Abstract
This article attempts to fill two gaps in the literature on individual party choice. First, it deals with hitherto unanswered questions about relationships between social cleavages and party choice in Israel. Second, the article attempts to overcome methodological problems arising in the multivariate analysis of multiparty systems by utilizing Goodman's method of log-linear contingency table analysis. In the sample, occupation is not as strongly related to party choice as is a nonhierarchical dimension of economic position, sector of the economy. Ethnicity is modestly related to party choice, but hypotheses that the relationship is affected by place of birth, age, or other variables are disconfirmed. Hypotheses that the relationship between religiosity and party choice is affected by economic position are also disconfirmed. The advantages of using log linear contingency table analysis are demonstrated.