Escolar Documentos
Profissional Documentos
Cultura Documentos
Lecture aim
to provide students with an outline of the process of data gathering from the preparation and organization stage to the actual carrying out of the research fieldwork.
Lecture Outcomes
To outline the importance of the methodological choice for the research topic To identify and explain the various suitable for research topics To outline the merits and demerits of various methodologies To gain an appreciate of how to execute various research methods
Priming Exercise
Jot down how you might research your objectives. Do they involve primary data collection?
How might you research the objectives?
Research Design
Why this research design?
E.g. cohorts is a small numbers phenomenon
Respondents (in)appropriate? Units of analysis? Operational measurement (confounding?) Sparse reporting, scale validity Include design controls (e.g. firm size)
Access!
Adopted from Tashakkori and Teddlie (1998), Greene et al. (1989), and Howe and14 Eisenhart (1990)
Two Phases (i) Network analysis (i.e. issue crawler) (ii) Rhetorical strategies (IT artefacts and 28 indepth interviews). In stage one we employed a network sampling method to locate and map authoritative sites on the World Wide Web, which are involved issue debates concerning the case company under investigation Tesco.
To do this, we used a software tool called IssueCrawler (http://issuecrawler.net) to identify the authoritative actor network websites associated with the firms activism on the Web (this methodology has been used in political science projects, see for example http://www.govcom.org). Significantly, not only does it seek relevance as defined by the issue but it also made on the basis of network representation (Rogers, 2002; Rogers and Marres, 2002; Rogers and Ben-David, 2008). This sweep resulted in IssueCrawler yielding fifty one websites, using the prescribed URL. This allowed the researcher to produce an issue network of sufficient scope and transdiscursivity (Rogers and Marres, 2002).
Research Designs
Quantitative Methods
Quantitative Methods
A precise level of measurement The greater the range of analytical techniques Impact of data coding on subsequent analyses Existing measurement scales pretested items in for example the Handbook of Marketing Scales. Testing of hypotheses or predictions Relationships between variables
Exogenous dynamics Endogenous dynamics
Stata
R Listeral
1) Experiments
Experimental
Limitations
2) Survey
Analytic Survey?
Descriptive Survey?
Identify the independent, dependent and other variables Eg relationship between accounting controls and business strategy
Identify the phenomena that you wish to describe eg what motivates employees in a particular context?
Determine the sampling strategy by defining the research problem And designing a means of accessing a (representative) sample. Is the data to be collected through one approach to respondents?
..continued:
Alternatively the use of surveys to explore a substantive area, or Descriptive surveys Inductive approach Use of open-ended questions Merges the survey approach with styles of research which are more ethnographic Much survey design may begin with an unstructured investigation using overtly ethnographic methods Theory developed inductively to be tested later using a more structured questionnaire as a part of the main study
If the population is small, there is no requirement to sample Probability or representative sampling: mostly
used in survey based strategies when trying to estimate the characteristics of the population
Approach to sampling
For a positivistic study, sampling is a fundamental element For a social contructionist study, sampling is less important. You may well only gather data from one person, such as the chairman of the company
Simple random
Stratified random
quota
snowball
Conven -ience
cluster systematic
purposive
Selfselection
Probability sampling
Seeking a REPRESENTATIVE sample Select the sampling frame Decide on a suitable sample size Statistical inference and margin of error
Purposive sampling -where you use your judgement Snowball sampling-technique used when it is difficult to gain
access to individuals
Sample size Type of questions Wording of intelligible and unambiguous questions Importance of pilot testing Instructions and design of questionnaire Any accompanying letter Method of distribution and return Tests for reliability and validity Collation and analysis of data Any action for non-response
Methods of Distribution
Self-administered
Internet-mediated (Surveymonkey) etc Postal Delivery and collection
Interviewer-administered
Telephone Structured interview
Advantages
On-line questionnaire Large geog area Easy to administer Can use internet tools
Disadvantages
Only include IT literate respondents Avoid complex questions Low/moderate response rate Time consuming Expensive
Similar advantages to postal Collection increases response rate Wide geog area Sample selection easy Respondents have own time Anonymity preserved Token incentives increase responses
Postal
Low response rates Self selection can lead to bias Clarification of questions not provided Non-respondents need chasing Expensive
Eg Research question= To establish existing employees attitudes towards the introduction of a no-smoking policy in the workplace.
Investigative questions
How do employees feel about the introduction. (attitude) Do employees wish to see a smoking room set aside for smokers? Do employees views differ according to Age? Whether a smoker?
