Saturday, October 24, 2015

Sunday, October 18, 2015

Interpretive Research in Information Systems

Philosophical basis of interpretive research


The ethnographic research tradition in anthropology is a valuable starting point for a consideration of the philosophical basis of interpretive case studies.

What is called data in interpretive studies are  constructions of  people of what they and their
compatriots are up to.


Van Maanen (1979), writing in the tradition of organizational ethnography, calls the interviewee's
constructions first-order data and the constructions of the researcher second-order concepts.

Second-order concepts rely on good theory and insightful analysis, and mere collection of in-depth case study data does not provide these concepts in itself. Examples of second-order concepts
in the IS literature, derived from interpretive case studies, include the 'automate' concept from the work of Zuboff (1988), and the concept of 'technological frames' in Orlikowski & Gash (1994).

A second feature of the anthropological tradition is its concern with 'thick description'. The ethnographer is faced with a multiplicity of complex conceptual structures, many of them superimposed upon or knotted into one another and which must be first grasped and then rendered
intelligible to others.

 An IS researcher can only access the subtleties of changing interpretation by the use of approaches based on 'thick' description.

The goal is not to generate truth or social laws, and this interpretive approach can be
clearly distinguished from the positivist tradition. This should not be taken to imply that interpretive work is not generalizable, although the nature of such generalizations is different in the two traditions. This point will be considered in some detail later.


In 'nonpositivism' facts and values are intertwined and hard to disentangle, and both are involved in
scientific knowledge; and 'normativism' which takes the view that scientific knowledge is ideological and inevitably conducive to particular sets of social ends. Either of these  two positions is open for the interpretive researcher to adopt.

 'internal realism' views reality-for-us as an intersubjective construction of the shared human cognitive apparatus, and 'subjective idealism' where each person is considered to construct
his or her own reality. The usual ontological stance for an interpretive IS researcher would involve one of these  two positions, particularly with regard to the human interpretations and meanings associated with computer systems.

 Mingers (1984) identified the existence of at least four substantively
different strands of thought in nonpositivistic research: phenomenology, ethnomethodology,
the philosophy of language, and hermeneutics.

For example, Zuboff (1988) drew on phenomenology, Suchman (1987) on ethnomethodology,
and Boland & Day (1989) and Lee (1994) on hermeneutics.


Reference

Interpretive case studies in IS research: nature and method
G WALSHAM
Department of Management Science, The Management School, Lancaster University, Lancaster LAI 4YX, UK
Eur. J. Inf. Systs. (199S) 4,74-81



Sunday, October 11, 2015

The Case Study Research Method - Dan Remenyi - Chapter Summary




The case study is becoming an important research tactic in business and management research and its use is on the increase for both masters and doctoral dissertations.

The Case Study in Research

In research the case study has two distinct features. Firstly, the case study can be used in establishing valid and reliable evidence.  This evidence may be collected and analysed from either a positivistic or a phenomenological (interpretivist) perspective.  The evidence can be used  to produce a theoretical conjecture (grounded theory method)  or it can be  used as evidence to support or contradict a theoretical conjecture.

The case study can be used as a vehicle for creating a story or narrative description of the situation being studied, in such a way that the resulting narrative represents a research finding in its own right (narrative development was discussed in chapter 7 of the book)



Case Study Research -  Definitions


More formally, a case study may be defined as an empirical enquiry that investigates a contemporary phenomenon within its real life context, when the boundaries between phenomenon and context are not clearly evident, and in which multiple sources of evidence are used.

Case study, is that it tries to illuminate a decision, or a set of decisions: why they were taken, how they were implemented, and with what result.


The aim of the case study to provide a rich, multi-dimensional picture of the situation being studied.

Scope of the Case Study


Scope of the Case Study as an evidence-collection approach

Cases concerning business decisions, processes, or organisational change have the inherent difficulty of knowing where to start and where to start and where to end.

Comparison of cases can lead to the formulation of a theoretical conjecture, or in some circumstances the confirmation of hypotheses or empirical generalisations.

