatisfied
(4) somewhat dissatisfied
(5) very dissatisfied
(6) DK
Module C: Educational attainment and lifelong learning GPS
Transiti on C
IF IN4=1 or IN13=1 or IN7=1
Now I would like to ask you a few questions about training and learning.
C2
IF IN6=2,3
Did you participate in some kind of work-related training activities that
were provided either by your company or by an other organisation, in
the last four weeks?
(1) yes
(2) no
(3) DK
C9b
IF IN7=1 or IN6=1
Did you participate in some kind of training activities with the aim of
preparing you for a future job, in the last four weeks?
(1) yes
(2) no
(3) DK
No
Branching

91
C14a
IF IN4=1
Apart from the training that may have been provided by others, did you
engage in some kind of self-directed learning related to your work, in
the last four weeks?
(1) yes
(2) no
(3) DK
C14b
IF IN7=1 or IN6=1
Apart from the training that may have been provided by others, did you
engage in some kind of self-directed learning which was aimed at
preparing you for a future job, in the last four weeks?
(1) yes
(2) no
(3) DK
C18
IF A1=1 and (C2=1 or C9b=1 or
C14a=1 or C14b=1)
Did you use, in the course of your training and learning in the last four
weeks, electronic learning materials such as learning programmes on
CD-ROM, in company-internal computer systems or on the Internet?
(1) yes
(2) no
(3) DK
C19
IF C18=1
What did you use? Did you use
(a) CD-ROMs or other so-called offline media such as diskettes, audio
or video tapes etc.?
(b) online learning materials provided on the internal computer system
of your organisation or through the Internet
FOR EACH
(1) yes
(2) no
(3) DK
C20
IF IN2=2 and A1=1
Did you use, in the course of your studies in the last four weeks,
electronic learning material such as learning programmes on CD-ROM,
on the internal computer system of your school/university or through
the Internet?
(1) yes
(2) no
(3) DK
C21
IF C20=1
What did you use? Did you use
(a) CD-ROMs or other so-called offline media such as diskettes, audio
or video tapes etc.?
(b) online learning material provided on the internal computer system of
your school/university or through the Internet?
FOR EACH
(1) yes
(2) no
(3) DK
Module Z: Standard demography GPS
Finally we would like to ask you a few more questions for statistical
purposes:

Z17
ALL
How many people live in your household, yourself included? |__|__|
[DK]
Z18a
IF Z17>1
How old is the youngest? |__|__|
[DK]
Z18b
IF Z17>1
How many are 15 years and older? |__|__|
[DK]
[PROGRAMMER: Build in check with Z17 and
Z18a]
Z14
ALL
Do you have any long-standing illness, disability or infirmity that limits
your activities in any way? By long-standing I mean anything that has
troubled you over a period of time or that is likely to affect you for a
period of time.
(1) yes
(2) no
(3) DK
Z19
ALL
We also need some information about the income of this household to
be able to analyse this survey.
What is your household's monthly net income (after tax)? Please count
the total wages and salaries per month of all members of this
household; all pensions and social security benefits; child allowances
and any other income like rents etc.
[ADD IF NECESSARY: Of course, your answer (as all other answers in
this interview) will be treated confidentially and referring back to you or
your household will be impossible.]
Is it less or more than <income 1>, <income 2> or <income 3>.
(1) less than <income 1>
(2) <income 1> to less than <income 2>
(3<income 2> to less than <income 3>.
(4 <income 3> or more
(5) DK
(6) Refusal
No
Branching

