An empirical study of the relationships of Catholic practice and occupational mobility to fertility.

MPIRICAL research in differential fertility as in other areas operates with two general types of independent variables: hypotheses variables and control variables. Interest is focused on the relationship of the hypothesis variable and the dependent variable, but a careful testing of the hypothesis demands that the influence of other relevant factors, the control variables, be eliminated or held constant. These variables have their origin in deductive theory and/or in empirical fact. Deductive theory usually contributes more heavily to hypotheses variables; empirical generalizations to control variables. Accordingly, this section presents a brief review of some of the literature on differential fertility as a source of control variables. Four topics are considered: the Catholic-non-Catholic fertility differential within the United States, the declining importance of this differential, fertility differentials within the Catholic population, and finally, fertility differentials within the general American population. The Catholic-non-Catholic Differential. In an early correlational analysis of the relationship of religion and fertility, Hornell Hart concluded from a study in Iowa that the "tendency for married Catholics to have more children than married non-Catholics is offset by the fact that Catholics in Iowa