Qualitative Research. 1. Qualitative Research v. other research strategies 2. Qualitative Analysis –Fluid/flexible analytical frames –Analytical induction.

Qualitative Research

Qualitative Research

1. Qualitative Research v. other research strategies

2. Qualitative Analysis– Fluid/flexible analytical frames– Analytical induction– Theoretical sampling

3. Case Studies

4. Qualitative Methods

1. Qualitative Research v. Other Approaches

Common Assumptions (not Ragin)

• Holistic understanding of the world• Cases have to be considered as a whole (in-depth)• Cases are embedded in time and space. Generalisation across cases

is not possible• Specific features have different implications according to the exact

configuration in which they occur • Therefore, isolation of (independent) variables is not possible• Aim is to understand specific individual cases, not to explain general

features• Inductive research methodology

Please note: these assumptions make qualitative research the opposite of quantitative research, working according to a different research logic. Ragin’s account emphasises the common features. He sees them along a continuum rather than as completely different

3 Strategies of Social Research

Trade-off between number of cases and number of features

1. Qualitative: few cases – many features

2. Quantitative: many cases – few features

3. Comparative: balance between 1. and 2.

Data handlingWhat does qualitative research do to the data?

Ex: study of gun enthusiasts

Qualitative Research: enhances dataGoal: exhaustive representation of a casewant to see details behind what is apparentWhy do you collect guns, when did you start, how does it feel to posess

a gun etc.

Quantitative Research: condenses datain order to reveal the bigger picture behind the data get rid of all the particularities of special cases, and look for remaining

pattern (which social group is most inclined to collect guns?)

Comparative Research: elucidates subtle patterns in data in order to examine differences between cases (more case oriented than quanti, less case oriented than quali). Would look for different types of gun enthusiasts

The goals of qualitative research

• Giving Voice

• Interpreting Historically and Culturally Significant Phenomena

• Advancing Theory

Basic interest of qualitative research

(according to Ragin)

• In-depth knowledge of case

• Identify the basic features/characteristics of a case

• Identify common features across different instances of a case

• Often in order to clarify key aspects and to correct common misrepresentations

2. Qualitative Analysis

• General: Start with flexible/fluid analytical frame (in order to engage in “reciprocal clarification” between concepts and images – retroduction

Analytic Induction (deviant case analysis): search for contrary evidence in order to improve images and concepts– reconceptualize (extend the concept) to make new evidence fit, or– narrow the category, thus excluding contrary evidence (from being a

deviant case, it becomes an irrelevant case, because it now falls outside the frame of reference)

Theoretical sampling: – add further cases and compare with original study object (not in order to

capture full variation, but in order to deepen understanding of research subject and to clarify concepts) (<-> random sampling in quantitative research)

Example: Study on professional politicians (British MPs)

• Analytic Induction– Observation: interviews have shown, MPs (category) strive to become government

ministers (concept)– Deviant observation: one MP (who also holds a boardroom position in a private

company) strives for a position in top-management of the private sector

2 possible ways to deal with this problem– Broadening concepts: MPs strive to advance their career (concept)– Narrowing categories: Full-time MPs (category) strive to become government

ministers

• Theoretical Sampling– Observation: interviews have shown, MPs (category) strive to become government

ministers (concept)– Extend interviews to different types of MPs (perhaps regarding personal

background, regional origin etc. – features that might have an influence on career ambitions)

• Qualitative R. : one/few cases!• Ragin‘s examples (e.g. legal assistance lawyers): many

cases• Solution: not really treated as different cases, but as

incidents of the same case• Goal is NOT to find general pattern over many cases,

nor to find variation among cases, but to know more about what one case (group) has in common. Looking at different incidents in order to have full picture of your case (look at new cases until saturation) and to refine concepts and categories.