92
Z20
ALL
Looking back over the last three years, has your household income
increased, decreased, or remained roughly the same?
(1) increased
(2) decreased
(3) remained roughly the same
(4) DK
(5) Refusal
Z21
ALL
Gender
[INTERVIEWER: Ask only if in doubt]
(1) male
(2) female
Data provided by survey organisation Categories
P0 Survey Number 101438
P1
Country Code
|__|__|
P2 Interview Number |__|__|__|__|
P3 Date of Interview: Day |__|__|,
Month |__|__|
P4 Time of the beginning of the interview (USE 24 HOUR CLOCK): Hour |__|__|,
Minute |__|__|
P5 Number of minutes the interview lastet |__|__|__|
P6 Size of locality |__|__|
P7
Region
|__|__|
P8a Postal Code / Area code
must be convertible into NUTS 2 regions
|__|__|__|__|__|__|__|__|
P8b NUTS 2 regions |__|__|__|__|__|__|__|__|
P9 Interviewer Number |__|__|__|__|__|
P10 Weighting Factor |__| . |__|__|__|__|__|
P11 Language of interview (Luxembourg, Belgium, Finland, Switzerland) |__|

93
9.3.2 Questionnaire for the Decision Maker Survey (DMS)

Structure of the questionnaire:

Introduction and Screener Section

Module A: Basic characteristics
Number of staff (employees)
Turnover

Module B: Module B: Basic ICTs take-up and intensity of use (e-Business)
e-Mail
Internet
Intranet
EDI
Video-conferencing
Call-centre
Staff access to ICTs

Module C: e-Commerce
Website/ Internet presence
Online sales
Barriers to e-commerce (selling)
Benefits from / Outcomes of e-commerce
Online procurement
Barriers to online procurement
Benefits from/ Outcomes of online procurement
Online supply chain integration
e-Marketplaces

Module D: e-Business security
Security breaches
Information security strategy
Barriers to security
Security provisions

Module F: e-Government
Use of e-Government services
Barriers to e-Government

Module G: Website accessibility
Design for all / universal design principle awareness

Module E: R&D
R&D staff
Computer staff in R&D unit(s)
IT staff providing computer services to R&D
Outsourced computer services for R&D
Vacancies in IT for R&D

No
Branching

94
Introduction and Screener Section DMS
Database/address information: Categories
A11
ALL
Main business activity
PROGRAMMER: Copy fromdatabase

NACE code (2-digit level)
I__I__I

1 Mining, Energy
(includes NACE 10 - 14/ 40, 41)
2 Manufacturing
(includes NACE 15 - 37)
3 Construction
(includes NACE 45)
4 Distribution
(includes NACE 50, 51, 52)
5 Hotels, Restaurants
(includes NACE 55)
6 Transport, Communication
(includes NACE 60, 61,62, 63, 64)
7 Banking, Insurance
(includes NACE 65, 66, 67)
8 Business Services
(includes NACE 70, 71, 72, 73, 74
[except: 74.13])
9 Public Administration
(includes NACE 75 [except 75.2])
10 Education
(includes NACE 80)
11 Health and Social Work
(includes NACE 85)
12 Other personal or social services
(includes NACE 90, 91, 92, 93)

A12
ALL
Establishment/ size (if available)
PROGRAMMER: Copy from database
According to database
a) OPEN (if available)
I__I__I__I__I__I__I 6-digit
numerical
[1] not available from database

and

b) in categories, i.e.
(1) 0 - 9
(2) 10 - 49
(3) 50 - 199
(4) 200 - 499
(5) 500+
(6) not available from database
No
Branching

95
S1 (INTRO)
ALL
At reception/switchboard:
Good morning/good afternoon. My name is ... . I am calling for
... [name of institute].
We are presently conducting a scientific survey in several
European countries. The topic is the use of information and
communications technologies. I would like to talk to the person
who is responsible for DP/IT decisions at your location.
INT.: NOTE:
THIS SHOULD BE THE HEAD OF THE DP/IT DPT. OR A
SENIOR PERSON IN THE DP/IT DPT. IN SMALLER FIRMS IT
CAN ALSO BE THE MANAGING DIRECTOR, THE GENERAL
MANAGER OR THE OWNER.
INT.: ADD, IF NECESSARY:
Your participation is very important to us, because your firm has
been selected through a statistical procedure that will result in a
typical selection of firms in [COUNTRY]
The interview will last approx. 15 minutes

