However, case studies do not strive for this type of external validity. There are two types of generation: statistical and theoretical. Statistical generalization is achieved by using representative randomsamples. On the basis of statistical probability we generalize our findings to a wider population that our sample is designed to represent.
Theoretical generalization involves generalizing from a study to a theory. Rather than asking what a study tells us about the wider population (statistical generalization) we ask, ‘What does this case tell us about a specific theory (or theoretical proposition)?’ In Chapter 13 I argued that case study designs are fundamentally theoretical. They are designed to help develop, refine and test theories. They do this using the logic of replication.
Replication
Tounderstand the concept of theoretical generalization it is necessary to understand the logic of replication in research design (Yin, 1989). This is the logic that lies at the heart of generalizing from experiments. Since experiments typically do not use representative probability samples they provide no basis for statistical generalization to a wider population. An experiment might employ an excellent design that ensures a high degree of internal validity but unless it is based on a probability sample we do not know if those findings will occur in the wider population. There may be something about the particular sample that means that the findings will apply only to the people who participated in the experiment.
Experiments using non-probability samples argue for external validity on the basis of replication logic. That is, if the experiment can be repeated again and again under the same conditions and produce the same results we can be confident that the experimental results will hold up more