Case study research is a time-honored, traditional approach to the study of topics in social science and management. Because only a few instances are normally studied, the case researcher will typically uncover more variables than he or she has data points, making statistical control (ex., through multiple regression) an impossibility. This, however, may be considered a strength of case study research: it has the capability of uncovering causal paths and mechanisms, and through richness of detail, identifying causal influences and interaction effects which might not be treated as operationalized variables in a statistical study, As such it may be particularly helpful in generating hypotheses and theories in developing fields of inquiry.

In recent years there has been increased attention to implementation of case studies in a systematic, stand-alone manner which increases the validity of associated findings. However, although case study research may be used in its own right, it is more often recommended as part of a multimethod approach ("triangulation") in which the same dependent variable is investigated using multiple additional procedures (ex., also grounded theory, survey research, sociometry and network analysis, focus groups, content analysis, ethnography, participant observation, narrative analysis, archival data, or others).

Key Concepts and Terms

• Types of case studies. Jensen and Rodgers (2001: 237-239) set forth a typology of case studies, including these types:
  1. Snapshot case studies: Detailed, objective study of one research entity at one point in time. Hypothesis-testing by comparing patterns across sub-entities (ex., comparing departments within the case study agency).
  2. Longitudinal case studies. Quantitative and/or qualitative study of one research entity at multiple time points.
  3. Pre-post case studies. Study of one research entity at two time points separated by a critical event. A critical event is one which on the basis of a theory under study would be expected to impact case observations significantly.
  4. Patchwork case studies. A set of multiple case studies of the same research entity, using snapshot, longitudinal, and/or pre-post designs.This multi-design approach is intended to provide a more holistic view of the dynamics of the research subject.
  5. Comparative case studies. A set of multiple case studies of multiple research entities for the purpose of cross-unit comparison. Both qualitative and quantitative comparisons are generally made.

• Representativeness. Unlike random sample surveys, case studies are not representative of entire populations, nor do they claim to be. The case study researcher should take care not to generalize beyond cases similar to the one(s) studied. Provided the researcher refrains from over-generalization, case study research is not methodologically invalid simply because selected cases cannot be presumed to be representative of entire populations. Put another way, in statistical analysis one is generalizing to a population based on a sample which is representative of that population. In case studies, in comparison, one is generalizing to a theory based on cases selected to represent dimensions of that theory.

• Case selection should be theory-driven. When theories are associated with causal typologies, the researcher should select at least one case which falls in each category. That cases are not quantitative does not relieve the case researcher from identifying what dependent variable(s) are to be explained and what independent variables may be relevant. Not only should observation of these variables be part of the case study, but ideally the researcher would study at least one case for every causal path in the model suggested by theory. Where this is not possible, often the case, the researcher should be explicit about which causal types of cases are omitted from analysis. Cases cited in the literature as counter-cases to the selected theory should not be omitted. In public administration, "best practices" lists may provide cases for selection, though it is necessary also to pick contrasting cases.
  • Cross-theoretic case selection. As multiple theories can conform to a given set of data, particularly sparse data as in case study research, the case research design is strengthened if the focus of the study concerns two or more clearly contrasting theories. This enables the researcher to derive and then test contrasting expectations about what would happen under each theory in the case setting(s) at hand.

  • Other selection criteria: Yin (1984) points out that researchers may select cases not only when they are critical (to testing a theory), but also when they are revelatory (reveal relationships which cannot be studied by other means) or unusual (throws light on extreme cases).

• Pattern matching is the attempt of the case researcher to establish that a preponderance of cases are not inconsistent with each of the links in the theoretical model which drives the case study. For instance, in a study of juvenile delinquency in a school setting, bearing on the theory that broken homes lead to juvenile delinquency, cases should not display a high level of broken homes and simultaneously a low level of delinquency. That is, the researcher attempts to find qualitative or quantitative evidence in the case that the effect association for each causal path in the theoretical model under consideration was of non-zero value and was of the expected sign.