(1) put through to target person
CONTINUE
(2) target person currently unavailable
MAKE APPOINTMENT FOR CALLBACK
(3) no such person TERMINATE
(4) refusal to participate END
S2 (INTRO)
ALL
At target person:
Good morning/good afternoon. My name is ... . I am calling for
... [name of institute].
We are presently conducting a scientific survey in several
European countries. The topic is the use of information and
communications technologies. We are talking to people who are
responsible for DP/IT decisions at their respective locations.
Can I just check: Would you be the right person to talk to at your
location and can we do the interview now?
Your participation is very important to us, because your firm has
been selected through a statistical procedure that will result in a
typical selection of firms in [COUNTRY]
The interview will last approx. 15 minutes

(1) yes, interview now CONTINUE
(2) yes but no time at the moment
MAKE APPOINTMENT FOR CALLBACK
(3) no, other person responsible at this
location ASK TO BE PUT
THROUGH TO THAT PERSON ,
RESPECTIVELY ASK FOR CONTACT
DETAILS. AT NEW TARGET
PERSON START AGAIN WITH
QUESTION S2
(4) no, other person responsible at another
location TERMINATE
(5) refusal to participate
TERMINATE

A13
ALL
Function of target person
What is your position in your establishment? What of the
following is the most appropriate?
INT.: READ OUT. SINGLE ANSWER.
(1) Owner/Proprieter
(2) Managing Director/Board Member
(3) Head of Establishment/Site
(4) Head of IT/DP
(5) Other senior member of IT/DP
Department
(6) Other TERMINATE
Module A: Basic characteristics DMS
Transition A
ALL
Let us start with some general questions about your
establishment.

No
Branching

96
A2
ALL
Does your organisation have only one establishment, or has it
more than one establishment?
By establishment we mean a single indentifiable unit at a
particular address.
[TRANSLATOR: Be very careful to identify a correct
translation for "establishment"]
(1) only one establishment
(2) more than one establishment
(3) DK
A4
IF A2=2
How many employees does your organisation have in total in
[country], including yourself?
INT.: IF "DK" SAY:
If you do not know it exactly, can you give me an estimate?

numerical
[DK]
A5
ALL
And how many employees work for your organisation AT THIS
ESTABLISHMENT, including yourself?
INT.: IF "DK" SAY: If you do not know it exactly, can you give
me an estimate?

PROGR.: CHECK:
IF A2=(2), Answer in A5 MUST be <Answer in A4!
IF NOT RE-ASK A4 / A5

numerical

[DK] TERMINATE INTERVIEW

PROGR.: CHECK QUOTA (according to answer in A5)

1 up to 9 employees QUOTA
2 10 - 49 employees QUOTA
5 500+employees QUOTA

IF "DK" TO QUESTIONS A5

A3
IF A2=2
Is your establishment ...?
INT.: READ OUT ALL ANSWER CATEGORIES. SINGLE
ANSWER.
(1) the headquarters of an internationally
(2) the headquarters of an organisation
that only operates in this country
(3) a division or branch operation of an
internationally operating organisation
(4) a division or branch operation of an
organisation that only operates in this
country
(5) other [INT.: DO NOT READ]
(6) DK
No
Branching

97
A8
ALL
EXCEPT IF A11 (NACE Code)
= 75, 80, 85

Please indicate your establishment's turnover in the last
financial year.
INT.: IF "DK", SAY:
If you do not know itexactly, can you give me a rough estimate?
INT.: PLEASE TRY TO GET AT LEAST AN ESTIMATE.
INDICATE IF ANSWER IS GIVEN IN EURO OR IN
PREVIOUS NATIONAL CURRENCY (/UK: RESP. OR
IN GBP)

(1) Turnover given IN EURO
(2) Turnover given IN PREVIOUS
NATIONAL CURRENCY (UK:
Always use GBP)
(3) DK, no answer to turnover

Turnover given:
I__I__I__I.I__I__I__I.I__I__I__I.I__I__I
__I
12-digit numerical

A9
ALL
EXCEPT IF A11 (NACE-Code)
=75, 80, 85
Has the turnover of your establishment increased, decreased or
roughly stayed the same when comparing the last financial year
with the year before?
(1) increased
(2) decreased
(3) roughly stayed the same
(4) DK
Module B: Basic ICTs take-up and intensity of use (e-Business) DMS
Transition B
ALL
Now we would like to ask you some questions about the use of
Information and Communications Technologies in your
establishment.