The case study comes into its own when a How?, Who? Or Why? Question is being asked about a contemporary set of events, over which the researcher has little or no control.

Depth of Enquiry of the Case Study


The depth of enquiry possible through the case study method is significantly greater than some other research methods such as surveys, focus groups, in-depth interviews, experiments and analysis of archival evidence.
                                                                                                                             
Prejudices about the Case study Method

Traditionally there have been prejudices against the case study method, especially in business and management research.


The evidence collected from informants will have to have been triangulated by using other evidence sources and this is discussed later in this chapter.

Competent case study research, leading to real understanding and explanation in business and management, is difficult to perform.
One of the main problem is that it is difficult to screen would-be case study researchers to ascertain their ability to do case studies.

Case Studies and Experimental logic

Because the case study follows the logic of an experiment, rather than the logic of a survey, Yin(1989) argues that it is not necessary to replicate a case study many times.

A case study researcher is clearly closer to an experimentalist than a theorist, in that the case study requires the collection of empirical evidence.  The researcher tries to get close to the phenomenon being studied in much the same way as a physical or life scientist does by collecting empirical evidence.
On the other hand, case studies differ from experiments in that case studies try to use passive observation and thus to leave unchanged by the enquiry the phenomenon being studied.

Rich and Complex Evidence
Purposes of theory creation or theoretical conjecture development .

There are at least three difficulties in obtaining unbiased testimonials from observers:
The difficulties encountered by individuals in their being able to recall events accurately;
The difficulty individuals have in disclosing important feelings;
The suspicion individuals have about revealing information that might reflect poorly on themselves or their superiors.

Bias may be minimised by the process of triangulation.
This involves a number of distinct activities.
Firstly, multiple sources of evidence should be sought to support all important assertions.
Secondly, corroborative evidence may be obtained from several different informants in the firm.

Bias may be minimised by the process of triangulation.
This involves a number of distinct activities.
Firstly, multiple sources of evidence should be sought to support all important assertions.
Secondly, corroborative evidence may be obtained from several different informants in the firm.

Finally, on the issue of bias, it is naïve to assert that any form of research, or perhaps human activity generally, is without bias.
As the form of the case study is primarily narrative, accurate description is essential.
All relevant facts need to be included, and circumstances having no bearing on the situation should be omitted.
The Case Study Process

Producing competent case study research is a process that needs to be understood and carefully followed.
Uniformity when Recording Evidence
Thus uniformity of recording should be sought as this facilities comparison between enterprises or situations, which allows the highlighting of similarities and differences.

In fact, the case study research approach requires a distinctly formal approach.  Before the research can proceed a protocol needs to be drawn up.

The protocol is a formal and detailed master plan that specifies full particulars of the research, a summary of the questions to be asked during structured interviews,  field procedures for the researcher, details of all types of evidence required, as well as the structure of the final report.

Case Study Protocol

A protocol is in important instrument for ensuring the reliability of the study.
The protocol is in fact a detailed statement of what the research is trying to achieve, as well as a plan that indicates how the objectives will be met.
The general rule is that the more time and attention which is given to the planning of the research the better the results are likely to be.

Protocol Overview

The protocol overview should include the main objectives and the detailed issues the researchers will focus on.

Field Procedures

Field procedures need to detail a number of issues
including :

• defining who should be interviewed;
• how to access to the right people;
• ensuring resources are available including    time, paper, tape recorders, etc;
• developing a procedure for obtaining assistance     from other researchers;
• making a schedule of the required evidence-collection    activities
• providing for contingencies.

 
Case Study Questions

At the centre of the protocol is a set of questions reflecting the actual inquiry.  Firstly, the protocol questions are set for the researcher and not for the respondent.

At least one informant per firm should be a senior manager.

A friendly gatekeeper or guide should be found as soon as possible.

Documentary evidence should be sought to support the verbal information.