  • Process tracing is the a more systematic approach to pattern matching in which the researcher attempts, for each case studied, to find evidence not only that patterns in the cases match theoretical expectations but also that (1) that there is some qualitative or quantitative evidence that the effect association which was upheld by pattern matching was, in fact, the result of a causal process and does not merely reflect spurious association; and (2) that each link in the theory-based causal model also was of the effect magnitude predicted by theory. While process tracing cannot resolve indeterminancy (selecting among alternative models, all consistent with case information), it can establish in which types of cases the model does not apply.
    • Controlled observation is the most common form of process tracing. Its name derives from the fact that the researcher attempts to control for effects by looking for model units of analysis (ex. people, in the case of hypotheses about people) which shift substantially in magnitude or even valence, on key variables in the model being investigated. In a study of prison culture, for instance, in the course of a case study an individual may shift from being free to being incarcerated; or in a study of organizational culture, an individual may shift from being a rank-and-file employee to being a supervisor). Such shifts can be examined to see if associated shifts in other variables (ex., opinions) also change as predicted by the model. Controlled observation as a technique dictates that the case study (1) be long enough in time to chronicle such shifts, and (2) favor case selection of cases where shifts are known to have occurred or are likely to occur.

    • Time series analysis is a special and more rigorous case of process tracing, in which the researcher also attempts to establish not only the existence, sign, and magnitude of each model link is as expected, but also the temporal sequence of events relating the variables in the model. This requires observations at multiple points in time, not just before-after observations, in order to establish that the magnitude of a given effect is outside the range of normal fluctuation of the time series.

    • Critical incident technique (CIT). CIT is a method of case selection and analysis developed in the 1950s by the Air Force at the University of Pittsburgh (see Flanagan, 1954). It is a methodology associated with the American Institutes for Research (AIR), founded by John C. Flanagan as a more empirical data collection method thought to be superior to conventional survey research. AIR has provided an extensive bibliography on CIT. Cases are selected based on their being important, significant, and critical to determining either an effective or ineffective outcome. Effective outcomes are those which solve a problem or resolve a situation. Ineffective outcomes are those which only partially solve problems or resolve situations, but also create new problems or new needs for resolution. By collecting a large number of brief, factual reports on critical incidents, researchers attempt to identify common factors associated with effective outcomes. These factors may be set forth in "critical requirements" or "acceptable performance" standards for the organization.

  • Congruence testing is an even more systematic approach to pattern matching which requires the selection of pairs of cases which are identical in causal type, except for the difference of one independent variable. Differences in the dependent variable are attributed to incongruency on the independent. Where there are a large number of cases, it may be possible to replace congruence testing with statistical methods of correlation and control.

• Explanation-building is an alternative or supplement to pattern matching. Under explanation-building, the researcher does not start out with a theory to be investigated. Rather, the researcher attempts to induce theory from case examples chosen to represent diversity on some dependent variable (ex., cities with different outcomes on reducing welfare rolls). A list of possible causes of the dependent variable is constructed through literature review and brainstorming, and information is gathered on each cause for each selected case. The researcher then inventories causal attributes which are common to all cases, common only to cases high on the dependent variable, and common only to cases low on the dependent variable. The researcher comes to a provisional conclusion that the differentiating attributes are the significant causes, while those common to all cases are not. Explanation-building is particularly compelling when there are plausible rival explanations which can be rebutted by this method. Explanation-building can also be a supplement to pattern matching, as when it is used to generate a new, more plausible model after pattern matching disconfirms an initial model.

• Meta-Analysis is a particular methodology for extending grounded theory to a number of case studies. In meta-analysis the researcher creates a meta-analytic schedule, which is a cross-case summary table in which the rows are case studies and the columns are variable-related findings or other study attributes (ex., time frame, research entity, case study design type, number and selection method for interviewees, threats to validity like researcher involvement in the research entity). The cell entries may be simple checkmarks indicating a given study supported a given variable relationship, or the cell entries may be brief summaries of findings on a given relationship or brief description of study attributes. The purpose of meta-analysis is to allow the researcher to use the summary of case studies reflected in the meta-analytic table to make theoretical generalizations. In doing so, sometimes the researcher will weight the cases according to the number of research entities studied, since some case studies may examine multiple entities. See Hodson (1999); Jensen and Rodgers (2001: 239 ff.). Hodson (1999: 74-80) reproduces an example of a meta-analytic schedule for the topic of workplace ethnography.