B1
ALL
Does your establishment use e-mail? (1) yes
(2) no
(3) DK
B2
ALL
Does your establishment have access to the World Wide Web,
i.e. the Internet?
(1) yes
(2) no
(3) DK
B3
ALL
Does your establishment have an Intranet, i.e. an internal
computer network that uses the Internet protocol?
(1) yes
(2) no
(3) DK
B5
ALL
Does your establishment use EDI, i.e. electronic data
interchange using the EDI standard?
(1) yes
(2) no
(3) do not know what this is [IF
SPONTANEOUSLY SAID]
(4) DK
B6
IF B5=1
Is your EDI Internet based? (1) yes
(2) no
(3) do not know what this is [IF
SPONTANEOUSLY SAID]
(4) DK
B7
ALL
Does your establishment use video-conferencing in your own
facilities?
(1) yes
(2) no
(3) DK
No
Branching

98
B8
ALL
Does your establishment use a call center for communication
with customers or other external contacts?
(1) yes
(2) no
(3) DK
B11
IF B1=1
Which applications can be accessed by the majority of your
office workers?
Can the MAJORITY OF YOUR OFFICE WORKERS
... send e-mails to external addresses?
(1) yes
(2) no
(3) DK
B12
IF B2=1
(What applications can be accessed by the majority of your
office workers?)
... browse Internet sites?
(1) yes
(2) no
(3) DK
B13
IF B3=1
(What applications can be accessed by the majority of your
office workers?)
... browse INTRANET sites?
(1) yes
(2) no
(3) DK
Module C: E-commerce DMS
Transition C
ALL
Now we would like to ask you some questions about E-
commerce. Please refer to your establishment when answering.

C1
ALL
Does your establishment put information on the Internet, for
example by means of a website?
(1) yes
(2) no
(3) DK
C2
IF C1=1 or 3
Do you sell goods or services via the Internet? (1) yes
(2) no
(3) DK
C3a
IF C1=1 or 3
Do you offer online reservation? By this we mean that your
customers can make a reservation for a product or service
through the Internet.
(1) yes
(2) no
(3) DK
C3b
IF C2=1
Do you distribute digital products or services online? By this we
mean that the product is transferred to the customer online, or
the service is provided online.
(1) yes
(2) no
(3) DK
C4a
IF C2=1
Are some of your online sales to businesses? (1) yes
(2) no
(3) DK
C5a
IF C4a=1
How large a share of your total sales to businesses are
conducted online?
Would you say ...
INT.: READ OUT. SINGLE ANSWER
(1) less than 5%
(2) 5 up to 25%
(3) 26 up to 50%
(4) 51 up to 75%
(5) more than 75%
(6) DK
C4b
IF C2=1
Are some of your online sales to consumers? (1) yes
(2) no
(3) DK
No
Branching

99
C5b
IF C4b=1
How large a share of your total consumer sales are conducted
online?
Would you say ...
(1) less than 5%
(2) 5 up to 25%
(3) 26 up to 50%
(4) 51 up to 75%
(5) more than 75%
(6) DK
C4c
IF C2=1
Are some of your online sales to the public sector? (1) yes
(2) no
(3) DK
C5c
IF C4c=1
How large a share of your total sales to the public sector are
conducted online?
Would you say ...
(1) less than 5%
(2) 5 up to 25%
(3) 26 up to 50%
(4) 51 up to 75%
(5) more than 75%
(6) DK
C6
IF C2=1
Are your online sales MAINLY to a local, national or global
market?
INT.: SINGLE ANSWER.
(1) local market
(2) national market
(3) global market
(4) DK
C7
IF C1=2
OR
IF C2=2 or 3
I am now going to read you a list of statements about selling
online. For each statement, please tell me whether you agree
completely, agree somewhat or do not agree from the point of
view of your establishment.
How about the statement ... [item].
Do you ...
INT.: READ OUT ANSWER CATEGORIES. ONE ANSWER
PER ITEM.

