January 31: Brainstorming ideas for research topic and questions
For February 7: Cara will post outline for Ch1&2 - finalize topic
For February 14: Zinnia will post outline for Ch 3&4
For February 21: Dylan will post outline for Ch 5&6
For February 28: Prensentation design

I encourage you to attempt to align the topics you read about with the topics we are covering in a given week (see the weekly table in the syllabus). (That won’t be possible in all cases, but it’s worth an attempt.)

Please find a balance between challenging yourself to meaningfully engage with substantive

portions of the text and overwhelming yourself. I want you to walk away with a clear idea of the

approach you are studying, but you need not read every page of the book. You are building a

strong foundation, not placing “the final brick.”

Chapter 2 deals with designing case studies, though the author points out that there is no standard “criteria” for this research method as there are for other, more established research methods. In case studies, it is important to identify a research design (qualitative, quantitative, etc.) and the theoretical framework behind it. Although “quality indicators” do not necessarily exist for case study research, there are several components that should be considered and included in case studies in order to increase the quality of a study: 1) construct validity; 2) internal validity; 3) external validity; and 4) reliability. The specifics of these components, of course, depend on the specific research method or theoretical model, but they are integral to valid case study, just as they are for other research methods. A case study has five components: 1) the research question (“how” or “why”); 2) its proposition (correctly identifying the approach or direction of how to examine the phenomenon); 3) its case (defining the problem and keeping identifying what is “part” of the problem and what is “outside” of the problem); 4) linking data to propositions (how is the researcher actually measuring the data that the proposition is purporting to measure); and 5) establishing criteria for interpreting the strength of a case studies findings. One of the last issues discussed in this chapter were single- vs. multiple-case design. The author encourages multiple-case design, or even two-case design, in order to decrease the vulnerability that comes with single-case design studies. Multiple-case design allows for direct replication and contrasting situations. 

Chapter 3  highlights the preparation stage of case study research. This chapter highights the skills needed to be a case study researcher and training needed for specific case study. Yin (2017) also highlights how to develop a case study protocol. There is also a walkthrough of getting human subjects protection approval, how to screen candidates and select final cases, and conducting pilot case studies. 

     The desired skills of a case study researcher are asking good questions, being a good listener, staying adaptive, having a firm grasp of issues being studied, and ethically conducting research. Preparation and training for specific case study include protecting human participants and ensuring ethical care of their well being and the information that is acquired through the case study. The training required to do the case study the timing is important as often the case study is completed within the natural environment. The training process begins with definitions of the research questions. 

     Training to be a “senior” researcher requires key steps because often a case study often has a team for the following reasons a case study often requires intensive data collection at the same site requiring a team of researchers, case studies might include multiple cases with different people covering each site, or there is a combination of the two. This chapter offers a few examples of these that are easy to follow and helpful to use as a reference.

 The case study protocol is related to a survey in that both are directed at a single focus for data collection the single case or single respondent. The case study protocol has four sections a) an overview of the case study which states the objective, case study questions, theoretical framework, and role of protocol to guide, b) data collection procedures, c) protocol questions, d) tentative outline for the case study report. This section offers a few examples that are really helpful. 

     The screening of candidates for a case study includes two approaches. A one-phased approach where after getting a dozen or so candidates, asking about them from other people who know information about them as well as documentation (e.g. demographics, etc). It was cautioned that a mini case review per each of the candidates should not be done. Having a set criteria prior would be beneficial. A two phased approach includes the following for a larger number of eligible candidates phase one is collecting relevant information about the candidate pool from archival sources, once this is done, a set criteria should help reduce the number of eligible participants.

     The pilot case study includes a selection of pilot cases, scope of the pilot inquiry, and  reports from the pilot cases. These serve as “dress rehearsals” to the possible case study that will be conducted. It offers information of field questions that would support the objectives of exploring the “case” and the hypothesis. 

Chapter 4 highlights the collection process of a case study. This chapter addressing six sources of evidence, how to triangulate evidence from different sources, assembling data into a comprehensive case study database, maintaining chain evidence, ways to use data from electronic sources and the strengths and weakness of the evidence. There is a table that highlights the six sources of evidence that is easily digestible they are 1) documentation 2) archival records, 3) interviews, 4) Direct Observations, 5) participant observation, 6) physical artifacts. In the text they go into depth what would be acceptable evidence and where it would fall in these categories. 

The four principles of data collection highlighted by Yin (2017) were Principle 1: Use of Multiple Sources of Evidence which include the process of triangulating sources this part of the chapter named four types of triangulation there is a useful figure/diagram that shows ways to address convergent and non-convergent evidence from multiple sources. Principle 2: Create a Case study database which helps organize the data that is collected. Principle 3: Maintaining a Chain of Evidence is Case study findings⇐> Case study database ⇐> Citations to Specific evidentiary sources in the Case Study Database ⇐> Case Study protocol (linking questions to protocol topics) ⇐> Case Study questions

            

            To this the authors offer four general strategies; relying on general theoretical proposition, working your data from the ground up, developing a case study description, and explaining plausible rival explanations. 

Relying on general theoretical proposition: The theoretical proposition that drove the data collection also informs the data analysis.  It helps structure the analysis and informs the context and explanations.

