Case Study
A case study is unique in that it can be thought of both as a method, and as a research design, depending on the school of thought to which one subscribes. Though it is named here as the former, the following explanation is constructed such that it is treated as the latter. The author understands that a case study consists of multiple methods, but that those methods vary depending on the particular case. As such, each case study must be carefully designed with several considerations in mind. Before discussing those considerations, it is helpful to examine the definition of “case study.”
Case Study: A Working Definition
To begin with, case study methodology is inherently constructivist. In a constructivist approach to any phenomenon, reality is seen as being socially constructed, and relative to an individual’s lived social experience (Baxter and Jack, 2008). Because a case study “investigates a contemporary phenomenon (the ‘case’) in depth and within its real-world context,” the background and experiences of the case’s key players are an integral part of the case (Yin, 2014, pp. 16-17; Gillham, 2000). In other words, it is impossible to separate an individual from his or her socially constructed reality.
Second, a case study uses several sources and methods to obtain data about the phenomenon under investigation (Yin, 2014). In fact, the researcher may not be fully aware of what defines the particular case under study until all sources and methods have been thoroughly examined, and the data triangulated (Ragin and Becker, 1992). As Masoner (1988) pointed out, “the uniqueness of a case study lies not so much in the methods employed (although these are important) as in the questions asked and their relationship to the end product” (p. 15).
Finally, a case study is flexible. A case study can (1) be adapted to multiple disciplinary and philosophical perspectives; (2) be used to test or build theory; (3) include either qualitative or quantitative data, or both; and (4) accommodate more than one sampling method (Masoner, 1988). Further, a case study can be used to examine current, observable phenomena or to reconstruct a particular historical case (George and Bennett, 2005). A case study may include only one, or a combination, of these features.
Given the five interdependent features, case study is a complex approach. However, Feagin, Orum, and Sjoberg (1991) provide a good working definition of the case study that demonstrates this complexity while still being succinct: “A case study is here defined as an in-depth, multifaceted investigation, using qualitative research methods, of a single social phenomenon. The study is conducted in great detail and often relies on the use of several data sources” (p. 2).
Planning and Designing a Case Study
After a working definition has been established, considerations for planning and designing a case study can be explored. It is important to know when to use a case study, as opposed to another type of method or design. To this end, Yin (2014) provides some specific guidelines for when a case study might be the optimal design: a case study is advantageous when “a ‘how’ or ‘why’ question is being asked about a contemporary set of events, over which a researcher has little or no control” (p. 14). Yin recommends conducting a thorough literature review on the proposed research topic in order to determine what is not yet known about the topic, and to formulate a research question. Once the question has been formalized, the guidelines above can be used in order to determine whether or not a case study is feasible.
Assuming that a case study is determined to be optimal, the researcher must then carefully design his or her study. This involves consideration of five key components: “(1) a case study’s questions; (2) its propositions, if any; (3) its unit(s) of analysis; (4) the logic linking the data to the propositions; and (5) the criteria for interpreting the findings” (Yin, 2014, p. 29). With these components in place, the ideal type and category of case study for the proposed research should begin to emerge.
Categories and Types of Case Studies
Case studies can be classified by both type and category. The type of case study refers to its scope: how many individual cases are under study and how many units of analysis are included. The category of case study refers to the study’s purpose: why the researcher is conducting the investigation.
Categories. There are two main categories of case studies: single-case and multiple-case. Each of these can then also be either holistic or embedded. A single-case holistic study would look at a single case and a single unit of analysis. A single-case embedded study would examine a single case, but have multiple units of analysis. Similarly, a multiple-case holistic study would look at multiple cases, but only a single unit of analysis for each case. A multiple-case embedded study would examine multiple cases and multiple units of analysis. In multiple-case studies, the units of analysis would be the same across sites, or cases. Single-case studies are appropriate when the case is either “critical, unusual, common, revelatory, or longitudinal” (Yin, 2014, p. 51). Multiple-case designs are appropriate when the researcher wants to show replication, or make a more compelling case in addressing the research questions (Yin, 2014). Types. In addition to these categories, there are also five types of case studies: exploratory/evaluative, descriptive/illustrative, explanatory/interpretive, intrinsic, and instrumental. The type of case study one chooses to conduct depends on the purpose of the study, and on the research questions (Baxter and Jack, 2008).
