A Bayesian approach to Configural Frequency Analysis

Configural Frequency Analysis (CFA) is a method for cell-wise inspection of cross-classifications. CFA searches for types, that is, patterns of variab le categories that occur more often than expected from some chance model, a nd for antitypes, that is, patterns observed less often than expected. Thus far, CFA has been plagued by the difficulties involved when looking for pa tterns of types and antitypes. This article introduces Bayesian CFA. Using Bayesian CFA one can (1) search for types and antitypes as before with the advantage that adjustment of the experiment-wise significance level ct is n ot necessary; and (2) test whether groups of types and antitypes form compo site types or composite antitypes. This option is crucial when patterns of types or antitypes must exist for a concept to be retained. Empirical examp les use data from alcohol research and from sleep research to illustrate bo th new options. Characteristics of Bayesian CFA and extensions are discusse d.