  Problems of meta-analysis include what even case study advocates admit is the "formidible challenge" (Jensen and Rodgers, 2001: 241) involved in developing a standardized meta-analytic schedule which fits the myriad dimensions of any sizeable number of case studies. No widely accepted "standardized" schedules exist. Moreover, for any given proposed schedule, many or most specific case studies will simply not report findings in one or more of the column categories, forcing meta-analysts either to accept a great deal of missing data or to have to do additional research by contacting case authors or even case subjects.

  Considerations in implementing meta-analytic schedules:

  1. Variables: In addition to substantive variables particular to the researcher's subject, methodological variables should be collected, such as date of data collection, subject pool, and methodological techniques employed.

  2. Coder training. It is customary to provide formal training for coders, who ideally should not be the researchers so that data collection is separated from data interpretation.

  3. Reliability. The researcher must establish inter-rater reliability, which in turn implies there must be multiple raters. Reliability is generally increased through rater debriefing sessions in which raters are encouraged to discuss coding challenges. Duplicate coding (allowing 10% or so of records to be coded by two coders rather than one) is also used to track reliability. In larger projects, rating may be cross-validated across two or more groups of coders.

  4. Data weighting. Meta-analysis often involves statistical analysis of results, where cases are studies. The researcher must decide whether cases based on a larger sample size should be weighted more in any statistical analysis. In general, weighting is appropriate when cases are drawn from the same population to which the researcher wishes to generalize.

  5. Handling missing data. Dropping cases where some variables have missing data is generally unacceptable unless there are only a very small number of such cases as (1) it is more likely that missing-data cases are related to the variables of the study than that they are randomly distributed, and (2) dropping cases when the number of cases is not large (as is typical of meta-analytic studies) diminishes the power of any statistical analysis. There is no good solution for missing data. See the separate section on data imputation, but maximum likelihood estimation of missing values carries fewer assumptions about data distribution than using regression estimates or substituting means. SPSS supports MLE imputation.

  6. Outliers. Metapanalysis often involves results coded from a relatively small number of cases (ex., < 100). Consequently, any statistical analysis may be affected strongly by the presence of outlier cases. Sensitivity analysis should be conducted to understand the difference in statistical conclusions with and without the outlier cases included. The researcher may decide that deviant case analysis may be appropriate, based on a finding that relationships among the variables operate differently for outlier cases.

  7. Spatial autocorrelation. It is possible that central tendencies and conclusions emerging from meta-analytic studies will be biased because cases cluster spatially. If many cases are from a spatially neighboring area and if the relationships being studied are spatially related, then generalization to a larger reference area will be biased. If the researcher has included longitude and latitude (or some other spatial indicator) as variables, then many geographic information systems packages and some statistical packages can check for spatial autocorrelation (see Land and Deane, 1992). However, a visual approach of mapping cases to identify clusters, then comparing in-cluster and out-of-cluster statistical results usually is a sufficient check on spatial autocorrelation.

Assumptions

• Cases selected based on dimensions of a theory (pattern-matching) or on diversity on a dependent phenomenon (explanation-building).

• No generalization to a population beyond cases similar to those studied.

• Conclusions should be phrased in terms of model elimination, not model validation. Numerous alternative theories may be consistent with data gathered from a case study.

• Case study approaches have difficulty in terms of evaluation of low-probability causal paths in a model as any given case selected for study may fail to display such a path, even when it exists in the larger population of potential cases.

Frequently Asked Questions

• What are common standards for case studies based dissertations?
• Is case study research a social science substitute for scientific experimentation?
• Aren't case studies unscientific because they cannot be replicated?
• Aren't case studies unscientific because findings cannot be generalized?
• Where can I find out more about case study research?
• What is NUD*IST?
  
