In the field of educational research, one often encounters situations in which the goal is to study the association between two variables, by means of a 2 × 2 contingency table, in the presence of an additional variable that, when taken into account, may lead to a better understanding of the relationship between these variables. Situations of this kind are most commonly found in quasi-experimental studies, in which the researcher has a classification variable and wishes to study the association between two other variables for each level of the classification variable as well as in meta-analytic studies and, especially, in the analysis of differential item functioning (DIF). In these three contexts, there are always k levels, studies, or groups, respectively, and in each case, the ...