Assignment: Small-Scale Qualitative Research Project
ashe.a
Fivelevelsofquestions.docxYin, R. K. (2018). Case study research and applications: Design and methods (6th ed.). Thousand Oaks, CA: Sage.
As the second critical feature, the content of Section C should not confuse five different levels of questions:
· Level 1: questions verbalized to specific interviewees;
· Level 2: questions about each case, which represent your line of inquiry, as just discussed;
· Level 3: questions asked of the pattern of findings across multiple cases;
· Level 4: questions asked of an entire study—calling on information beyond the case study evidence and including other literature or published data that may have been reviewed; and
· Level 5: normative questions about policy recommendations and conclusions, going beyond the narrow scope of the study.
Of these five levels, Section C of the protocol should concentrate on Level 2.
The difference between Level 1 and Level 2 questions is highly significant. The two types of questions are most commonly confused because case study researchers think that their questions of inquiry (Level 2) are synonymous with the specific questions they will emote to the interviewees in the field (Level 1).
To disentangle these two levels in your own mind, think about a clinician. Based on previous experience, the clinician may silently entertain ideas about the course of events in an illness (Level 2), but the actual questions that the clinician poses to the patient (Level 1) do not directly reflect the clinician’s conjectures. The clinician’s verbal line of inquiry differs from the mental line of inquiry, and this is the difference between Level 1 and Level 2 questions. For the case study protocol, accurately articulating the Level 2 questions in Section C is therefore of much greater importance than any attempt to identify the Level 1 questions.
In the field, retaining the Level 2 questions in the back of your mind, while simultaneously articulating Level 1 questions in conversing with an interviewee, is not easy. In a like manner, you can lose sight of your Level 2 questions even when examining a detailed document that will become part of the case study evidence (the common revelation occurs when you ask yourself, “Why am I reading this document?”). To overcome these problems, successful participation in the earlier training helps. Remember that being a “senior” investigator means maintaining a working knowledge of the entire case study inquiry. The (Level 2) questions in the case study protocol embody this inquiry.
The other levels also should be understood clearly. A cross-case question for a multiple-case study of organizational units, for instance (Level 3), may be whether the larger organizational units among your multiple cases are more responsive than the smaller ones, or whether complex bureaucratic structures make the larger ones more cumbersome and less responsive. However, this Level 3 question should not be part of the protocol for collecting data from the single case, because the single case only can address the responsiveness of a single organizational unit. The Level 3 question can only be addressed after the data from all the single-case studies (in a multiple-case study) have been examined. Thus, only the multiple-case analysis can cover Level 3 questions. Similarly, the questions at Levels 4 and 5 go well beyond the empirical data from the full case study, and you should be aware of this limitation if you include such questions in the case study protocol (they will most likely fit somewhere in Section A of the protocol). Remember: The protocol is for the data collection from a single case (even when part of a multiple-case study) and is not intended to serve the entire project.
Undesired confusion between unit of analysis and unit of data collection.
Related to the distinction between Level 1 and Level 2 questions, a more subtle and serious problem can arise in articulating Section C’s questions. They should cater to the unit of analysis of the case study (the “case”), which may be at a different level from the unit of data collection of the case study (a particular source of evidence about the case). Confusion will occur if, under these circumstances, the data collection process leads to an (undesirable) distortion of the unit of analysis.
The common distortion begins because the data collection sources may be individual people (e.g., interviews with individuals), whereas your unit of analysis (the “case”) may be a collective (e.g., the organization to which the individual belongs)—a frequent design when a case study is about an organization, community, or social group. Even though your data collection may have to rely heavily on information from individual interviewees, your conclusions cannot be based entirely on the interviews as a source of information (your case study would have transformed into an open-ended survey, not a case study). In this example, Section C’s protocol questions need to be about the organization, not the individuals. The second row in Figure 3.5 covers such an organizational case study, indicating the kind of evidence that might be obtained from either individual interviewees (Cell 1) or the organization’s policy records and documentable outcomes (Cell 2).
However, the reverse situation also can be true. Your case study may be about an individual, and the sources of information can include archival evidence (e.g., personnel files or student records) from an organizational source (Cell 3). In this situation, you also would want to avoid basing your conclusions about the individual on the organizational sources of information only. In this example, Section C’s protocol questions therefore need to be about the individual, not the organization. The first row in Figure 3.5 covers such a case study about an individual person.
Figure 3.5 Design Versus Data Collection: Different Units of Analysis