Attempt to interview informants in their office rather than in an interview room.
Secondly, each question should be accompained by a list of probable sources of evidence which cover interviewees comments, documents, artifacts and observations.




Case study Report Guidelines

The following are the major headings that were established as the key focal points of the case study reports.  These were established early in the research process so that they could be used as a supplementary aide-memoire for the author in conducting unstructured interviews with informants.
Introduction and general background of the firm.
The state of IT within the firm.
The reasons for the current move towards strategic information systems (SIS).
The IS strategy:
a What is actually is.
b  How it is formulated in the organisation.
The implementation of the strategic information system
What are the consequential effects of the SIS investment?
What conclusions have been drawn by the organisation themselves?



Collecting the Evidence

For a masters or doctorate degree the case study approach to research requires comprehensive and intensive study of the subject and thoroughness is one of the first requisites.

Different Types of Case Study Evidence

Evidence can be obtained from documents, archival records and interviews as well as from any person who has knowledge of the subject, observations of the researcher, participant-observer interactions as well as physical artefacts.

There are six important sources of evidence used in case studies:

documents;
interviews;
direct observations;
participant-observation situation;
physical artifacts;
archival records.

Documents

Documents are primarily used to corroborate and augment evidence from other sources.
They are helpful in verifying spellings and titles.
For business and management researchers the emphasis may be placed on obtaining copies of proposals, contracts, accounts, personal correspondence between informants, as well as corporate publicity material.
Interviews
If the emphasis of the interview moves substantially towards obtaining insights then the respondent is referred to as an informant.
A focused interview is one in which the informant is interviewed for a short period of time, for example an hour.
Interviews are an essential part of case study evidence.  However, they are verbal reports only and as such are subject to the problems of bias as well as poor and inaccurate articulation and listening.

Corroboration may be verbal, by other members of the firm, by documents, by observation or by the firms suppliers or clients.
By conducting a personal interview the researcher will probably have visited the organisation and will perhaps have been shown around the premises.
This type of visual contact is also useful for the purposes of triangulation.

Direct Observation

All evidence other than observation is essentially hearsay and its reliability needs to be examined.  Observation is thus one of the most valuable ways of collecting reliable evidence.
The researcher may, for example, observe locations, individual behavior, behavior at meetings, morale, dress codes, and corporate culture, to mention a few examples.
Observed evidence is clearly stronger than hearsay evidence.
Observations may be so important that it is necessary to take photographs or to make a video of the case study site .
Participant-Observation Situation

However, it is not always easy for a researcher at the masters or even at doctoral level to be given this level of access to an organisation.

Physical Artefacts

Physical artefacts include books, technological devices, tools, instruments, and ledgers.
A computer printout may be considered an artefact.
The presence of books on MS-DOS 2, for example, would be evidence of the organisation having used personal computers for more than 15 years.
 
Archival Records

Archival records include staff service and payroll records, old correspondence, old product or service descriptions and so on.
Often archival evidence is highly quantitative.  
The same general caution about the original purpose of the information applies to archival records as well as to documents in general.

Principles for Good Practice

Three principles of evidence collection.
Use multiple sources of evidence;
Create a case study database;
Maintain a chain of evidence.

Use Multiple Sources of Evidence

An example of triangulated evidence is obtaining a view about the corporate strategy from the managing director, the workers in the factory and from a stock exchange broker or the organisation’s bank manager.
The evidence may then be corroborated by reference to annual accounts and also to articles published in the business press.

Multiple sources of evidence tend to help with the problem of construct validity because this provides several measures  of the same phenomenon.
The multiple nature of evidence collection also allows the researcher to attempt to find information convergence.


Create a Case Study Database

A case study database is a critical part of the evidence supporting the case study research strategy.
There are two aspects to the case study database : the evidence collected and the reports written by the researcher.
A researcher creates a formal, retrievable database of the evidence collected so that other researchers can easily review the original material.  The main item in the database will usually be the case study notes.  These notes should be organised and categorised and as complete as possible.  They should be available in the original medium i.e. in writing, as well as supported by tape recordings if these were used.


