Although existing literature demonstrates that the analysis of case study evidence is one of the least developed and most challenging facets of doing case studies, the proposed study will employ four techniques named in Yin (1994), namely pattern matching, explanation building, time-series analysis, and logic models. These analytic techniques are explained as follows:
Pattern Matching
Pattern matching is one of the most favored approaches as it enables the researcher to compare an empirically based pattern with a predicted one (or with several alternative predictions), to not only strengthen the internal validity of the case study if patterns coincide (Yin 1994) but also to infer strong causal inferences if the predicted alternatives are confirmed to be present in the case study (Wynn & Williams 2012). The predictions can be made based on “…a pattern of nonequivalent dependent variables, a pattern based on rival explanations, or a simple pattern” (Yin 1994, p. 140-141). However, despite the pattern used, it is of essence to always remember that the essential comparison between the predicted and the actual pattern does not have to involve any quantitative or statistical methodologies.
Explanation Building
This analytic model aims to analyze data arising from the case study by generating explanations about the case that could be employed to develop ideas and concepts for further study rather than conclude a study (Yin 1994), as is the intention in the proposed study. The explanation-building approach is also applicable in situations where the study is concerned with revising theoretical positions; however, this is not the aim of the proposed study, hence the approach may be inapplicable in this context.
Time-Series Analysis
The essential logic underlying a time-series analytic approach is grounded on “…the match between the observed (empirical) trend and either of the following: (a) a theoretically significant trend specified before the onset of the investigation or (b) some rival trend, also specified earlier” (Yin 1994, p. 147). It is important to note that in situations that demand the employment of time-series analysis as the relevant approach to a case study, the researcher(s) must always take time to identify the specific indicators or variables that need to be traced over time, not mentioning that they are also required to identify the specific time intervals they aim to cover and the alleged sequence of relationships that may have been noted before undertaking the actual data collection exercise (Barth & Thomas 2012). Yin (1994, p. 149) reinforces this view by suggesting that “…only as a result of such prior specification are the relevant data likely to be collected in the first place, much less analyzed properly and with minimal bias.”
Logic Models
This analytic approach, which is increasingly gaining popularity among scholars and practitioners engaged in doing case study evaluations, consciously specifies a complex sequence of events over an extended period and is staged in repeated cause-effect-cause-effect patterns, whereby a single dependent variable (the phenomenon of interest) at an earlier phase transition into the independent variable (causal phenomenon) for the next phase. As an analytic model, extant literature demonstrates that “…the use of logic models consists of matching empirically observed events to theoretically predicted events” (Yin 1994, p. 149). Conceptually, therefore, it can be argued that logic model techniques resemble pattern matching analytic techniques with the exception that the latter does not make use of sequential phases in the analysis of the case study (Wynn & Williams 2012).
Reference
Barth, M & Thomas, I 2012, ‘Synthesizing case-study research – ready for the next step?’, Environmental Education Research, vol. 18 no. 6, pp. 751-764.
Wynn, D & Williams, CK 2012, ‘Principles for conducting critical realist case study research in information systems’, MIS Quarterly, vol. 36 no. 3, pp. 787-810.
Yin, RK 1994, Case study research: Design and methods, Sage Publications, Thousand Oaks, CA.