Working your data from the ground up: An inductive strategy where researchers explore and analyze the data without the lens of a theoretical proposition. Researchers can look for themes or outcomes that they may have not realized were influential to the context.

Developing a case study description: If clear patterns are developing from either the perspective of a theoretical proposition or inductive work then using a descriptive framework is another way to organize data. Creating a structure to describe research in context is a way to present data in a meaningful way.

Explaining plausible rival explanations: This strategy works well in tandem with the previous three. You don’t need to identify and analyze all possible rival theories, but rather identify what is plausible and how it may also impact the data researchers collected. (Figure 5.1 explains this well!)

            

            The authors indicate that the best way to prepare for case study analysis regardless of your strategy is to engage specific techniques. They identify five techniques that are appropriate for engaging with the strategies outlined above. These techniques are: pattern matching, explanation building, time-series analysis, logic models, and cross-case synthesis.

Pattern Matching: This is identified as a “most desirable” logic. In this you compare the patterns identified in research from existing hypotheses or rival explanations. It is a good way of teasing out the “how” and “why” of a context, while being aware that there may be other influences that inform outcomes.

Explanation Building: This is less a separate technique as much as it is an extension of pattern matching. It extends the identifying of the “how” and “why” from matching a pattern to creating a linear explanation. You “explain” why context and causal sequences resulting in an outcome. This is more difficult than just pattern matching because it requires explaining and justifying more refined ideas.

Time-Series Analysis: This is a form of pattern matching that is at the same time simpler and more complicated than traditional pattern matching. It is simpler in the way that tracks a single measure over time, and looks at its interactions with context. It is more complicated in the way that researchers must justify why the specific time intervals observed are meaningful. Why this time? What happens outside this time? Additionally within an observed time-series outcomes can present difficult to interpret. If trends are not obvious or linear than there may be contextual factors at play that are outside the series limits.

Logic Models: This seems like projecting the logic model of the context observed onto the data collected. When a context exists with a planned outcome then taking the data collected and projecting it onto the model can result in meaningful analysis. Projected/anticipated outcomes versus actual outcomes within a tight system would be a good example of this, but even loose systems/contexts have logic to them. 

Cross-Case Synthesis: This technique only applies to analysis of multiple-case studies. Wherein quantitative comparisons often there is a focus on reducing variables between cases to create data that most closely resemble each other in cross-case synthesis those variables are kept in tact. The purpose is to compare two holistic cases and their differences, while identifying patterns in outcomes and how differing variables may be impacting them.

            Audience for Case Study Research: Any time one is reporting results or composing it is important to be aware of a target audience, however when composing case studies researchers must be aware of a broader spectrum of potential audiences. Rather than simply addressing academic peers and colleagues, as with much experimental research case studies can additionally reach policy makers, special groups (dissertation advisers, etc.), or funders/sponsors of research. While research procedures are relatively defined, when composing it is important to identify which audience is your target.

            Varieties of Case Study Compositions: Choices for the myriad of decisions to be made when preparing to report findings are broken into four categories: 

            Reporting Formats: The primary formats are single-case study, multiple-case study, alternate for single or multiple-case study (more of a Q & A of a longer format study), and alternate for multiple-case only (chapter by chapter cross case comparison).

            Illustrative Structure: There are six structures identified that work particularly well for case study composition styles (explanatory, descriptive, and exploratory). Linear-analytic, comparative and chronological structures are common approaches for research reports and fit all composition styles.  Theory building, “suspense” (inverted linear-analytic) and unsequenced structures are better suited to specific composition styles.

            Methods and Research Literature Portions of a Case Study:  Every case study will also discuss research methods used and description of the relevant research literature. Methods should generally include a carefully worded research question, selection of cases, data collection profile, and an explicit, clear analytic strategy. (Figure 6.2 offers an excellent table for this on page 234.)         The length of a literature section should be dependent on the audience (it may or may not necessitate a full literature review) but it should serve to show your mastery over the topic and support the importance of your research.

            Case Studies as Part of Larger, Mixed Methods Studies: There are situations where a case study is part of a larger mixed methods study – this can change the context of how case study research is presented. Awareness of the purpose of the research within the larger study is important.

            Procedure for Composing: The how and when of creating a composition for a case study is relatively obvious – as a cohort we have discussed the procedures for writing research quite a bit. But aspects to consider include the how and if of identifying subjects and the review process. Remember that including peers for review is good, but using participants (informants) from the case study can provide meaningful feedback.

            Characteristics of an Exemplary Case Study:  Quality indicators for case study research include:

Ensuring the case study is significant – Be of meaningful, relevant public interest. 

The case study must be complete – This can be difficult to describe operationally but three guidelines presented are: clear attention to boundaries, clearly demonstrate exhaustive effort to collect all relevant evidence, absence of artificial conditions (the study didn’t end due to outside influences).

The case must consider alternate perspectives: Rival propositions must be considered.

The case must display sufficient evidence: Present unbiased  evidence including the details of the research, evidence that the research is accurate and valid

 The research must composed in an engaging manner: Clear, engaging, enticing, seductive. Well presented social research is all of these things (?!).