Jeremy E. Uecker
8 min
This study examines why some young adults in the United States marry early, despite broader social trends toward delayed marriage. Specifically, it investigates how religion acts as a cultural institution—encompassing both symbolic schemas and material practices—to influence the timing of marriage. The author explores how an individual's religious affiliation, service attendance, and beliefs, alongside the religious context of their high school, contribute to the decision to marry early.
Using longitudinal data from the National Longitudinal Study of Adolescent Health (Add Health), the study conducts multilevel event-history analysis on a sample of over 14,000 young adults followed from adolescence through their late twenties. The research models the hazard of first marriage by incorporating individual-level religious characteristics (tradition, attendance, salience, and belief in scriptural inerrancy) and school-level religious context. To understand the mechanisms at play, the author tests several mediators, including educational attainment, beliefs about money, and attitudes toward cohabitation and marriage.
Mormons and conservative Protestants are the most likely to marry early, significantly outperforming other religious groups and the nonreligious. The study finds that religion influences this timing through two main channels: material practices (like attending religious services) and schematic orientations (like believing in scriptural inerrancy). Surprisingly, religious context matters: even when controlling for an individual’s own religiosity, students attending high schools with a higher percentage of conservative Protestants and Mormons are more likely to marry early themselves. The author suggests that these peer environments may facilitate marriage by providing a social structure where early marriage is viewed as normative. While individual religious effects are partially explained by attitudes toward cohabitation and marriage, the influence of school-level religious context remains robust, suggesting it may operate through structural network effects rather than just individual belief changes.
Understanding the drivers of early marriage is critical because marrying young is often associated with lower educational attainment, reduced earnings, and decreased marital stability. By identifying religion as a significant cultural force that shapes these life-course decisions, the study moves beyond purely economic explanations of marriage timing. It highlights that the cultural environments—including school networks—that adolescents inhabit can have long-lasting effects on their family formation paths.
Early marriage has important consequences for individuals in the United States. Several studies have linked religion to early marriage but have not examined this relationship in depth. Using data from Waves 1, 3, and 4 of the National Longitudinal Study of Adolescent Health, I conduct multilevel event-history analysis to examine how religion, at both individual and contextual levels, is associated with early marriage. Further, I test mediators of the religion-early marriage relationship. I find significant variation in early marriage by religious tradition, religious service attendance, religious salience, belief in scriptural inerrancy, and religious context in high school. The individual religious effects—but not the school context effects—are explained in part by differential attitudes toward marriage and cohabitation.
Sam: It follows adolescents into adulthood using the National Longitudinal Study of Adolescent Health, called Add Health. The analysis includes about fifteen thousand people. They came from a nationally representative school-based study involving one hundred thirty-two schools.
Alex: Following people helps establish sequence. How does the study turn those histories into a comparison?
Sam: It uses event-history analysis: modelling whether first marriage occurs in each year someone remains unmarried. After marriage, that person leaves the risk set. The models also account for students being clustered within schools, rather than treating every observation as independent.
Alex: And the religious measures come before those marriages, rather than asking married adults to reconstruct their teenage beliefs?
Sam: Religion is measured during adolescence. The author then adds dimensions in stages: affiliation, service attendance, and religious salience—how important religion feels to someone. Another measure is scriptural inerrancy: believing that one’s sacred scriptures contain no mistakes.
Alex: The sequence should show whether an apparent denomination difference is really a difference in religious involvement. What happens?
Sam: Initially, Mormons had more than twice the yearly odds of marriage compared with nonreligious respondents, after individual controls. Conservative Protestants had about thirty-seven percent higher yearly odds than that same comparison group. Mainline Protestants, Catholics, Black Protestants, and Jewish respondents did not differ significantly from the nonreligious at this stage.
Alex: Those are yearly odds among people who have not yet married—not a statement that twice as many eventually marry.
Sam: Correct, and the differences change as the models expand. Adding attendance reduces the affiliation associations. Once attendance and religious importance are both included, conservative Protestants no longer differ significantly from nonreligious respondents.
Alex: Does that mean denomination contributes nothing once you account for commitment?
Sam: No, because the comparison matters. Conservative Protestants still marry earlier than mainline Protestants and Catholics in the adjusted analyses. Mormons also remain distinct, although their association becomes smaller and only marginally significant once family attitudes enter.
Alex: What about scripture beliefs? Are they simply another way of measuring religious commitment?
Sam: They carry an additional association. Belief in inerrant scripture was linked to about eighteen percent higher yearly marriage odds, after accounting for affiliation, attendance, and religious importance. Adding that belief also made the association with reporting religion as very important statistically nonsignificant.
Alex: Let’s turn to the school finding. Does “religious context” mean attending a religious school?
Sam: Not just that. The study measures schoolmates’ affiliations, attendance, religious importance, and scripture beliefs, alongside religious-school attendance. When those context measures enter together, the significant association is the share of schoolmates who were conservative Protestant or Mormon.
Alex: Could that simply reflect religious students choosing schools full of similar students?
Sam: The models account for respondents’ own religious characteristics and several demographic, family, and school characteristics. The schoolmate association remains. But these are observational comparisons; adjustment does not establish that schoolmates caused anyone to marry earlier.
Alex: So there’s an association beyond measured personal religion, but not a demonstrated peer effect. What happens when the author tests explanations involving education, money, and family life?
Sam: College attendance and placing more importance on money for a successful relationship are associated with later marriage. But they explain little of the religious differences. Family attitudes do more: accepting cohabitation without marriage plans predicts later marriage, while expecting to marry young predicts earlier marriage.
Alex: That sounds consistent with religion shaping how people view marriage and its alternatives. Do actual cohabitation and sexual histories support that explanation too?
Sam: Not in the same way. Neither premarital cohabitation nor premarital sex significantly predicts earlier marriage in these models. Adding family attitudes removes the significant association with weekly service attendance and reduces several other religious associations, without explaining them all.
Alex: And the school association survives those attitudes. Does the author think schools change what students believe, or whom they can marry?
Sam: The author proposes that religious environments might facilitate matching by exposing people to potential partners willing to marry young. But the document explicitly calls that explanation speculative. These data do not test the matching mechanism.
Alex: The individual-attitude explanation also needs the timing to work. Were those attitudes always measured before marriage?
Sam: No, and that is the central limitation for interpreting them as an explanation. Some cohabitation attitudes were measured after respondents had already married. Marriage could have changed those attitudes, so the analysis may overstate how much they explain religious differences.
Alex: Then we should trust the description of linked patterns more than the proposed direction of influence.
Sam: That’s my reading. The longitudinal sample, multiple religious measures, and school-level analysis provide substantial evidence of associations. They do not settle whether religion changes attitudes, attitudes guide marriage, or unmeasured differences explain part of the pattern.
Alex: Who should read the full document, and where should they start?
Sam: Researchers studying family formation, religion, or social networks should read it. Start with the Results comparisons that add attendance and religious importance, then family attitudes. Those show why affiliation alone is an incomplete measure. Then read the Discussion’s measurement cautions, especially the timing of cohabitation attitudes and the speculative explanation for school context.
Alex: And what should everyone else remember?
Sam: Religion’s link to early marriage is not just about what someone believes; it also reaches into their social surroundings.
Alex: Keep the beliefs and the surroundings in view.