College-educated women in the U.S. are the group least likely to divorce, yet when their marriages do end, they initiate the split at rates approaching 90%. The two facts sound contradictory. They aren't.
Education lowers the overall rate of divorce but concentrates who leaves, and why, in ways that reveal what marriage demands of women
seng kui Lim / Getty Images
College-educated women in the U.S. are the group least likely to divorce, yet when their marriages do end, they initiate the split at rates approaching 90%. The two facts sound contradictory. They aren't.
Michael Rosenfeld, a sociologist at Stanford University, found that women initiated 69% of all divorces, compared to 31% for men, using data from the How Couples Meet and Stay Together survey. Among college-educated women, that figure climbs to about 90%.
At the same time, researchers at the National Center for Health Statistics estimate that 78% of college-educated women who married for the first time between 2006 and 2010 could expect their marriages to last at least 20 years. Among women with a high school education or less, the share is only 40%, according to the Pew Research Center. Data from the Bureau of Labor Statistics confirm an inverse correlation between education and divorce: Among college graduates, 30% of marriages ended in divorce, compared with more than half among those without a high school diploma, according to a BLS analysis of longitudinal data.
So the paradox resolves into a question: Why does education suppress divorce overall while concentrating its initiation so lopsidedly among women?
Part of the answer is economic. College-educated couples tend to have higher household incomes, more accumulated assets, and children enrolled in expensive school systems. The opportunity costs of divorce are higher. Most studies estimate the magnitude of the household income drop for women at 23% to 40% during the year following divorce, according to a review published in the Journal of Marriage and Family. For women in higher-income marriages, the absolute loss can be larger.
Those costs function as a barrier. They suppress the overall divorce rate by keeping marginally dissatisfied couples together. But they also raise the threshold of unhappiness required to leave, meaning that the divorces that do occur among college-educated couples tend to reflect deep dissatisfaction rather than impulsive decisions.
A foundational study by sociologists Liana Sayer and Suzanne Bianchi at the University of Maryland found that "measures of marital commitment and satisfaction are better predictors of marital dissolution than measures of economic independence." Their analysis, published in the Journal of Family Issues, concluded that the independence effect found in prior research "may have been measuring the role of wives' economic independence in exiting bad marriages, not in exiting all marriages."
In other words, economic resources don't cause women to leave marriage. They enable women who are already unhappy to act on it.
The pattern is reinforced by who marries whom. According to a study published in the American Sociological Review, among couples who married between 2005 and 2009, more than 60% of those with different levels of education featured a wife who was more educated than her husband, an increase from about 35% in the early 1950s, according to researchers Christine Schwartz and Hongyun Han. Among couples who married in the 1990s or later, a wife's educational advantage was no longer associated with an increased risk of divorce, the American Sociological Association reported.
Schwartz, a sociologist at the University of Wisconsin-Madison, described these trends as "consistent with a shift away from a breadwinner-homemaker model of marriage toward a more egalitarian model." Educationally matched or female-advantaged couples, once divorce-prone, now form the most stable pairings.
Yet stability and satisfaction are different things. Rosenfeld's data showed that married women reported lower relationship quality than married men, whereas women and men in non-marital relationships reported equal levels of relationship quality. The gender gap in satisfaction exists specifically within marriage, not outside it. "I think that marriage as an institution has been a little bit slow to catch up with expectations for gender equality," Rosenfeld told the American Sociological Association.
The economic independence explanation, while necessary, is incomplete. Education also shapes expectations. Research published in the journal Demographic Research found that educated women are "able to bargain for a more even distribution of childcare and housework," which improves marital satisfaction. A 2024 study in the same journal found that "it is not the male partner's but rather the female partner's tertiary education" that predicts variation in both partners' gender role attitudes and day-to-day experiences of housework, according to researchers Liat Raz-Yurovich and Barbara Okun at Hebrew University, using U.K. data spanning 1992 to 2018.
College-educated women, in other words, enter marriage with higher expectations for equality. When those expectations collide with persistent imbalances, the resulting dissatisfaction is sharper. And because these women have the financial capacity to leave, they do. The overall divorce rate remains low because most of these marriages meet a high standard of quality. But the departures that do occur are overwhelmingly female-initiated, because the gap between what women expected and what they received is the primary fault line.
The paradox, then, is not a paradox at all. It is two effects of the same cause. Education raises the floor of marriage quality, helping keep divorce rates down. But it also raises the bar for what women expect from the institution, meaning that when a marriage fails to meet that standard, the woman is almost always the first to recognize it and act.
Join 500,000+ readers who start their day with Quartz.
By subscribing, you agree to our Terms of Service and Privacy Policy.