Deck 10 Flashcards
1
Q
What is a case study?
A
- A case study isn’t really a study design. A case study is a case within which you study. You do a case study when you want to get a complete understanding of the phenomenon as it presents itself in this case but you don’t trust your theory fully. You start with an interest in a phenomenon, then you select a case, and then you study it thoroughly. A case is kind of like the sampling frame in relationship to a population. If your case is bad cannot really interpret what you learn about the phenomenon. Data sources in your case study are population of observation units within your research unit of the case.
2
Q
What is a phenomenon?
A
- A phenomenon is anything that could be of interest to you:
- Process (e.g. the bleaching of coral reefs.)
- Historical outcome (e.g. like civilization collapse)
- Anomaly (eg. reefs that survive increased temperatures/ acidity.)
3
Q
What is the objective of a case study?
A
- The objective is to understand something that is too complex or unknown to support the use of pre - determined (reductive) strategies.
4
Q
What is a case?
A
- A case is bounded in time and place. It is where the phenomenon of interest is useful (serves your purpose) to study. You need to be very clear when designing your case study what the phenomenon is, because when you get into your case you can get lost. Case studies are intensive studies of a few cases under high uncertainty, so you typically want to look at it from different angles.
5
Q
What are the characteristics of a case study?
A
- The idea is that you restrict yourself a lot by just studying one or a few cases and statistical generalisation is out of the window, but at least you can now focus all your attention on just those few cases. Make it worth your time and start looking beyond the surface, from different angles, using multiple methods, consulting multiple data sources:
- A full understanding is to be achieved.
- Holistic approach as opposed to reductionist (partial).
- One case or a few cases is/ are investigated deeply.
- Multiple sources of data are typically used.
- Multiple methods are typically used.
6
Q
What is reductionism vs. holism?
A
- Reductionism is a representation through values of a known relevant subset of features, whereas, holism is an exhaustive identification and account of all features. Relevance of features is not yet known so there is no basis for selection.
7
Q
What are the steps in sampling in case studies?
A
- Definition of the theoretical and operational population of cases.
- Specification of frame and selection of cases.
- Selection and definition of theoretical and operational population of observation units within cases.
- Specification of a frame and selection of units within each population within each case.
8
Q
What are the considerations that you can a use to justify selection of a case?
A
- Typicality: extent to which a case is (ab)normal on a parameter of interest (e.g. centrality on outcomes, a condition etc.)
- Intensity: phenomenon density on a parameter of interest (e.g. frequency, heterogeneity)
- Stage: where on the empirical cycle the use of the case falls (e.g. induction, testing).
- Instrumentality: extent to which generalizability is relevant.
9
Q
What is single vs. multiple case studies?
A
- In a single case study, only one case is investigated, multiple case study covers several cases.
10
Q
What are the types of multiple case study?
A
- Parallel all cases are selected analytically prior to study. => More appropriate for theory testing.
- Sequential: selection of next case influenced by results of previous cases. => More appropriate for theory building,
11
Q
What are the logics of selecting multiple cases?
A
- Heterogeneity: extent to which cases cover anticipated variability on some analytically relevant dimensions.
- Iterativity: extent to which you reference one case in selecting the next. (sequential multiple case study)
- Stages: extent to which you reference one case in moving to other stages on the empirical cycle.
12
Q
What is parallel case selection in multiple case study?
A
- All cases are selected prior. First, individual cases are studied. Then, comparison between cases is made. A theory may say that effects or relevant mechanisms are different depending on a single factor. To explore this you may study multiple cases in parallel, covering the different valuesof the distinguishing factor. You may also use parallel case selection in theory - generating e.g. as a way to maximize variation:
- when testing a theory you generally try for cases that have maximum differences on interesting analytically relevant characteristics and minimum differences on others.
13
Q
What is sequential case selection in multiple case study?
A
- The choice of the second case is based on the results of the first case. Further cases to expand on earlier findings or refine or modify preliminary conclusions using dimensions of interest (either specified in advance or discovered as as you go along)
14
Q
What is triangulation?
A
- ” Looking at the object from three angles”. More than one method/ case/ source/ researcher are used to check the results and to become more confident about the results.
15
Q
What is the purpose of using multiple researchers in triangulation (case study)?
A
- To counter biased measurement + interpretation made by a single researcher and assess inter - researcher reliability by using multiple observers/ raters/ coders.
