People in same-sex relationships tend to partner with individuals who are different from them in age, race, and education more often than people in different-sex relationships do. A recent study published in the Journal of Marriage and Family suggests that dating apps do not entirely explain these demographic differences, even though same-sex couples are much more likely to meet online. Instead of encouraging people to date broadly outside their age group, online dating actually tends to bring together individuals who are closer in age.
Homogamy is a sociological term for the tendency of individuals to form romantic relationships with people who share similar social characteristics, such as race, educational background, and age. This pattern plays a major role in society by concentrating resources like wealth and education within certain households. Most previous research on partner selection has focused on heterosexual populations.
Jisu Park, a research scientist at the Max Planck Institute for Demographic Research, conducted the study to understand the mechanisms driving partner selection among different types of couples. Prior evidence indicates that same-sex couples exhibit lower levels of homogamy than different-sex couples. They are generally more likely to form partnerships with people of a different race, a different education level, or a notably different age.
Park explored whether the ways people meet might explain this gap. Sexual minority individuals often face smaller, more geographically scattered pools of potential partners in their daily lives. Because of this, they rely heavily on digital platforms. Online dating could theoretically reduce partner similarity by exposing users to a wider, more diverse pool of people. On the other hand, the detailed search filters and algorithms on dating apps might help users seek out partners who perfectly match their preferences, potentially increasing similarity.
To test these ideas, Park analyzed data from the 2017 How Couples Meet and Stay Together survey. The nationally representative sample included 3,292 English-speaking adults in the United States. Within this group, 2,994 respondents were in different-sex relationships and 298 were in same-sex relationships.
The survey provided information on whether the respondents met their current or most recent partner through an online dating platform. Park also measured whether the partners matched in three specific areas: race, educational attainment, and age. Age similarity was divided into three categories: a gap of zero to two years, a gap of three to five years, and a gap of six or more years. Statistical models accounted for several variables, including the respondent’s gender, employment status, household income, marital status, relationship length, age, race, and education.
The initial analysis highlighted demographic differences in how couples form. Nearly one-third of same-sex couples, about 31.9 percent, met their partner online. This compares to just 10.1 percent of different-sex couples.
Same-sex couples were also less similar to their partners across all three measured categories. About 30.2 percent of same-sex couples were in interracial relationships, compared to 19.9 percent of different-sex couples. Large age differences were also more common in these partnerships. Around 41.3 percent of individuals in same-sex relationships reported an age gap of six or more years, compared to 24.8 percent of people in different-sex relationships. Similar patterns emerged for educational differences, with same-sex couples having roughly 38.8 percent higher relative odds of partnering across different education levels.
When looking at whether online dating explained these gaps, Park found that digital platforms did not account for the differences in racial or educational matching between the two groups. Meeting online did not show a statistical association with whether couples shared the same race or education level on a broad scale.
However, online dating did play a role in age matching. Meeting a partner online was associated with roughly 37.4 percent lower odds of having an age gap of six or more years. Because same-sex couples are highly likely to meet online, and online dating tends to match people closer in age, digital platforms actually suppressed the overall age gap. If same-sex couples did not use dating apps at such high rates, their age differences would likely be even larger than what the data currently shows.
Park also looked at how online dating operates differently depending on the type of relationship. For same-sex couples, meeting online was associated with a higher probability of being in an interracial relationship. For different-sex couples, meeting online showed no such association with racial mixing. Regarding age, online platforms were linked to smaller age gaps for both types of couples, though the specific mathematical categories varied slightly between the groups.
The study relies on observational data, meaning it cannot confirm that online dating directly causes changes in partner preferences. It is possible that individuals who choose to use dating apps already possess different attitudes about dating outside their race or age group compared to people who meet partners in person. The specific features of dating apps, such as how profiles display age and education, might influence choices, but the available data cannot separate platform design from user preference.
The relatively small number of same-sex couples in the dataset limits the statistical power of the analysis. A smaller sample makes it harder to detect nuanced differences, such as variations between gay male and lesbian couples, who often exhibit different relationship patterns.
Future studies could examine how age and generational changes influence digital dating habits. Online dating has become much more normalized over the past two decades. People who grew up with dating apps might show different partner selection patterns compared to older generations who adopted the technology later in life. Additionally, exploring how different rates of relationship breakups affect the demographic makeup of long-term couples would provide a more complete picture of modern romance.
The study, “How I Met My Partner: Online Dating and the Homogamy Gap Between Same-Sex and Different-Sex Couples,” was authored by Jisu Park.