A POSITIVE MODEL OF THE DETERMINATION OF RELIGIOUS AFFILIATION

Objective. This paper integrates, for the first time, two separate str ands of the literature on the determinants and effects of religiosity. Of primary importance is the relationship between income and religiou s affiliation. Methods. Data are a pooled cross section, with states t he units of observation for 1952, 1971, and 1980. Per capita income le vels and percentage membership in four groups (fundamentalist Protesta nt, nonfundamentalist Protestant, Catholic, and Jewish) are estimated simultaneously. Results. Empirical results for the income equation ind icate that states with larger Jewish populations are associated with h igher per capita income levels, while states with larger fundamentalis t Protestant populations are associated with lower income levels. For the religious membership equations, income is a significant determinan t of variations in membership for both Protestant groups: increases in per capita income are associated with lower membership levels in both . This latter result is consistent with Iannaccone's (1988) sect/churc h theory, which explains the historical Protestant switching pattern. Conclusions. Further research is necessary to answer questions regardi ng the differing impacts of other socioeconomic variables on denominat ional choice. In addition, newly available religious membership data w ill provide answers regarding recent changes in religious choice and d enominational mobility.