A large number of documents will be collected and the researcher will find it useful to organise the filing of these into primary and secondary systems.  It is also essential to cross-reference the documents for subsequent retrieval.  Large amounts of numerical data could be included in the database, which should be held on disk or tape.
As a by-product of the case study database, some of the summary material written by the researcher will become part of the text of the dissertation.

Maintain a Chain of Evidence

The principle of maintaining the chain of evidence states that an external observer or reader of the case study should be able to follow the argument and the derivation of the evidence from the original research design and questions to the eventual conclusions.  The observer should be able to trace the argument both forwards and backwards.  The chain of evidence principle also states that no evidence should be lost or omitted.  If these rules are followed then the case study will have addressed the methodological issues of determining construct validity and the overall quality of the case will have been enhanced.

How to Judge Case Study Design

From a positivist point of view, case study design may be judged on the basis of four tests, which are construct validity, internal validity, external validity and reliability.

Construct Validity

Construct validity refers to establishing correct operational measures for the concepts, ideas and relationships being studied.  More formally, construct validation is a scale evaluation criterion that relates to the following question :  ‘What is the nature of the underlying variable or construct measured by the scale?’

To meet  the test of construct validity the researcher should be sure to have covered two steps:
Carefully identify ideas, concepts, relationships and issues which are to be studied;
Demonstrate that the selected measures to be used in the research actually address the ideas, concepts, relationships and issues being studied.

For example, if the research aims to determine the extent of awareness of Porter’s Generic Strategies model in those making strategic information systems decisions, it is necessary to demonstrate that the questions will actually measure this.  Three tactics are available to increase construct validity.
Triangulation;
Establish a chain of evidence to show how each link in the chain relates to the  next;
Draft a case study reviewed by key informants who have detailed knowledge of the relevant ideas, concepts and relationships.


Construct validity is an important issue because criticisms are often made of the validity of research in business and management studies.
This is indeed a fair comment and is one of the reasons why rigorous research in business and management studies is so challenging.    
  
Internal Validity

Internal validity is demonstrated by sound argument, as has been discussed in Chapter 3, and this is a fundamental pillar on which the validity business and management research is based.

External Validity

External validity is concerned with knowing whether the researcher’s findings are generalisable to a wider universe beyond the immediate research environment.  For positivists this is a central issue, while phenomenologist are less concerned with external validity and more concerned as to whether the research is authentic and properly represents the events being studied.

Whereas case studies rely on in-depth evidence that is evaluated on the basis of analytical generalisations.  In analytical generalisation, the researcher is striving to associate a particular set  of results to some broader theory, and thus the sample size is not such a relevant issue.

The logic of replication in case studies is the same as that which underlies replication in experiments.  Most business and management researchers who use the case study approach would not attempt to argue that their findings are generalisable in the same sense as a physical or life scientist would, or for that matter as a social scientist who had conducted a  large-scale survey would.

Reliability

Research reliability refers to the issue of whether the evidence and the measures used are consistent and stable.  This is especially important if the findings of the research are to be applicable to other situations and not only to the original environment or environments in which the research was conducted.

A positivist using multiple case studies will claim reliability but a phenomenologist will not regard this issue as pertinent.  To the positivist the purpose of ensuring reliability is to reduce or minimise errors and bias in conducting the study.

Single versus multiple Case Study Design

Sometime a single case study is sufficient while on other occasions multiple case studies are preferred or even necessary.  There are an increasing number of dissertations being concluded on the basis of a few case studies.  The idea that a large number of observations of a phenomenon is not necessary is well expressed by Gummesson (1988) when he said:
It no longer seems so ‘obvious’ that a limited number of observations cannot be used as a basis for generalization.  Nor does it appear to be ‘obvious’ any longer that properly devised statistical studies based on large numbers of observations will lead to meaningful generalisations.