(a) Selling our products and services requires face-to-face
interaction with customers
(b) The necessary technology is expensive
(c) The costs for the promotion of the online offer are high
(d) The revenue potential of online sales is low
(e) Customers might be concerned about data protection or
security issues
(f) Adapting corporate culture to e-commerce is difficult
(g) The necessary skills are not readily available
(h) Handling the delivery process causes problems

FOR EACH:
(2) agree somewhat
(4) DK
No
Branching

100
C8
IF C2=1
You said earlier that you make sales online.
According to your experience, what effect has selling online on
... [item]?
Would you say the effect is ...
PER ITEM.
(a) your sales
(b) your costs
(c) your sales area
(d) the quality of your customer service
(e) the efficiency of your internal business processes

FOR EACH:
(1) very positive
(2) rather positive
(3) neither positive nor negative
(4) rather negative
(5) very negative
(6) DK
C9
IF B2=1 or 3
Do you use the Internet or other online services to purchase
goods or services?
(1) yes
(2) no
(3) DK
C10
IF C9=1
Roughly what proportion of the maintenance, repair and
organisation goods your establishment buys are purchased
online, measured in amount spent?
Would you say ...
(1) less than 5%
(2) 5 up to 25%
(3) 26 up to 50%
(4) 51 up to 75%
(5) more than 75%
(6) DK
C11
IF B2=2
OR
IF C9=2 or 3
I am now going to read you a list of statements about
purchasing online. For each statement, please tell me whether
you agree completely, agree somewhat or do not agree from the
point of view of your establishment.
How about the statement ... [item].
Do you ...
PER ITEM.

(a) Purchasing procurement products or services requires
face-to-face interaction with suppliers
(b) Our suppliers do not sell online
(c) The necessary technology is expensive
(d) The cost advantage is negligible
(e) We are concerned about data protection or security issues
(f) The legal protection of online contracts is not sufficient
(g) The necessary skills are not readily available
(h) Suppliers technical systems are not compatible with ours

FOR EACH:
(2) agree somewhat
(4) DK

No
Branching

101
C12
IF C9=1
You said earlier that you purchase goods or services online.
According to your experience, what effect has online
procurement on ... [item]?
Would you say the effect is ...
PER ITEM.
(a) your procurement costs
(b) stock-keeping of maintenance, repair and organisation
goods
(c) the number of suppliers
(d) your relations to suppliers
(e) the efficiency of your internal business processes

FOR EACH:
(1) very positive
(2) rather positive
(3) neither positive nor negative
(4) rather negative
(5) very negative
(6) DK
C13
IF C1=1
Does your establishment have an EXTRANET, i.e. a private,
secure network running on the Internet protocol and accessible
for selected external users?
(1) yes
(2) no
(3) DK
C14
IF C13=1
For which of the following purposes do you use your Extranet?
Do you use it for ... [item]
INT.: ONE ANSWER PER ITEM.
(a) communication with customers or clients?
(b) communication with suppliers?
FOR EACH:
(1) yes
(2) no
(3) DK
C15
IF B2=1
Do you have access to the Extranet of one of your supplier,
partner or customer organisations?
PROGR.: IF C1=2 or 3, add:
By Extranet I mean a private, secure network running on the
Internet protocol and accessible for selected external users.
(1) yes
(2) no
(3) DK
C19
IF B2=1
Does your establishment trade goods or services through an e-
marketplace? By e-marketplace I mean a business-to-business
Internet trading forum in which multiple buyers and sellers
exchange goods and services within an industry group or
geographic region.
(1) yes
(2) no
(3) DK
C20
IF C19=1
On e-marketplaces, different types of business transactions can
be accomplished. In which of the following types is your
establishment actively involved?
INT.: READ OUT AND CODE ALL THAT APPLY
(1) catalogue-based offering of products or
services
(2) catalogue-based purchasing of
products or services
(3) auctions -- as a seller
(4) auctions -- as a bidder
(5) launching calls for tenders
(6) answering calls for tenders
(7) powerbuying, i.e. joint purchases
together with other organisations to save
costs
(8) none of these
(9) DK
Module D: e-Business security DMS
Transition D
IF C1=1
Let us now turn to the topic of information security. Again,
please refer to your establishment when answering.