  
  
• What are common standards for case studies based dissertations?
  Doctoral programs do not dictate methodology but rather leave issues of research design and empirical procedure to dissertation committees. For this reason few programs articulate formal bars to any particular methodology, case studies included. Although it is true that skepticism toward the case study method is widespread, prompted by fears of low quality, the great majority of doctoral programs in public administration and policy nonetheless allow case studies. Many, however, do so with guidelines or stipulations. Based on replies to a survey of 35 doctoral programs in public administration and public policy, a composite set of guidelines has been constructed below.

  Common Guidelines for Case Study Based Dissertations

  Case study dissertations should represent original research, be analytic, well-written, insightful, systematic, explicitly related to the literature of the field, and should cover their focus in depth. This focus must test propositions which are relevant to significant theoretical issues. Theoretical issues may be political-theoretic, decision-theoretic, economic or market-theoretic, or public policy or action-theoretic, to name some of the possible dimensions of theory. In this way the criteria for acceptable case study dissertations do not differ from those for other types of dissertations.

  To test propositions derived from theory, one must have some variance in the dependent variable(s) under study, which in turn requires there be some type of comparison such as might be provided by before-after studies of a policy intervention or by examining a phenomenon in a public compared to a private setting. That is, case study dissertations must have a longitudinal, cross-sectional, or other comparative perspective. In some but not all dissertations, it may be necessary to study multiple cases to achieve the requisite variance in the object of study. Non-longitudinal "single shot" case studies of a given organization or policy event do not provide a basis for comparison and testing of propositions and are not acceptable no matter how detailed the description. In fact, description not directly germane to the theoretical concerns of the thesis should be relegated to appendices or dropped from the dissertation altogether.

  Because case study dissertations seek to provide theoretical or policy insight based on a small number of cases or even on a single case, a "triangulation" approach to validation is strongly recommended. Such a rigorous approach involves a multi-method design in which key constructs and processes are traced using more than a single methodology. Multiple methods may include structured and unstructured interviews, sample surveys, focus groups, narrative analysis, phenomenological research, ethnography, symbolic action research, network analysis, advocacy coalition research (Sabatier), content analysis, participant observation, examination of archival records, secondary data analysis, experiments, quasi-experiments, and other methods. Testing the same propositions through data gathered by multiple methods helps address some of the validation problems in case study designs.

  The standard reference for public administration and public policy graduate students doing case study research, formally recommended by many programs, is Robert Yin's Case Study Research: Design and Methods (1984, 1994, 2002). Other references which were cited by survey respondents as the basis for standards for certain types of case study research included Goetz and LeCompte (1984), Ragin (1987), Strauss and Corbin (1990), Sabatier (1993), and Morgan (2001).

• Is case study research a social science substitute for scientific experimentation?
  It is interesting to note that case study research plays an important role in the natural sciences as well as social sciences. Many scientific fields, such as astronomy, geology, and human biology, do not lend themselves to scientific investigation through traditional controlled experiments. Darwin's theory of evolution was based, in essence, on case study research, not experimentation, for instance.

• Aren't case studies unscientific because they cannot be replicated?
  It is true that a later researcher using case methods will of necessity be studying a different case, if only because he or she comes later, and therefore may come to different conclusions. Similarly, in experimental and quasi-experimental research the subjects will differ, meaning relationships may differ. What makes research replicable in either case study or experimental research is not the units of analysis but whether the research has been theory-driven. If the case researcher has developed and tested a model of hypothesized relationships, then a future case researcher can replicate the initial case study simply by selecting cases on the basis of the same theories, then testing the theories through pattern matching. If pattern matching fails to uphold theories supported by the first case researcher, the second case researcher may engage in explanation building, as discussed above, to put forward a new model.

