For more than a century, sociologists have predicted a decline in faith. Measuring religious belief is very difficult, but what we do know is that Americans born after 1980 are less likely to identify with a religion than were members of any previous generation at the same age [1]. It remains to be seen, though, whether their beliefs will change in adulthood: members of other generations have become more religious after marrying and having kids [2].

If Millennials do eventually become religious, they may reap some unexpected fringe benefits. For one thing, compared with the nonreligious, religious people report more satisfaction with their love lives [3] and sex lives [4]. Frequent church attendance has also been associated with lower rates of smoking and drinking, a greater tendency to exercise [5], reduced risk of cancer, and improved cardiovascular health [6]. A research review conducted on behalf of the National Institutes of Health went so far as to declare that “church/service attendance protects healthy people against death” [7].

So religion affects the course and even duration of your life. What about the converse: How does your life—specifically your early life—affect your religion? Over the decades, studies have identified a number of contributors to religiosity, many of them unsurprising—like attending religious services as a child [8]. Family also matters a lot: studies have found that kids with married parents of the same faith are most likely to keep the faith they grew up with [9]. Those with divorced parents, however, are less likely than their peers to affiliate with a religious institution (though they’re no less likely to feel “closeness to God”) [10]. And kids of certain faiths, such as young Mormons, evangelicals, and Jews, are more likely to stay in the fold than are those of other denominations [9].

Other studies have added some surprising factors to this list. For one thing, birth order seems to matter: middle children are more likely than their older or younger siblings to lose their attachment to religion as they get older. Perhaps this is because of underlying personality differences. Compared with their brothers and sisters, these kids were rated more “rebellious” as well as less “agreeable” and “conscientious”—two traits the authors associated with religiosity [11]. In another twist, although those with bachelor’s degrees have historically had lower levels of religious affiliation, a study published this year found that this isn’t true for people born after the 1950s. In fact, college grads born in the 1970s are more likely than nongrads of the same age to identify with a particular faith [12]. Maybe there’s something about contemporary campus life that makes people more, not less, likely to gravitate toward traditional institutions—or maybe college grads have simply learned that religion is pretty good for you.

The Studies:

[1] Taylor et al., “Millennials in Adulthood” (Pew Research Center, March 2014)

[2] Wilson and Sherkat, “Returning to the Fold” (Journal for the Scientific Study of Religion, June 1994)

[3] Neto and da Conceição Pinto, “Satisfaction With Love Life Across the Adult Life Span” (Applied Research in Quality of Life, April 2014)

[4] Neto, “The Satisfaction With Sex Life Scale” (Measurement and Evaluation in Counseling and Development, Jan. 2012)

[5] Strawbridge et al., “Frequent Attendance at Religious Services and Mortality Over 28 Years” (American Journal of Public Health, June 1997)

[6] Hoffa et al., “Religion and Reduced Cancer Risk” (European Journal of Cancer, Nov. 2008)

[7] Powell et al., “Religion and Spirituality: Linkages to Physical Health” (American Psychologist, Jan. 2003)

[8] Merino, “Irreligious Socialization?” (Secularism & Nonreligion, Jan. 2012)

[9] Bengtson et al., Families and Faith: How Religion Is Passed Down Across Generations (Oxford, Nov. 2013)

[10] Zhai et al., “Parental Divorce and Religious Involvement Among Young Adults” (Sociology of Religion, Summer 2007)

[11] Saroglou and Fiasse, “Birth Order, Personality, and Religion” (Personality and Individual Differences, July 2003)

[12] Schwadel, “Birth Cohort Changes in the Association Between College Education and Religious Non-Affiliation” (Social Forces, Aug. 2014)