On July 24, New York will join the league of states that allow gay marriage. Meanwhile, demographic data show that this group is already united in another significant way: lower-than-average divorce rates. Interesting, but does this mean that same-sex marriages in New York will last longer? Are the two characteristics even related? Perhaps, as data show that factors like education level and marriage age tend to be related to both a state's divorce rate and its stance on same-sex marriage.
According to provisional data from the Census Bureau and the Centers for Disease Control's National Vital Statistics System, 5 of the 10 states, plus the District of Columbia, with the lowest divorce rates per thousand people (of the 44 states, plus D.C., that had available data) are also among the nine jurisdictions (a group that includes eight states and the District of Columbia) that currently perform or recognize gay marriages. Of course, states with more marriages naturally have more chances for divorce. But the trend also holds up when one looks at divorces as a share of marriages. In states that recognize or perform gay marriages, the number of divorces in 2009 was 41.2 percent of the number of marriages. In the 36 other states for which 2009 data are available, it was 50.3 percent. Remove the outlier Nevada, the state with by far the lowest divorce rate by this metric (16.3 percent), likely due in part to Las Vegas's status as a wedding hotspot for out-of-state couples who may get married there but divorced elsewhere, and the figure jumps to 53.2 percent.
In early 2010, the New York Times's Nate Silver (then writing at his blog, fivethirtyeight.com) analyzed state divorce rates over time and noted a similar correlation. He found a statistically significant relationship between states' gay marriage laws and changes in divorce rates over a five-year span.
But the above data is from 2009--a year in which only two states performed gay marriages for the full year, joined later by Iowa and Vermont. And Gary Gates, Williams Distinguished Scholar at the UCLA School of Law's Williams Institute, a think tank that works to advance sexual orientation law, says that same-sex couples divorce at rates comparable to those of different-sex couples. Data from Massachusetts, the state with the longest track record on gay marriage, seems to support this. The numbers show that divorce rates in that state have not changed since same-sex unions became legal in 2004. Though it fluctuated in the intervening years, the rate of 2.2 divorces per 1,000 people in 2004 was the same in 2009.
So why does there seem to be correlation but no causal linkage? The answer is that there appear to be many other related factors at play. Untangling the correlation between gay marriage and divorce rates means examining other interrelated demographic factors that are associated with a state's politics as well as its residents' propensity to divorce. Two chief explanatory variables among these are marriage age and education.
According to D'Vera Cohn, a senior writer at the Pew Research Center, there is a link between education, marriage age, and divorce rates. "From what I understand from the research, people who take the time and have the discipline to complete a college degree or more advanced education--those personality traits are also useful in sustaining a marriage," she says.
Some studies have also shown rates of higher education to be related to gay marriage support. A 2009 University of Florida study, for example, found that for every additional 1 percent of a given county's population with a bachelor's degree, there was a nearly equal decrease in support for an amendment that would have enacted a gay marriage ban. Daniel Smith, one of the study's authors, posited that education "increases exposure to those who are different" and added, "Studies show very clearly that the more educated people are, the more tolerant they are of differences." Indeed, according to data from the Census Bureau's 2009 American Communities Survey, of the 10 states (plus the District of Columbia) with the highest rates of residents 25 and older with bachelor's degrees or higher, seven allow or recognize gay marriage.
Higher education levels are also associated with later marriage, another key factor that links to states' residents' politics and divorce rates alike. A 2009 Pew Research Center study authored by Cohn found that states with high shares of college-educated adults also have populations that tend to marry at an older age. The study found that residents of states with high shares of Democratic votes also tended to marry at older ages than states with low Democratic vote shares.
Census data also bear out this point. Of the 10 states with the highest median age for males at their first marriage, eight recognize or perform gay marriages, with median ages between 29.3 (Maryland) and 31.5 (D.C.). Iowa is the outlier here, with a median age of 26.9. Altogether, these statistics point to a relationship between older marriage age, low divorce rates, and liberal views on same-sex marriage.
Incidentally, same-sex couples who choose to marry also tend to be older than newly married, opposite-sex couples and same-sex married couples on the whole also tend to have higher levels of education than those in different-sex marriages, according to Gates. So does this mean that a slew of new marriages between older, well-educated, same-sex couples will drive down New York's already low divorce rate? Not so fast, says Gates. "There are 100 [different-sex] married couples for every 1 [same-sex] couple," Gates told U.S. News in an E-mail. "Even if all of the [same-sex] couples marry, at only 1 percent of married couples, there are just not enough [same-sex] couples to really alter broad statistics about marriage and divorce."