No
Branching

102
D1
IF C1=1
Many establishments are affected by security breaches such as
identity theft, online fraud, manipulation of software applications,
computer viruses or unauthorised entry to internal networks.
Have any breaches of your information security occurred in your
establishment in the last 12 months?
(1) yes
(2) no
(3) DK
D2a
IF D1=1
Progr.: Note for D2a to D2b:
For each itemin D2a=1, ask directly D2b; then go
to next itemin D2a!!

Which of the following types of information security breaches
have occurred in your establishment in the last 12 months? Did
you experience cases of ... [item]?
INT.: READ OUT. ONE ANSWER PER ITEM.
(a) Identity theft
(b) Online fraud
(c) Manipulation of software applications
(d) Computer virus infections
(e) Unauthorised entry to internal networks

FOR EACH:
(1) yes
(2) no
(3) DK
D2b
(For Each Item) IF D2a=1
And how substantial were the consequences of this security
breach for your establishment? Would you say they were ...
INT.: READ OUT ANSWER CATEGORIES. SINGLE
ANSWER (PER ITEM ASKED)
FOR EACH ITEM IF D2a=1

(1) very substantial
(2) rather substantial
(3) not substantial
(4) DK
D3
IF D1=1
Where do you believe these breaches mainly came from? Do
you think the largest threat to online security came from ...
INT.: READ OUT ANSWER CATEGORIES. CODE ALL
THAT APPLY
MULTIPLE ANSWERS
(1) Customers
(2) Suppliers/competitors
(3) Former employees
(4) Computer hackers
(5) Internal users
(6) Others, not mentioned yet
(7) DK
D4
IF D1=1
How have you learned about these breaches, in most cases?
Were you ... [item]
INT.: READ OUT, CODE ALL THAT APPLY

MULTIPLE ANSWERS
(1) alerted by a customer/supplier
(2) alerted by employees or did you notice
yourself
(3) notified by your own information
security system
(4) made aware by damage or loss of data
(5) alerted by the providers of outsourced
security services
(6) in another way (DONOT READ)
(7) DK
No
Branching

103
D5
IF C1=1
Does your establishment or your organisation have an
information security policy?
(1) yes
(2) no
(3) DK
D6
IF D5=1
How would you describe it? As formal or informal? (1) formal
(2) informal
(3) DK
D7
IF D5=1
Which are your information security priorities?
How much priority is given to ... [item]
PER ITEM.
(a) Blocking of unauthorised access
(b) Expanding budget for security measures
(c) Defining the security architecture
(d) Outsourcing security management

FOR EACH
(1) high priority
(2) medium priority
(3) low priority
(4) DK
D8
IF C1=1
How important are the following factors as barriers to effective
information security inside your establishment?
How about ...[item]:
Is this factor as a barrier to effective information security inside
your establishment...
PER ITEM.
(a) High costs for security measures
(b) Lack of staff training
(c) Lack of staff time
(d) Complexity of the technology
(e) Lack of employee co-operation

FOR EACH:
(1) very important
(2) fairly important
(3) not important
(4) DK
D9
IF C1=1
Which of the following tools do you use for information security
in your establishment? Do you make use of ... [item]

(a) Control of access to the computer system
(b) Cryptography/ data encryption
(c) Vulnerability Assessment Tools
(d) Firewalls
(e) Security Training and Awareness Rising Activities
(f) Intrusion Detection Systems
(g) End-user Security Training Classes

FOR EACH:
(1) yes
(2) no
(3) DK
Module F: e-Government DMS
Transition F
IF B2=1
Now lets turn to the topic of using online services for interacting
with public administration.

