Another preparatory step is the final selection of the case(s) to be part of your case study. Sometimes, the selection is straightforward because you have chosen to study a unique case whose identity has been known from the outset of your inquiry. Or, you already may know the case you will study because of some special arrangement or access that you have. However, at other times, there may be many qualified case study candidates, and you must choose your final single case or array of multiple cases from among them. The goal of the screening procedure is to be sure that you identify the final cases properly prior to formal data collection. The worst scenario would occur when, after having started formal data collection, the case turns out not to be viable or to represent an instance of something other than what you had intended to study.
When you have only a score or so (20 to 30) of possible candidates that can serve as your cases (whether these candidates are “sites” or individuals or some other entity depends on your unit of analysis), the screening may consist of querying people knowledgeable about each candidate. You even may collect limited documentation about each candidate. To be avoided, at all costs, is an extensive screening procedure that effectively becomes a “mini” case study of every candidate case. Prior to collecting the screening data, you should have defined a set of operational criteria whereby candidates will be deemed qualified to serve as cases. If doing a single-case study, choose the case that is likely, all other things being equal, to yield the best data. If doing a multiple-case study, select cases that best fit your (literal or theoretical) replication design.
When the eligible number of candidates is larger, a two-stage screening procedure is warranted. The first stage should consist of collecting relevant quantitative data about the entire pool, from some archival source (e.g., statistical databases about individual schools or firms). You may have to obtain the archival data from some central source (e.g., a federal, state, or local agency or a national association). Once obtained, you should define some relevant criteria for either stratifying or reducing the number of candidates. The goal is to reduce the number of candidates to 20 to 30 and then to conduct the second screening stage, which consists of carrying out the procedure in the previous paragraph. Such a two-stage procedure was followed in a case study of local economic development, and the experience is fully reported in the companion text (Yin, 2003, chap. 6, pp. 9-14).
In completing the screening process, you may want to revisit your earlier decision about the total number of cases to be studied. Regardless of any resource constraints, if multiple candidates are qualified to serve as cases, the larger the number you can study, the better.
Source: Yin K Robert (2008), Case Study Research Designs and Methods, SAGE Publications, Inc; 4th edition.