• Aren't case studies unscientific because findings cannot be generalized?
  Generalizability of findings is a function of the range and diversity of settings in which a theory is tested, not of the testing methodology per se. It is true that randomization of subjects in experimental research and random sampling in quasi-experimental research, along with larger sample sizes, mean that research of this type can more easily lay claim to range and diversity than can case study research projects. Nonetheless, judicious case selection to identify cases illustrating the range of a theory (ex., a theory about causes of divorce) may result in more generalizable research than, say, the attempt to test the same theory based on a random sample of students in one university. Moreover, if case research is replicated (discussed above), generalization of case-based findings can be enhanced further.

• Where can I find out more about case study research?
  • Introduction to case study, by Winston Tellis, The Qualitative Report, Vol. 3, No. 2, July, 1997
  • Text Analysis Software Links
  • More Content Analysis Software Links

• What is NUD*IST (N6)?
  NUD*IST is software standing for Non-numerical Unstructured Data Indexing, Searching, and Theorizing, which can be used in conjunction with grounded theory to create and analyze theories and provide a framework for understanding. At this writing, the latest version is called simply N6. It can handle coding categories and sub-categories, supporting hierarchical indexing; browse and code documents and indexing databases; search for words and word patterns and combine them in indexes; "memo link" emerging codes and categories with their associated documents; and create new indexing categories out of existing ones. See Richards and Richards (1991). Information on NUD*IST and other content analytic software is linked from http://www.qsr.com.au/products/productoverview/product_overview.htm.