No
Branching

104
F1
IF B2=1 AND A11 (NACE-
Code) NOT =75 (Public
Admin)
Progr.: Note for F1 to F2:
For each itemin F1=2, ask directly F2; then go
to next itemin F1!!

I am going to read you a list of activities for which
establishments have to get in touch with public administration.
For which of these activities do you already use online media
such as EDI or the Internet?
What about ...[item]? Do you use online media such as EDI or
the Internet for this?

(a) Payment of social contribution for employees
(b) Corporation tax declaration
(c) VAT declaration
(d) Submission of data to statistical offices
(e) Obtaining environment-related permits
(f) Participation in public invitation to tender

FOR EACH
(1) yes
(2) no
(3) DK
F2
(For Each Item)
IF F1=2
Would your establishment prefer to use online media such as
EDI or the Internet for this purpose?
FOR EACH ITEM IF F1=2
(1) yes
(2) no
(3) DK
Transition F3
IF B2=2 or 3
Now lets turn to the topic of using online services for interacting
with public administration.
It is now possible to conduct atlest some of the interaction with
public administration online, i.e. by using EDI or the Internet.

F3
ALL
Now I will read you a list of statements about using online media
for interacting with public administration. Please tell me for each
statement whether you agree completely, agree somewhat or do
not agree.

Public services on the Internet ... [item].
Do you ...
PER ITEM.
(a) are not useful enough
(b) are faster than the traditional way
(c) require thatyou install special equipment or software
(d) reduce the number of mistakes public authorities make
(e) do not seem as safe as using the traditional way
(f) make it possible to deal with the authorities at more
convenient times
(g) make it possible to deal with the authorities at more
convenient locations, e.g. from the workplace
(h) are difficult to use

FOR EACH
(2) agree somewhat
(4) DK
No
Branching

105
Module G: Website accessibility DMS
Transition G
IF C1=1
Now a few questions about the accessibility of your website for
people with special needs.

G1a
IF C1=1
What priority has making your website user friendly for ... [item]
in your establishment?
PER ITEM.
(a) People with visual disabilities or sight difficulties
(b) People with reduced or limited dexterities
(c) People with limited literacy

FOR EACH:
(1) high priority
(2) medium priority
(3) low priority
(4) DK
G1b
IF G1a (a) =2,3
or IF G1a (b) =2,3
or IF G1a (c) =2,3
Bearing the these groups in mind: Would you say that your
website could be adapted rather easily, would prove difficult to
adapt, or could not at all be adapted to these people's needs?

(1) could be adapted rather easily
(2) would prove difficult to adapt
(3) could not at all be adapted
(4) DK
G2
IF G1a (a) =1,2
or IF G1a (b) =1,2
or IF G1a (c) =1,2
Does your establishment or your organisation have formal
Guidelines for making your website accessible to people with
such special needs? By guidelines I mean rules which have to
be followed by your website developers?
(1) yes
(2) no
(3) DK
G3
IF G1a (a) =1,2
or IF G1a (b) =1,2
or IF G1a (c) =1,2)
Was your website ever evaluated concerning its accessibility for
people with such special needs?
(1) yes
(2) no
(3) DK
G4
IF G3=1
Was this evaluation done internally or using external
evaluators?
(1) internal evaluation
(2) using external evaluators
(3) both
(4) DK
Module E: R&D DMS
E1a
ALL
You said before that xyz [PROGR.: Insert answer to question
A5] employees work for your organisation at this establishment.
From this, how many work in research & development, i.e.
R&D? Please add up possible part time R&D personnel to full-
time personnel.
INT.: IF "DK", PROMPT:

INT.: IF NECESSARY, EXPLAIN:
Among R&D we include all creative work undertaken on a
systematic basis in order to increase the stock of knowledge
and the use of this stock of knowledge to devise new
applications.

