How Early Life Religious Exposure Relates to the Timing of First Birth

• DOI: http://doi.org/10.1111/jomf.12364
• Date: 01 October 2016
• Published date: 01 October 2016
• Author: Shannon N. Davis,Lisa D. Pearce

L D. P University of North Carolina at Chapel Hill

S N. D George Mason University∗

How Early Life Religious Exposure Relates to the

Timing of First Birth

This article examines intermediary processes

explaining how religious socialization and

involvement early in life are related to the

timing of rst births for women in the United

States. The theory of conjunctural action forms

the basis for hypotheses for how religious

schemas and materials operate to inuence

birth timing. Using the National Longitudinal

Survey of Youthdata and event history methods,

the study nds evidence for expected family

size, work–family gender ideology, educational

attainment and enrollment, cohabitation, and

age at marriage as mediators of associations

between early life religious exposure (afliation

and attendance) and the timing of nonmaritally

and maritally conceived rst births. These

ndings corroborate other research identifying

the long reach of religious socialization and

involvement in youth, elucidate some of the path-

ways for these connections, and motivate further

work to understand linkages between religion

and family behaviors in the United States.

In the past few decades, research on religion’s

relationship to childbearing has become more

Department of Sociology, Universityof North Carolina at

Chapel Hill, 155 Hamilton Hall – CB 3210, Chapel Hill,

NC 27599-3210, (ldpearce@unc.edu).

∗Department of Sociology and Anthropology, George

Mason University,4400 University Drive, MSN 3G5,

Fairfax, VA 22030.

Key Words: fertility, National Longitudinal Survey of Youth

(NLSY), religion.

sophisticated in its conceptualization of reli-

gious factors relating to fertility. Whereas earlier

studies restricted their examination of religion

to broad denominational differences in fam-

ily size (Ryder & Westoff, 1971; Whelpton,

Campbell, & Patterson, 1966), more recent

studies have shown differences across Protes-

tant denominations and considered factors such

as the frequency of people’s service attendance

and the declared importance of religion in peo-

ple’s lives (Hayford & Morgan, 2008; Mosher,

Williams, & Johnson, 1992; Zhang, 2008).

Their ndings suggest that religious traditions

may inuence adherents’ fertility in numerous

ways. For instance, belonging to a more prona-

talist religious tradition that emphasizes family

and parenthood and sanctions contraception or

abortion is associated with having more children

(Goldscheider, 1971; McQuillan, 2004), and

women who attend religious services more

frequently are likely to desire and have more

children (Hayford & Morgan, 2008; McQuillan,

2004; Zhang, 2008). Nevertheless, contempo-

rary studies of religion and fertility continue to

have important limitations in their theoretical

bases, outcomes of interest, and measurement

of religion and its mechanisms.

In their theoretical models, studies of reli-

gion and fertility have been largely motivated

by what Goldscheider (1971, p. 293) calls

the “supercial” and “inadequate” assumption

that it is primarily pronatalist religious values

that encourage pronatalist behavior. Gold-

scheider (1971, 1999) and McQuillan (2004)

have called for more encompassing theoretical

1422 Journal of Marriage and Family 78 (October 2016): 1422–1438

DOI:10.1111/jomf.12364

Religion and Timing of First Birth 1423

models attending to the wide variety of values

that religious institutions promulgate and the

means through which institutions promote and

reinforce these values.

In addition, studies of religion and fertility

have focused almost exclusively on completed

family size as the outcome of interest. The tim-

ing of rst birth has been relatively neglected,

although it carries signicant consequences for

women’s socioeconomic attainment and long-

term health and well-being, along with the health

and well-being of their children (Augustine,

Prickett, Kendig, & Crosnoe, 2015; Loughran

& Zissimopoulos, 2009; Royer, 2004; Williams,

Sassler, Addo, & Frech, 2015). As women who

have earlier rst births are likely to have more

births overall (Mills, Rindfuss, McDonald &

te Velde, 2011; Rindfuss, Morgan, & Swice-

good, 1988), studying the relationships between

religion and the timing of rst birth may addi-

tionally help us anticipate how macro-level

religious change might affect population growth

or decline.

Two analytical limitations also characterize

earlier studies of religion and fertility. First,

they usually rely on measures of religious

characteristics that were gathered after women

completed their fertility. As a result, it is unclear

whether ndings are evidence of religious

involvement’s inuence on fertility or if having

children increases religious afliation and prac-

tice (Stolzenberg, Blair-Loy, & Waite, 1995),

although results are often interpreted as the for-

mer. To better understand and estimate religion’s

effects on fertility, we need more studies that

measure religious afliation and involvement

earlier in the life course, before women have

begun having children. A second methodologi-

cal issue with studies of religion and fertility is

that they rarely conceptualize or test a compre-

hensive set of mechanisms for religion’s inu-

ence. Existing studies tend to control for as many

spurious factors as possible, such as education

or income, and then attribute residual religious

differences in family size to particularized,

pronatalist theologies (Goldscheider, 1971).

Rarely have studies incorporated measures of

ideology to directly test for mediation, or empir-

ically evaluated other mechanisms for religion’s

inuence such as educational attainment or

enrollment, employment, or union formation.

This article addresses these limitations by

applying the theory of conjunctural action

(Johnson-Hanks, Bachrach, Morgan, & Koher,

2011) to situate, integrate, and broaden prior

research on why young women’s early life

religious exposure might inuence the timing

of their rst births. Prior research shows that (a)

women raised in conservative Protestantism are

more likely to have a nonmaritally conceived

rst birth than others, (b) greater religious ser-

vice attendance early in life lowers women’s

risk of nonmaritally conceived birth, but (c)

greater religious service attendance hastens the

timing of women’s maritally conceived births

(Adamczyk & Felson, 2008; Pearce, 2010). We

hypothesize that family attitudes and experi-

ences with education, employment, and union

formation help explain these ndings.

We test our hypotheses with discrete-time

hazard models of women’s rst birth timing

using the rst 22 waves of the National Longi-

tudinal Survey of Youth 1979 (http://www.bls.

gov/nls/nlsy79.htm). We use a competing risks

framework to separately estimate the risk of

nonmaritally and maritally conceived rst

births, as religion and other covariates are

theorized to relate differently to each. In sum,

we provide a comprehensive conceptualization

and empirical analysis of how women’s early

life religious exposure comes to play a part in

the timing of their rst births, contributing to a

better understanding of the long and complex

reach of religious upbringing in the life course.

B

Theoretical Framework

We draw on the theory of conjunctural action

(TCA), a framework for the study of demo-

graphic behavior, to delineate the complex ways

in which early life religious exposure can affect

the timing of women’s rst births. TCA inte-

grates recent developments in cognitive science,

developmental psychology, social psychology,

cultural anthropology, and the sociology of cul-

ture (Johnson-Hanks et al., 2011). Following

Sewell (1992, 1999), TCA observes that soci-

eties are the sites of hundreds, if not thousands,

of patterned and overlapping social phenom-

ena, or structures, such as language, traditions,

or economic systems. Structures are sustained

through the repeated instantiations of schemas

(ideas, values, or concepts) in materials (external

resources such as objects or events). Individu-

als embody the social structures within which

they are embedded during their lives, forming