4. Case Studies (N = 1)(Gerring 2001: 215-222)

• Advantage:– Watching the relationship between X and Y over time and in detail (in its

context), may elucidate causal mechanisms• Disadvantage:

– Lack plenitude (problems of generalisation)

Is n really 1? The term case study might be ambiguous:– Look at one case in detail, yet embedd this case in context of other

cases (often done in the process of case case selection)– Across-case vs within-case variation

• Number of n is always connected to specific causal proposition (there can be more than one in one study)

• Ex: Study of Mau-Mau revolution in Kenia (n = 1)• But, why did it start from region A and not B? (n = number of Kenian regions)• Why did some people participate, others not (n = number of studied

individuals)

Types of Case Studies

• Extreme Case

• Typical Case

• Crucial Case– Deviant Case

• Counterfactual Case

What makes your case significant? By assigning your case study to one of these types, you are relating your case to the universe of other cases, your are setting it into context. Even though you might be doing a case study (n=1), you are “comparing” (going beyond that single case).

Extreme Case

• Definition: – choosing the most extreme case with regard to the outcome in

question• Example:

– Study facism in Germany or Italy, rather than where it was less extreme

– Study substate nationalism/regionalism in Scotland rather than Bavaria

• Adv./Disadv.+ Extreme cases often reveal the essence of a causal relation– Often sacrifice in representativeness– Extreme with regard to which dimension? (is there only one?)

Typical Case• Definition:

– Choose (not the extreme but) the most representative/typical case. A case that exemplifies the mean, median or mode on the dimension in question

• Example:– Robert and Helen Lynd studied Muncie, Indiana as a typical American

community• Adv./Disadv.

– + generalisation of results– - Quite often there is no typical case, as a population consists of quite

different subgroups (remedy: find typical case of each subgroup)– Neither mean nor median are necessarily typical

Note: the representativeness of a case cannot be established within a n=1 study. We can only choose with reference to other sources/general knowledge

Crucial Case

• Definition:– Cases that are critical for a concept or a body of

theory, cases that have come to exemplify a concept

• Example:– France in a study on revolutions

• Adv./Disadv.– If you want to add a new aspect to a theory, best do it

in a well-studied, crucial case, not in some obscure case hardly known

– There might not be a crucial case

Deviant Case(subcategory of crucial case)

• Definition: – case that is disconfirming a particular theory/concept (least-likely case

shown to be positive, most likely case shown to be negative)– Unlike extreme case, which is confirming a theory

• Example:– Country with Plurality Voting System but Multi-Party System

(disconfirming Duverger’s law)• Adv./Disadv.:

– Helps to figure out faults in a theory– There may not be a deviant case, and if, it might be explained easily,

without much impact on the original theory (exception proving the rule)

• Note: Modification, not falsification is the main purpose of crucial/deviant case analysis

Counterfactual Case

• Definition:– Thought experiments (no real case), “what-if”-scenarios– counterfactual case is most similar to real case, but differs in the

one variable that is of interest

• Example:– Proposition: US won WWII because invasion in 1944– Counterfactual case to support this proposition:

imagine WWII without the US invasion of 1944possible result: if US had not invaded in 1944, Germany would have had time to develop its own atomic bomb

• Adv./Disadv.– Not necessarily conclusive, but helpful– Not for purpose of general explanations, but to explain specific

outcomes

4. Qualitative MethodsHow to find/retrieve/create evidence?

• Interviews (in-depth)Narrative (life story, story of event), semi-structuredTranscription (words, emphasis, body language etc.), analysis (different

techniques)• Panel Discussions

Moderated, unmoderatedTranscription, analysis

• (Participant) observationethnographic method. Observe objects and their action in their natural context, often the researcher becomes part of the social world that is studied

• Field Notes record of observations, reflections and thoughts during the research

process• Document analysis

Official documents, letters, diaries, newspapers

Analysis/Interpretation

• Data enhancing (narrative interview v. structured questionnaire)

• Problem: how to condensate, analyse in order to form representation. How to find significant evidence

• Different techniques• General: much more room for researcher and

their interpretation• Retroduction (interplay analytic frame – image)

very important

You should know:

• Advantages / goals of qualitative research

• Characteristics of qualitative research

• Analytical induction

• Different types of cases studies

• Are case studies (n=1) comparative?

• Methods to retrieve evidence/data