[OPEN]
numerical
INT.: IF NONE, CODE "0".
[DK]

Progr.: Answer to E1a (Number
employed in R&D) must be Answer to
A5 (Total number employed in
establishment)
If not, re-ask E1a
E1b
IF E1a > 0
and E1a is NOT DK
R&D can be centralised in R&D units, or it can be distributed
over various units of an establishment.
Do you have at least one central R&D unit at your
establishment?
(1) yes
(2) no
(3) DK
No
Branching

106
E2
IF E1b=1
What is the size of the computer staff in your central R&D
unit(s)? Please add up part time computer staff to full-time staff.

INT.: IF NECESSARY, EXPLAIN:
By computer staff we mean all staff that
- manages the computers, networks and digital resources, or
- manages the Internet access and presentation, or
- carries out information searches and computations as their
major work tasks, or
- provides user training.


[OPEN]
numerical
[DK]

Progr.: Answer to E2 (Computer staff in
R&D) must be Answer to E1a (Number
employed in R&D)
IF NOT, re-ask E2
E3
IF E1a > 0
and E1a is NOT DK
Do you get IT services for R&D from internal computer staff that
are not members of your central R&D unit(s)?
(1) yes
(2) no
(3) DK
E4
IF E3=1
What is the size of the internal computer staff outside of your
R&D unit(s) who provide IT services for R&D projects? Please
add up part time computer staff to full-time staff again.

[OPEN]
numerical
[DK]

Progr.: Answer to E4 (Computer staff
outside R&D) must be Answer to A5
(Total number employed in
establishment)
IF NOT, re-ask E4
E5
IF E1a > 0
and E1a is NOT DK
Do you buy IT services for R&D from external service providers? (1) yes
(2) no
(3) DK
E6
IF E5=1
What is the number of additional computer staff in your
establishment that would be necessary to substitute for the IT
services for R&D projects which are currently obtained from
external service providers?

[OPEN]
numerical
[DK]
E7
IF E1a > 0
and E1a is NOT DK
Do your R&D activities suffer from a low supply of qualified
computer staff in your establishment?
(1) yes
(2) no
(3) DK
E8
IF E7=1
Please specify the number of open jobs for computer staff
needed to provide IT services for R&D projects in your
establishment?

[OPEN]
numerical
[INT.: IF NONE, CODE "0".
[DK]
No
Branching

107
X1
ALL
Finally I would like to ask you for a brief assessment:
In the course of the interview we talked, among others, about
the areas e-Commerce, i.e. selling and buying online, and e-
Government, i.e. interacting online with public administration.
That is about areas, which might notnecessarily fall into your
direct responsibility.

Thinking back to the questions about ... [item]: What would you
say: How familiar were you with the topics covered in those
questions? Would you say...
PER ITEM.

(a) e-Commerce, i.e. selling and bying online
(b) e-Government, i.e. interacting online with public
administration

FOR EACH
(1) very familiar
(2) fairly familiar
(3) not very familiar
(4) not at all familiar
(5) DK/ no answer
X2
ALL
And all in all: How interesting did you find the questionnaire as a
whole?
Would you say ...
INT.: READ OUT ANSWER CATEGORIES. SINGLE ANSWER.

End Text
ALL
These were all my questions. I would like to thank you very
much for participating in the interview.
Have a nice day/evening!

Data to be provided by survey organisation Categories
P0 Survey Number 1 0 1 4 3 9
P1
Country Code
|__|__|
P2 Interview Number |__|__|__|__|
P3 Date of Interview: Day |__|__|,
Month |__|__|
P4 Time of the beginning of the interview (USE 24 HOUR CLOCK): Hour |__|__|,
Minute |__|__|
P5 Number of minutes the interview lasted |__|__|__|
P9 Interviewer Number |__|__|__|__|__|__|__|__|

1-Benchmarking Telecommunication and Access in Information Society

Enviado por

Dados do documento

Descrição original:

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

1-Benchmarking Telecommunication and Access in Information Society

Enviado por

Direitos autorais:

Formatos disponíveis

SIBIS

Você também pode gostar