The first three types of case studies are defined in Yin (2014). The exploratory, or evaluative, case study is intended to serve primarily as the basis for further research. This type of study explores a phenomenon in depth in order to elicit research questions of interest. The descriptive, or illustrative, case study is designed to observe and, subsequently, describe some event or behavior “in its real-world context” (Yin, 2014, p. 238). Finally, the explanatory, or interpretive, case study seeks to discover causation or correlation.
The other two types of case studies are defined by Stake (1995). An intrinsic case study is best used when the intention of the researcher is to gain a more complete understanding of the phenomenon in question in that particular case. An instrumental case study, conversely, should be used when the case itself is not the center of interest, and the focus is not on understanding or even generalizing results. Instead, it is useful when the researcher is looking for some specific insight or working on refinement of a particular theory. Once the appropriate category and type of case study have been chosen, data collection methods should be considered.
Data Collection Methods in Case Studies
Case studies are unique in that data collection methods are almost unlimited. However, not all methods are appropriate for all cases. The research design and research questions should be carefully considered when determining which methods should be used. Also, though the case study has historically been primarily qualitative in nature, quantitative methods can be employed in complement to qualitative methods. For example, a survey can be useful as a starting point for a case study.
Surveys. A survey can be especially useful as a first step in designing a case study. A survey is designed to collect general information, under natural conditions, on particular variables for a specific sample. Ideally, survey samples are chosen such that the information collected can be considered to be generalizable to the larger population (Roberts, 1999). However, the use of the survey as a starting point for a case study does not need such considerations in design, as the population of the case study is the population of interest, and the purpose of conducting the survey is to uncover topics of interest for further exploration using qualitative methods, such as interviews and observations.
Interviews. “The qualitative interview is the most common and one of the most important data gathering tools in qualitative research” (Myers and Newman, 2007, p. 3). Interviews can be structured, semi-structured, or unstructured. A structured interview follows a very specific set of questions and does not deviate to topics outside of those questions. A semi-structured interview, on the other hand, consists of a series of open-ended questions with room to move off topic if deemed of interest. Finally, an unstructured interview often begins with a topic of interest about which the researcher wants further information. Many times, these unstructured interviews begin with a researcher’s field notes following a period of observation (DiCicco‐Bloom, B. and Crabtree, B. F., 2006).
Observations. Yin (2014) describes two types of observations: direct and participant. In direct observation, the researcher positions herself such that she can unobtrusively observe the phenomenon in question, leaving her free to make notes on what she observes. Notes may be taken freely, or a more structured form may be provided for data collection. Participant observation, as its name suggests, occurs when the researcher participates in the phenomenon, and may or may not be able to take notes while participating. Gillham (2000) further describes direct observation as “mainly analytical/categorical” and participant observation as “mainly descriptive/interpretive” (p. 52). There is, of course, room for overlap between the two.
Textual evidence. Questioning and observing the participants in a particular case often tells a compelling story, but sometimes additional evidence is needed to corroborate that story or expand it. This is where documents and archival records can be useful. Reviewing such textual evidence can help to verify factual information gathered during an interview or observation. Inferences can also be drawn from certain documents or records and used as indications that there is a need to investigate further. Finally, textual evidence can be used as a preliminary overview of a particular organization or case (Yin, 2014).
Physical artifacts. One final method of data collection worth mentioning is the study of physical artifacts. Though traditionally used primarily in anthropological research, technology and its products (computer code, activity logs, etc.) can be considered physical artifacts and may be relevant to the case under study (Yin, 2014).
Rigor in the Case Study
The case study has been criticized as a research design for its perceived lack of rigor (Yin 2014). According to Gibbert, Ruigrok, and Wicki (2008), this concern can be addressed by vigilant attention to study design and careful consideration of the four criteria for establishing rigor, as defined by the positivist tradition: construct validity, internal validity, external validity, and reliability.
Construct Validity. Construct validity refers to the use of appropriate “operational measures for the concepts being studied” (Yin, 2014, p. 46). Establishing construct validity is particularly challenging because a case study is often exploratory, and operationalization must occur during the course of the study. For this reason, the researcher must be extra diligent about refraining “from subjective judgments during the periods of research design and data collection” (Riege, 2003, p. 80). Further measures for increasing construct validity include the following: using a variety of sources to support findings, connecting evidence to form a chain that can be followed logically, and having case study participants review data for inconsistencies and misunderstandings (Riege, 2003; Yin, 2014).