Bibliography

• Allison, Graham T. (1971). Essence of decison: Explaining the Cuban Missile Crisis. Boston: Little, Brown, and Co. Considered a classic model case study.
• Annells, M. P. (1997). Grounded theory method, part 1: within the five moments of qualitative research," Nursing Inquiry, Vol. 4:120-129; and Grounded theory method, part II; options for users of the method, Nursing Inquiry, Vol. 4: 176-180. A good two-part historical overview of the grounded theory method, relating it to theory, methodological issues, and practical procedures.
• Bailey, Mary Timney (1992). Do physicists use case studies? Thoughts on public administration research. Public Administration Review, Vol. 52 (Jan./Feb.): 47-55. Much-cited article defends the case study approach as meeting scientific standards of generalizability, transferability, and replicability.
• Benbasat; Izak; David K. Goldstein; & Melissa Mead (1987). The case research strategy in studies of information systems. MIS Quarterly 11(3), 369-386.
• Bock, Edwin a., ed. (1962). Essays on the case method. NY: Inter-University Case Program.
• Campbell, Donald (1975). 'Degrees of freedom' and the case study. Comparative Political Studies 8( 2): 178-193. Classic defense of the case study method.
• Clarke, Adele (2005). Situational analysis: Grounded theory after the postmodern turn. Thousand Oaks, CA: Sage Publications. Based on Anselm Straus's ecological social worlds/arenas.discourses framework. Treats development of situational maps, social worlds/arenas maps, and positional maps.
• Denzin, Norman K. and Yvonna Lincoln (2003). Strategies of qualitative inquiry, Second edition. Thousand Oaks, CA: Sage Publications. Includes a chapter on case studies by Robert Stake.
• Eisenhardt, K.M. (1989). Building theories from case study research. Academy of Management Review, 14(4): 532-550.
• Flanagan, J. C. (1954). The critical incident technique. Psychological Bulletin 5(4), 327-358.
• Glaser, Barney G. and Anselm Strauss (1967). The discovery of grounded theory: Strategies for qualitative research.. Chicago, IL: Aldine Publishing Co. The seminal work in grounded theory.
• Goetz, J. P. and M. D. LeCompte (1984). Ethnography and qualitative design in educational research. London: Academic Press.
• Hamel, Jacques (1993). Case study methods. . Thousand Oaks, CA: Sage Publications. Covers history of the case study method.
• Hodson, Randy (1999). Analyzing documentary accounts. Thousand Oaks, CA: Sage Publications. Quantitative Applications in the Social Sciences Series No. 128. Describes random sampling of ethnographic field studies as a basis for applying a meta-analytic schedule. Hodson covers both coding issues and subsequent use of statistical techniques.
• Hoyle, R. H. (1999). Statistical strategies for small sample research. Thousand Oaks, CA: Sage Publications. Not on case studies per se, but relevant.
• Kennedy, Mary M. (1979). Generalizing from single case studies. Evaluation Quarterly, 3: 661-78.
• Land, K. C. and G. Deane (1992). On the large-sample estimation of regression models with spatial- or network-effects terms. In P. V. Marsden, ed. (1992: 221-248). Sociological Methodology. Washington, DD: American Sociological Association.
• Lee, Allen S. (1989). A scientific methodology for MIS case studies. MIS Quarterly, March: 33-50. Using management information systems as a focus, Lee addresses problems of and remedies for case study research.
• Lucas, W. (1974). The case survey method of aggregating case experience. Santa Monica, Calif.: Rand.
• Morgan, David L. (2001). Combining qualitative and quantitative methods. Portland, OR: Portland State University.
• Naumes, William and Margaret J. Naumes (1999). The art and craft of case writing. Thousand Oaks, CA: Sage Publications. The authors have led case writing workshops for the Academy of Management and the Decision Sciences Institute, and use their experiences to illustrate issues in case stduy research.
• Ragin, Charles (1981). Comparative sociology and the comparative method, International Journal of Comparative Sociology, 22(1-2): 102-120.
• Ragin, Charles (1987). The comparative method: Moving beyond qualitative and quantitative strategies. Berkeley: University of California Press. Ragin is a distinguished sociologist noted for his defense of the case study method.
• Ragin, Charles, and David Zaret (1983). Theory and method in comparative research: Two strategies. Social Forces, 61(3): 731-754.
• Richards, Tom J. and Lynn Richards (1991). The NUDIST qualitative data analysis system. Qualitative Sociology, Vol. 14: 307-324.
• Rhodes, Terrel R. (2002). The public manager case book. Thousand Oaks, CA: Sage Publications. Eight decision-making focused case studies.
• Sabatier, Paul C. (1993). Policy change and learning: An advocacy coalition approach. Boulder, CO: Westview Press.
• Sabatier, Paul C., ed. (1999). Theories of the policy process: Theoretical lenses on public policy. Boulder, CO: Westview Press.
• Scholz, Roland W. and Olaf Tietje (2002). Embedded case study methods: Integrating quantitative and qualitative knowledge. Thousand Oaks, CA: Sage Publications.
• Soy, Susan K. (1997). The case study as a research method. Unpublished paper. Retrieved 10/1/07 from http://www.gslis.utexas.edu/~ssoy/usesusers/l391d1b.htm.
• Stake, Robert E. (1995). The art of case study research. Thousand Oaks, CA: Sage Publications. Focuses on actual case (Harper School) to discuss case selection, generalization issues, and case interpretation.
• Strauss, Anselm and Juliet Corbin (1990). Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park, CA: Sage Publications. Probably now the leading contemporary treatment of grounded theory.
• Strauss, Anselm and Juliet Corbin, eds. (1997). Grounded theory in practice. London: Sage Publications.
• Travers, Max (2001). Qualitative research through case studies. Thousand Oaks, CA: Sage Publications. Discusses grounded theory, dramaturgical analysis, ethnomethodology, and conversation analysis.
• U. S. General Accounting Office (1990). Case study evaluations. Washington, DC: USGPO. GAO/PEMD-91-10.1.9. Available online in .pdf format. Government manual for doing evaluation research using case studies.
• Yin, Robert K. (1984). Case study research, design, and methods. 3rd edition. Thousand Oaks, CA: Sage Publications. Second ed., 1994. Third ed., 2002.
• Yin, Robert K. (2002). Applications of case study research, 2nd edition. Thousand Oaks, CA: Sage Publications. Completed case studies discussed in relation to themes in Yin (1984 ff.).
• Yin, Robert K. and Karen A. Heald (1975). Using the case survey method to analyze policy studies. Administrative Science Quarterly 20(3): 371-381.

Copyright 2002, 2008 by G. David Garson.