Over the past 50 years, the educational makeup of romantic couples has changed due to rapid increases in college enrollment and shifts in the gender balance of graduates. An analysis published in Research in Social Stratification and Mobility reveals that overall growth in education levels has as much of an impact on marriage patterns as the fact that women outnumber men in higher education.
For decades, social norms and structural barriers often resulted in partner matching where husbands had more formal schooling than wives. Sociologists call this dynamic hypergamy. Conversely, if the wife has more formal education than her husband, it is known as hypogamy. If the partners have the same educational background, the relationship is said to be homosexual.
Demographic changes in marriage selection have real-world implications for nations. How partners are matched can have a significant impact on economic inequality across society. Strict marriage among highly educated and high-income people naturally tends to widen the gap in household wealth between different social classes.
During the second half of the 20th century, the educational landscape changed dramatically in many Western countries. The total number of people attending university has expanded globally. At the same time, the traditional gender gap has been reversed. Women began earning college degrees at higher rates than men, fundamentally changing the pool of candidates available to young people looking for a partner.
Previous demographic research suggests that this reversal of gender differences is the main reason behind the reported increase in nonmarital relationships and apparent decline in hypergamous relationships. Julia Riesch of the Max Planck Institute for Demographic Research and Jan Skopek of Trinity College Dublin designed a study to test this story. They wanted to separate the personal effects of reversing the gender gap from the effects of general educational expansion.
To make a comprehensive comparison, researchers analyzed data from two countries with vastly different histories of public schooling. In the United States, a decentralized, demand-driven system led to a significant expansion of secondary education in the early 20th century. France relies on a formally stratified and highly centralized educational structure. Mass education was widely spread throughout the French population much later, primarily after World War II.
The researchers analyzed census and census records spanning 1962 to 2011 in France and 1960 to 2015 in the United States. Their analysis focused on the demographic group of partnered women between the ages of 25 and 34. This particular age group was chosen to capture statistical trends in first-time unions. For comparison purposes, the researchers divided educational background into three simple categories: less than middle school education, completed secondary school education, and completed college education.
To disentangle the overlapping trends of expanding degree programs and changing gender balance, the researchers used a statistical approach called counterfactual decomposition. This analysis tool relies on mathematical “what-if” scenarios. Riesch and Skopek calculated the outcome of a hypothetical marriage by keeping certain variables strictly constant while mathematically varying other variables over time.
For example, their model estimated what marriage patterns would be like if general educational attainment still increased, even though the statistical association between gender and education remained frozen in the 1960s. This method allowed the authors to identify the extent to which changes in romantic partnerships were related to the sheer volume of university degrees and the specific influx of female graduates.
The study highlighted several different dynamics between the two countries. In the United States, about 62 percent of young women in 1960 had exactly the same educational level as their partners. By 2000, this same-sex marriage rate had increased to nearly 71 percent and remained stable for several years.
This American trend was largely driven by changes in the relationship between gender and education, along with changes in the way people intentionally match candidates. The structural expansion of education was actually creating a countervailing effect behind the scenes. As fewer Americans ceased education at the basic secondary level, the bond between two uneducated individuals decreased significantly. This mathematical decrease was completely offset by a proportional increase in the bond between two highly educated individuals.
In France, a different statistical pattern was observed for couples matched on education level. The prevalence of same-sex marriage in France showed a U-shaped trend over the 50-year observation period. It first declined in the second half of the 20th century, but rose steadily again in the following decades.
The researchers found that the overall expansion of education in France acted as the main driver of this U-shaped curve. The initial sharp decline in marriages between less educated people outweighed the earlier slower increase in marriages between college graduates. By the 1990s, the burgeoning number of highly educated couples finally overtook the losses at the lower end, and the combined rate rose again.
When the researchers looked more closely at the relationship between educational level and inequality, they observed additional nuances. In the United States, the proportion of women who are more educated than their partners has reversed. This rate declined until the 1980s, then rose again towards the present day. At the same time, the proportion of women who were less educated than their partners began to decline markedly from the 1970s onwards.
In both countries studied, the reversal of the gender gap directly contributed to the overall decline in hypergamy. Less educated women are less likely to partner with more educated men simply because the availability of such candidates has changed. However, simulation models reveal that generalized educational expansion acts as a countervailing force against these trends.
Calculated estimates show that if overall educational attainment had increased in the United States, absent historical changes in the sex ratio, the country would actually have seen a structural decline in hypogamy. The sheer amount of degrees available has led to a downward trend in mathematics. A concrete increase in the number of female graduates also boosted it. These two distinct demographic currents often functioned in stark opposition.
To ensure the reliability of these numbers, our demographics team performed several sensitivity analyses. They expanded the mock date to include men up to 39 years old, with the aim of accounting for larger age differences in relationships. They also tested a more detailed four-level classification system for formal degrees. Across all alternative tests, the main conclusions remained statistically unchanged.
The authors pointed out that structural opportunity changes in the dating market do not operate in a social vacuum. Changing cultural tastes, the changing wage gap between men and women, and growing inequality within companies are also playing a compounding role in how modern couples choose each other. Factors such as the modern rise of online dating have dramatically changed the way strangers interact with each other and changed the geographic boundaries of the typical partner market.
Although the models utilized accounted for existing assortative mating patterns, the general public continually adapts their behavior as the direct dating pool changes. One of the limitations of the research framework lies in the basic assumption that individuals continually seek partners within a relatively narrow age range.
If you suddenly find yourself short on potential spouses with matching educational backgrounds, you might permanently expand your search to older or younger people. Future demographic research will need to map precisely how adults creatively adjust their search parameters across different generations in response to changing candidate pools.
Still others may choose to postpone cohabitation or marriage altogether until they find their ideal partner. Because this published study focused only on a specific snapshot of women, widespread delays in union formation may ultimately move the couple’s apparent indicators outside of the age window analyzed. The researchers noted that changes in international migration are also having an impact on the local single population.
Separating the effects of degree expansion from the effects of changing gender inequality remains a theoretical societal challenge, as both historical events occurred in parallel. Still, the statistical tools introduced in this analysis provide an objective window into deciphering the underlying mathematics at work behind modern romantic partnerships. The study, “Fifty years of marital sorting in France and the United States – The role of educational expansion and changing gender imbalances in education,” was authored by Julia Riesch and Jan Skopek.
Heading options
- How educational trends shape marriage partners
- The mathematics behind 50 years of modern love history
- Does a college degree change young people’s dating choices?
- How the surge in female graduates has changed the partner market
- Why couples still tend to match on educational attainment
- Tracking 50 years of demographic change and romantic partnerships
- Unraveling the relationship between secondary education and modern marriage
- Structural data that determines who to choose as a mate
- How reversing the gender gap in education will impact modern dating
- Uncovering the demographics that dominate the dating market