Internal Validity. Internal validity deals with the establishment of causation. Traditionally, the biggest challenge in establishing internal validity is in ferreting out spurious relationships that do not actually show causation (Yin, 2014). Case study research, however, is more concerned with establishing the ability to make inferences from the case that will hold up in the general population (Riege, 2003). Measures for increasing internal validity include pattern matching, explanation building, addressing alternate explanations, and using logic models (Riege, 2003; Yin, 2014).
External Validity. External validity involves the generalizability of research findings, and can be addressed initially by properly constructed research questions. Specifically, “how” and “why” questions should be included in order to arrive “at an analytic generalization” (Yin, 2014, p. 48). Other ways to increase external validity are to employ theory in the design of single-case studies, and replication logic in multiple-case studies, as well as to clearly define the scope of the study (Riege, 2003; Yin, 2014).
Reliability. Reliability refers to the ability to repeat research and get the same, or reasonably similar, results. The concern when dealing with qualitative research is the subjectivity of the researchers. Ways to circumvent this concern involve keeping detailed records and accounts of the research. Yin (2014) suggests using a “case study protocol” and developing and maintaining a “case study database” (p. 49).
Questions
- Design a case study using Yin’s five key components. Be sure to include the rationale for the type and category of case chosen, as well as the methods.
- Describe appropriate methods of analysis for case study research.
- Describe the compatibility between information worlds and case study research in the context of a specific research problem. Specify the problem and research questions, then describe how you would use case study and information worlds to approach the research.
- Choose two examples of case study research in education and/or LIS. Compare, contrast, and critique the studies and describe how you would have approached the case differently, if applicable.
References
Baxter, P., & Jack, S. (2008). Qualitative case study methodology: Study design and implementation for novice researchers. The Qualitative Report, 13(4), 544-559. Retrieved from http://nsuworks.nova.edu/tqr/vol13/iss4/2
DiCicco‐Bloom, B., & Crabtree, B. F. (2006). The qualitative research interview. Medical education, 40(4), 314-321.
Eisenhardt, K. M. (1989). Building theories from case study research. The Academy of Management Review, 14(4), 532–550. doi:10.2307/258557
Feagin, J. R., Orum, A. M., & Sjoberg, G. (1991). A Case for the case study. Chapel Hill: University of North Carolina Press.
George, A. L., & Bennett, A. (2005). Case studies and theory development in the social sciences. Cambridge, Mass: MIT Press.
Gerring, J. (2004). What is a case study and what is it good for? The American Political Science Review, 98(2), 341–354.
Gibbert, M., Ruigrok, W., & Wicki, B. (2008). What Passes as a Rigorous Case Study? Strategic Management Journal, 29(13), 1465–1474.
Gillham, B. (2000). Case study research methods. London ; New York: Continuum.
Jaeger, P. T., & Burnett, G. (2010). Information worlds: Social context, technology, and information behavior in the age of the Internet. New York, NY: Routledge.
Masoner, M. (1988). An audit of the case study method. New York: Praeger.
Merriam, S. B. (1988). Case study research in education : a qualitative approach (1st ed.). San Francisco: Jossey-Bass.
Myers, M. D., & Newman, M. (2007). The qualitative interview in IS research: Examining the craft. Information and organization, 17(1), 2-26.
Ragin, C. C., & Becker, H. S. (Eds.). (1992). What is a case?: exploring the foundations of social inquiry. Cambridge university press.
Riege, A.M. (2003). Validity and reliability tests in case study research: a literature review with “hands‐on” applications for each research phase. Qualitative Market Research: An International Journal, 6(2), 75–86. http://doi.org/10.1108/13522750310470055
Roberts, E. S. (1999). In defence of the survey method: An illustration from a study of user information satisfaction. Accounting & Finance, 39(1), 53-77.
Stake, R. E. (1995). The art of case study research. Thousand Oaks: Sage Publications.
Yin, R. K. (2014). Case study research : design and methods (Fifth edition.). Los Angeles: SAGE.