Single Case Study Design

A single case may be regarded as analogous to a single experiment.  Thus a single case study may be sufficient when a well-formulated theory is to be tested.   Then a single case could be used to confirm, challenge, or extend the theory.  Business and management studies has been slow to accept a single case study as representing sufficient evidence at doctoral level.

A single case study would have to draw on the principles of ethnography as discussed in Chapter 3.  It would be important for the researcher to spend a considerable amount of time with informants in the organisation being studied and it would be essential for there to be proof of careful triangulation of the evidence collected.  None the less, in most circumstances a single case study approach should be regarded as high risk by a business and management researcher.

Evidence from multiple case studies is more compelling and the results are more robust.

Multiple case studies should be considered in the same light as multiple experiments in that they follow the logic of replication.  In the hands of a positivist, the sampling logic requires identifying the target population followed by the use of statistical sampling procedure to select an appropriate subset to be surveyed.  This approach is applicable when the frequency or prevalence of a particular phenomenon is being measured.  The result of the survey assumed to reflect the characteristics of the universe from which it was drawn.  Inferential statistics are then used to establish confidence intervals that determine reasonable bounds for the particular findings.

This means that a large sample of case studies, which is not in keeping with the case study philosophy, would be required to allow for statistical inference.

If all case studies produce the same result then there is compelling evidence for the initial set of hypotheses.
Multiple case study design is now very common for business and management studies.  Of course the question which continually plagues researchers is how many case studies need to be undertaken and there is no simple answer to this.    For doctorate degrees, sometimes five to ten are required, but as mentioned above  there are an increasing number of instances where one case study has been deemed sufficient.

Flexibility within Case Studies

In fact, it is perfectly adequate for a masters or a  doctoral dissertation to assert that the hypotheses or empirical generalisations are rejected.
The flexibility of the case study approach refers to the ability to change cases if they are found to be inappropriate.  This is analogous to changing an experiment if it is found to be unsuitable to prove the hypothesis.  However, changing the theory is quite a different matter and to do this means restarting the research project from close to the beginning.
Case Studies in Use
As mentioned earlier, case studies may be used as part of a grounded theory approach or they may be used to validate an already established theory.

Case Studies and Grounded Theory

When using case studies as part of a grounded theory approach the researcher is attempting to find empirical evidence from which to develop a theoretical conjecture.  A process of induction described in Chapter 5 underpins this procedure.


Validating Established Theory

When case studies are used to validate an already established theory, the first step is to articulate the theory that is to be tested.  The next step is to select cases and to design an evidence collection protocol.  Then the case studies are conducted and individual case reports written.  Each report should show how each proposition was demonstrated.  Cross-case conclusions may be drawn.

This framework should state the conditions under which a particular phenomenon is likely to be found, as well as the conditions under which it is unlikely to occur.  This framework becomes the vehicle for generalisation to new cases in the same way that cross-experimental designs are used.

According to Yin(1989) there are nine basic steps in the case study approach to research :

Develop theory;
Select cases;
Design evidence collection protocol;
Conduct case studies;
Write case reports;
Draw cross-case conclusions;
Modify theory;
Develop policy implications;
Write cross-case report




The skills required for collecting evidence for a case study are much more demanding than for experiments and surveys and include the ability to ask suitable questions, the ability to listen, being adaptive and flexible, having a firm grasp of the subject, being unbiased and not having preconceived notions.

Quantative and Qualitative Case Studies
As has already been stated several times in this book, it is misunderstanding to assert that the case study method of research is by its nature qualitative.  In many instances purely descriptive evidence may be converted into quantitative evidence to which statistical techniques can be applied.

The Case Study as a Narrative or Story

The case study also plays an important role as a knowledge generation approach in its own right.
The result of a case study is often written or told as a story.  Stories have from time immemorial been the repository of knowledge and the main vehicle for transmitting knowledge from one generation to another throughout the world.

But in what way can stories, especially as represented by the case study, be regarded as research?
The telling of a story requires the presentation of the facts or the evidence, in such a way that it is intelligible and of course engaging to the listener or the reader.  This requires the writer/story-teller to process the evidence and structure it in such a way that a convincing proposition is established.  The listener or reader is then offered an explanation of how the issues are resolved.  In effect this is a direct equivalent to the research process.