Variables required
Attitude of employee as to their rights Attitude of employee to such a provision
Example of coding: Saunders et al (2009) Accountants as sources of business advice for small firms: questionnaire administered by telephone from research by
Gooderham in Norway
Variable
Degree to which a small firm uses its authorised accountant as business advisor Long-term relationship with accountant Perceived competence of accountant in statutory accounting services Firm size (No. of employees) Degree of interest in authorised accountants attempts to sell them advisory services Upper hourly billing rate the firm is willing to pay for business advisory services Degree of competition
Codes
Scale from 1 to 6 1=not at all, 6=to a very large degree 1=changed accountant within last 5 years 2=have not changed accountant within last 5 years Scale from 1 to 6 1=very limited competence 6=very high competence 1=1employee,3=2-4 employees, 7=5-9 employees, 15=10-20 employees Scale from 1 to 6 1=very little interest 6=very large interest 1=do not know or under NKr 300, 2= NKr300-450,3=NKr451-650,4=NKr651ormore Scale from 1 to 6 1=negligible,6=very tough
1/ Deducing a hypothesis ( a testable proposition between two or more events or concepts) from the theory. (H1,..H2) 2/ Expressing the hypothesis in operational terms (indicate exactly how the variables are to be measured) Show how the variables relate. 3/ Test the operational hypothesis (e.g. experiment)
If theory is true
X will occur
Test X
A hypothesis states that there is a relationship between two concepts and specifies the direction of that relationship.
Age
Key terms explained The elements in the boxes are called concepts. The lines between the boxes are called relationships. Theories are composed of concepts linked by relationships.
Variables
A positive sign shows a positive relationship, e.g. (hours of study rises, exam grades rise)
A negative sign shows a negative relationship, e.g. (Price of houses rises, demand falls)
Variables
A variable is a characteristic which has more than one category or value. e.g. Age The effect is called a dependent variable (Y); The assumed cause is called an independent variable (X) An intervening variable (Z) is the means by which X affects Y..
Research approaches
You need to be able to operationalise variables i.e. to make them measurable and quantifiable. e.g Professionalism product quality or Job satisfaction cannot be measured directly.
53
There can be more than one indicator or variable or value: CONCEPTS: Poverty INDICATORS: poor living conditions VARIABLES: provision of sanitary facilities VALUES: Numbers of people per bathroom, WC .
Example 1
Example 2
Quantitative Design
Testable Model
Hypotheses
Hypothesis Exogenous variable Endogenous variable
Productive tension Productive tension Unproductive Tension Unproductive Tension Productive tension Unproductive Tension
Functional conflict Dysfunctional conflict Functional conflict Dysfunctional conflict Interorganizational performance Interorganizational performance
57
Example of outputs
Example of outputs
The standardized direct relations between PT, UT and FC, DC, PER
64
Results
Parameter FC <--- PT DC <--- UT PER <--- PT PER <--- UT PER <--- ALIAGEO DC <--- PT FC <--- UT PER <--- MRKSHRL FC <--- ALIAGEO DC <--- ALIAGEO Estimate .288 -.270 -.283 -.283 -.076 .210 .073 .052 -.084 .008 Lower .117 -.490 -.480 -.549 -.231 -.019 -.118 -.069 -.209 -.132 Upper .455 -.059 -.091 -.055 .132 .422 .265 .250 .046 .161 P .009 .029 .013 .028 .620 .132 .578 .444 .252 .840
65
Result
66
67
Results
Research objectives, theory and findings must be congruent? Findings overly descriptive? Findings overly narrow additional robust tests for different types of relationships?
Results?
Exercise
Sketch out your research design for data collection. Are you exploring or examining? If you are examining, jot down the sources of your measures for your research
What are you control measures?
If you are exploring, jot down the boundary conditions of your research?
Reflection Questions
How will the proposed methodology achieve research objectives What other research methodologies might be employed? Why were these methodologies rejected? How do you go about assembling the data collected during your fieldwork? How to recognise any gaps in the data collection? What is the overall object when carrying out data analysis?
Reflection Questions
How do you consider the practical and operational aspects of gathering data for your dissertation and prepare an outline. How can you anticipate and plan for the possible barriers to the completion of your fieldwork? What do you do when co-operation is not forthcoming from key respondents?
References
Gill, J. and Johnson, P. (2010) Research Methods for Managers, London, Paul Chapman Collis, J and Hussey, R. (2009) Business Research a practical guide for undergraduates and postgraduate students, London, Macmillian Kahn, R. and Cannell, C. (1957) The dynamics of interviewing, New York, Wiley Saunders M. et al (2009) Research Methods for Business Students, Harlow, Prentice Hall Grant, A. M., & Wall, T. D. 2009. The neglected science and art of quasi-experimentation: Why-to, when-to, and how-to advice for organizational researchers. Organizational Research Methods, 12: 653-686.
References
Bell J (1999) Doing your research project (3rd Edn) Buckingham:Oxford Univ Press Bryman A and Bell E (2003) Business Research Methods New York:Oxford Univ Press Easterby-Smith M et al(2002) Management Research: An Introduction London:Sage Publications Robson C.(2002) Real World Research (2nd Edn) Oxford:Blackwell Saunders M. et al (2009) Research Methods for Business Students, (Edn 4) Harlow, Prentice Hall Sekaran V (2000) Research Methods for Business: A skills building approach (3rd Edn) New York:Wiley Ticehurst GW and Veal AJ (2000) Business Research methods: a managerial approach NSW: Pearson Education