At the risk of over-generalization, the first half of the twentieth century in the United States was marked by an increase in conformity and attachment to institutions. Rates of marriage and membership in labor unions and churches were on the rise, while income inequality and political polarization were falling. In the late twentieth century, those trends went into reverse. The American public became more divided in opinions and behavior along several different lines, and family patterns diversified.

In last month’s Annals of the American Academy of Political and Social Science, UCLA sociologist Megan Sweeney documented one facet of how the rise and fall of conformity played out in family life: variability in marriage timing. As you might surmise, the term refers to the range of ages at which people marry for the first time, which scholars sometimes measure by examining the ages at which 25 percent, 50 percent, and 75 percent of people born in a certain year had married. A small range between those milestones implies a “standardized” life course; a large range implies more diversity, whether due to a greater scope for individual choice or to structural factors like a scarcity of stable jobs.

Using data from the General Social Survey and the National Survey of Family Growth, Sweeney found that variability in marriage timing declined in the early twentieth century, and then began climbing quickly. For instance, the “cohort duration of marriage”—the average time it took for each birth year’s population to go from 25 percent married to 75 percent married—of men born between 1915 and 1919 was seven years: men typically married between the ages of 21 and 28. The cohort duration of men born between 1940 and 1944 was down to five years. And then the trend turned around. For men born between 1950 and 1954, the cohort duration of marriage was nine years.

Women tended to marry within a narrower (and younger) age range in the early part of the twentieth century, but they followed a similar pattern. The cohort duration of marriage for women born between 1915 and 1919 was five years, and then four years for several subsequent cohorts. Things changed beginning with women born in the 1940s, and cohort duration doubled in the course of three decades. Women born in the 1940s typically married between approximately 19 and 24 years of age. The median age at marriage increased and the range expanded in the decades that followed. Twenty-five percent of women born between 1970 and 1974 were married by age 21.4 years, and 75 percent were married by age 31.8, for a cohort duration of 10.4 years.

Further analyses of white men and women showed that the trends in marriage timing were similar across all social classes, as measured by several individual and parental indicators. The story about how members of various social classes time their first cohabiting union and first birth would likely be different, Sweeney noted, and could still be evolving.

Unfortunately, data limitations made it impossible to look beyond the 1950-54 cohort of men and the 1970-74 cohort of women. Based on the smaller changes between the most recent female cohorts she examined, Sweeney suggested that for variability in marriage timing was leveling off for those born in the 60s and 70s. But with the median age at first marriage continuing to climb, only a future analysis will reveal for sure where it stands now.

What should we make of these trends? On the one hand, as education and career options multiply, different individuals meet the right person in the right circumstances at different times, so it’s natural and fitting that people marry at a wider range of ages these days than their grandparents did in the industrial 1960s. On the other hand, many men and women who put off marriage are not putting off parenting, with all the consequences for their child(ren) and their relationships that that implies. The question we’re left with, then, is how we can combine family stability for kids with today’s unstable economy and the expectation that young people will experiment with intimate relationships, lifestyles, and jobs for a while before tying the knot—and there’s no easy answer to that.