The initial proposition of the story involves the definition of the ideas, variables or concepts and the relationships between them.  Then the story presents the situation in which some event will take place, which may be seen as the equivalent in the research process of the theoretical conjecture.
The way the propositions and situation develop in the story and the way they are challenged are similar to the testing of the hypotheses; the resolution of the challenges described in the story is equivalent to producing the findings.

Thus it is clear that the telling of a story has a structure to that of the research process and thus story-telling may be regarded, at least in some cases, as the creation knowledge.
The story needs to make sense or put another way, it needs to resonate with the listener or reader.
Thus the story as described here is quintessentially a phenomenological research tool.

In business and management studies the question of whether the story is a knowledge creation activity needs to be addressed in terms of the arguments expressed in Chapter 2 concerning the dialectic.  If the story or the case study is useful and it contributes to an understanding of the world or explains interesting phenomena, then the case study’s or story’s value will be acknowledged and it will become an integral part of society’s knowledge base.

Business and management studies are replete with such story-telling.  Townsend(1984), and Harvey-Jones(1988) are two of the better known narrators of their business experiences who have published their stories in full-length books.

 In addition to this there is a plethora of short case studies written in academic journals.

In simple terms, case studies or stories that are useful contribute to knowledge, while those that are not useful do not.  The question that then arises is, ‘to what extent has a useful case study or story to be true?
However, as an approach to research and the methodological options available, this type of story could not really be regarded as empirical but rather a theoretical treatise.  None the less case study research used in this way offers many opportunities to the business and management researcher especially when addressing the more complex aspects of the field of study.

What Makes an Exemplary Case Study?

There are five general characteristics required for exemplary case study research :
It should be significant;
It should be complete;
It should consider alternative perspectives;
It should display sufficient evidence;
It should be composed in an engaging manner.

In the business and management research context a significant case study is one which is of general interest to business and management professionals, i.e. the stakeholders as described earlier in this book.
For a case study to be regarded as complete the boundaries of the research need to have been given considerable attention.
This means there has to be a careful and rigorous definition of the research problem as well as the determination of the research questions.

Exhaustive effort should have been expended in collecting the relevant evidence and for an exemplary case study the evidence should have been considered from as many perspectives as possible.

 This involves extensive triangulation and the main individuals described in the case setting should be disclosed.

It is useful to have case studies reviewed by peers and by participants in the firms being studied.  If the comments are helpful then they could be published as part of the report.  This procedure is a way of corroborating facts and evidence.  The feedback from the firm will add to the case study’s construct validity.  However, these reviews can produce considerable delays, and the time required obtaining feedback should be planned for at the outset of the research programme.  When composed in this way the case study can make engaging and convincing reading.

Dan Remenyi indicated in one of the presentations. "Read at least two dissertations using your proposed research tactic and number of papers"


More on Case Study Research Method



Guidelines for conducting Positivist Case Studies in Information Systems
http://journal.acs.org.au/index.php/ajis/article/view/448/406
(Two case studies one by Darke and second by Sarker and Lee are given in this paper.

Using A Positivist Case Study Methodology to Build and Test Theories in Information Systems: Illustrations from Four Exemplary Case Studies
http://expertise.hec.ca/gresi/wp-content/uploads/2013/02/cahier0109.pdf



Interpretive Case Studies - Nature and Method by Walsham
http://www.palgrave-journals.com/ejis/journal/v4/n2/pdf/ejis19959a.pdf



Positioning the Case to Tell a Story - Developing the Narrative
http://epubs.scu.edu.au/cgi/viewcontent.cgi?article=1165&context=tlc_pubs




Case Study Research in Business and Management - Bibliography

Case Study Research - Theses, Dissertations and Papers


Harvey-Jones, Sir J. (1988) Making It Happen-Reflections on Leadership. Fontana Collines, London.