Sunday, June 2, 2013

Collecting Empirical Data - Dan Remenyi - Chapter Summary

Ch.8. Collecting Empirical Data

This chapter discusses various approaches to the acquisition of appropriate evidence which will be used to support or refute the theory or hypothesis developed by the researcher.
The chapter further considers the role of evidence in formulating a theory, as happens when using a grounded theory approach.

Why Collect Empirical Evidence?

Creating a Theory
A research project may, on the other hand, require evidence to be collected before a theoretical conjecture can be established and subsequently expressed in terms of hypotheses and empirical generalisations which can be formally tested, perhaps at a later date in another project.  
An important technique for doing this is grounded theory (Glaser and Strauss, 1967) in which the researcher uses empirical evidence to establish directly the variables, concepts and relationships which will be combined in the theory.

Understanding and Explaining Phenomena
From the phenomenological point of view, evidence may also be collected in a less instrumental way in order to understand and explain phenomena.
Here the researcher will have a much more open view as to how the evidence will be treated and what the evidence might reveal.

  Phenomenological evidence collection is considered in more detail in chapter 6.

Approaches to collecting Evidence
Several approaches to the collection of evidence are open to the researcher and the approach chosen will depend upon the research strategy and tactics being followed as well as the research question itself.

Primary and Secondary Sources
Evidence may be collected from primary or secondary sources.
Evidence is collected from a primary source when the researcher goes directly to the originator of the evidence.
An example would be an interview with the managing director about the organisation’s marketing strategy.

A secondary source would be information that is already published or available indirectly, so in the previous example the researcher might be able to obtain essentially the same information from the annual financial statements.
A number of databases containing useful evidence and information for research in business and management studies are available and the Internet and the World Wide Web are rapidly increasing in importance as source of secondary evidence in business and management research.    
                                       
Direct and Remote Evidence Collection
Primary evidence may be collected either directly or remotely.
In direct evidence collection the researcher interviews the informant personally and records the responses directly.
In contrast, remote evidence collection would correspond to a situation in which the informant completes a questionnaire without the interviewer being present.  There are of course intermediate approaches, for example where the researcher interviews the informant on the telephone or engages in a dialogue with him or her by E-mail.

Valid and Reliable Evidence
‘Triangulation’ which Loveridge (1990:18) defines as ‘using multiple methods to capture a sense of reality.’  
In business and management research the term triangulation refers to obtaining evidence from multiple sources to ensure that a biased view is not being obtained from one informant.

Evidence for Formal Testing
The formal testing of hypotheses and empirical generalisations usually requires numerical evidence that will be analysed statistically.
Numerical Evidence
Here the key issues are to identify the evidence required and to prepare an appropriate measuring instrument.

Non-Numerical Evidence
In these cases techniques such as content analysis may be used to convert qualitative evidence into numerical form so that standard statistical tests may be used in the analysis.
Converting qualitative evidence into a numerical form is problematical and relies on assumptions concerning the homogeneity of the responses since counting the frequency with which a particular opinion occurs implies that each occurrence of that opinion is given equal weight.

NUD.IST
A popular computer package for text analysis is QSR NUD.IST (Non numerical Unstructured Data Indexing Searching and Theorising) ‘ a computer package designed to aid users in handling non numerical and unstructured data in qualitative analysis.  One highlights a particular block of text and creates a reference to this block of text under a suitable heading.  The different headings can themselves then be linked conceptual a paradigm sections of text referring to different aspects of the argument.  Phenomenologist generally recommend a more hermeneutic approach to the interpretation of such evidence.

Evidence for Creating a Theory
At this level the evidence-collection process is creative, making it difficult to provide specific and detailed guidelines.  However there are some general principles which may be helpful.

Both primary and secondary sources of evidence should be consulted.
Evidence should be collected from as many different informants as possible.
Informants should where possible represent a spectrum of individuals who may have quite different perspectives on the problem under consideration.  It is important, for example, to involve both top management and relatively junior staff.
All evidence should be corroborated by means of some form of triangulation.

In the process of theory creation the evidence provides the raw material but it is the researcher’s imagination and creative talents which lead to the development and formulation of the theory.

While the evidence is critical, the way in which it is perceived, analysed, synthesised and understood will determine the extent to which it is used effectively.

Understanding and Explaining Phenomena
The non-positivistic or phenomenological research paradigm is discussed in Chapter 6.
Rather, the evidence will be collected in order to achieve a greater level of understanding or to develop an explanation of the observed phenomena.
Evidence Collection ; Planning and Design
Whatever research strategy and tactics are used, the quality of the evidence will be improved if the research is well planned and designed.
How to obtain access? In particular, how will the researcher gain acceptance to/by the organisations required and not be seen as an inconvenience?
How to be introduced to the right people in the organisation?
        Specifically the researcher needs to find out who are the gatekeepers to the required evidence and the departments or sections within the orgnisation.

Does the researcher have to collect the evidence personally or can some of it be collected remotely, i.e. by using questionnaires?
What arrangements can be made for triangulation to ensure the integrity of the evidence?
What sources of secondary evidence can these organisations supply?
Can external triangulation by means of, a trade organisation, a bank or a trade union be obtained?

Access to Informants
Sometimes researchers find it difficult to contact organisations directly.
On such occasions the researcher may try to use an intermediary to arrange an introduction.  Members of staff of the university or business school may sometimes be helpful.  Management consultants may be prepared to collaborate with researchers and thus introduce them to the appropriate organisations.
If the researcher offers to help with an in-house project or offers to provide the informants with  a copy of the results of the research, this may improve access to organisations.
Direct Evidence Collection
The golden rule is never to force the pace by trying to obtain more evidence than is naturally and comfortably offered.

Indirect Evidence Collection
E-mail is a form of asynchronous communication and therefore closer to remote evidence collection.
The main drawback of remote evidence collection is that it is difficult, if not impossible, to probe the informant.


However one of the problems associated with the use of the Web is the fact that there are not yet any really hard and fast rules about how to cite references properly.
As a general approach researchers should provide sufficient information about the Web site to enable another person to access the same information subsequently.

No comments:

Post a Comment