Sources of Influence and Influences of Agency 163
better understand the connection between aging, life course transitions, and religious
understandings and commitments.
Sociological investigations of religious socialization are also underdeveloped in how
they address denominational and congregational influences. Few studies explore dis-
tinctive religious preferences of particular denominations, and there are no studies
demonstrating congregational influences on individuals’ preferences. While congrega-
tional studies have proliferated in religious sociology, it has generally meant a shift
of focus to the organizational level of analysis. Ideally, we would have multilevel lon-
gitudinal data that would allow us to sort out the impact of family, congregations,
denominations, and peer influences. However, this is a tall order to fill in an era of de-
clining research support and in a subspecialty with an applied focus and strong religious
agendas in many funding agencies. To explore the nuances in religious understanding
and commitment, systematic ethnography would be ideal. We do have a few good ex-
amples, largely on socialization into new religious movements (e.g., Rochford 1985),
but most ethnographic treatments in the sociology of religion have failed to deal with
issues of socialization and tend to lack a rigorous approach to sampling and interview-
ing. Also, there are no longitudinal ethnographic works on religious socialization or
commitment over the life course (but see Dillon and Wink, Chapter 14, this volume).
Gender differences in socialization are also of immense importance. Scholars have
long assumed that gender differences in religiosity are a function of variations in so-
cialization, and that gender divides spheres of influence among parents (cf. Nelsen
and Potvin 1981; Suziedelis and Potvin 1981; Acock and Bengtson 1978; De Vaus and
McAllister 1987). Yet, no study has rigorously tested this – particularly by investigating
the effects of specific socialization efforts on siblings. Recently, scholars have claimed
that gender differences in religiosity may instead be a function of risk preferences that
may or may not be a product of socialization (Miller and Hoffman 1995; Miller and
Stark 2002). This is an intriguing proposition, which also calls into question the scope
of socialization models for explaining individual differences in religiosity. Perhaps in
the future we will be able to investigate further the biopsychosocial foundations of
religiosity (Gove 1994; Stark 2000). Such a perspective may well be a valuable tool for
explaining gender and sexuality differences in religious commitment.