Widespread gloating erupted on the right last week when a paper by sociologists Brad Wilcox of the University of Virginia and Nicholas Wolfinger of the University of Utah showed that Republicans say they have happier marriages than Democrats. Predictably, that study richocheted around the Internet — but other researchers are raising hard questions about its validity,
Doubts arise because more extensive and reliable data have consistently showed that couples living in red states tend to have higher divorce rates than those in blue states – and higher divorce rates, of course, indicate fewer, not more, happy marriages.
David Leonhardt’s conclusion in The Upshot, based on Wilcox and Wolfinger’s analysis of county and household data, was that the differences among self-reported Republicans and Democrats could be based on culture.
Conservatives stress traditional admonishments to remain married through thick and thin, but conservative policies – promoted in many red states – don’t necessarily support strong marriages and families. Abstinence-only education can leave teens and young adults ill prepared for an adult sexual relationship, and at worst, may pressure young people into marrying an unsuitable partner.
As Emma Green observes in The Atlantic, communities shape people’s lives in untold ways. “Kids who are raised in two-parent households, for example, and whose friends are mostly raised in two-parent households, tend to fare better. Plus, the civic life and institutions of a place really matter; this is the Robert Putnam theory of stability and happiness. It’s possible that the happily married couples Wilcox and Wolfinger have identified are just living in places that are more conducive to happiness.”
Moreover, as noted in Wilcox and Wolfinger’s conclusion, Republicans are largely white, meaning that “they are less likely to face the discrimination, segregation, and poverty that minority couples often experience in America, all of which can compromise the quality of married life.”
The database for the study is quite small — especially contrasted with the results gathered by sociologists Jennifer Glass of the University of Texas at Austin and Philip Levchak at the University of Iowa, who also studied marriage at the county level, using a larger sampling of government data.
Glass and Levchak found that counties with the highest proportions of conservative and evangelical Protestants were also the counties with the highest divorce rates. This finding isn’t just an artifact of regional poverty — although conservative religious groups are usually concentrated in areas with higher poverty rates and lower wages — because Glass and Levchak controlled the results for income and geography; nor does it result from rates of marriage overall (as compared to cohabitation, which they found no link for); or even a regional culture that in their words, promotes “factors that destabilize marriage.”
High divorce rates among religious conservatives, they found, correlate closely with earlier ages of first marriage and lower educational attainment, which often translates to lower incomes. But religion wasn’t a factor in predicting divorce; the best predictor was where a couple lived.
“Living in a cultural climate where most people expect to marry young and there is little support from schools or community institutions for young people to get more education and postpone marriage and children” encourages people to get married early, say the researchers:
“Abstinence-only education, restrictions on the availability of birth control and abortion, support for marriage as the resolution of unexpected pregnancies, and distrust of secular education (especially higher education) among the populace in religiously conservative counties work to create an environment where young people of every religious belief – or none – tend not to pursue higher education or job training, and instead to engage in early marriage and child-bearing.”
Their analysis doesn’t explicitly study political affiliation — and it further distinguishes among Christians by their level of church affiliation. Those most active in church are less likely to divorce, while those who are “nominal” Protestants – attending services fewer than twice a month — fare much worse than both the actively religious and those who aren’t religious at all.
Sociologist Philip N. Cohen of the University of Maryland, College Park, reminds readers that “Republican” and “conservative” aren’t interchangeable. Indeed, Wilcox tells The Atlantic that conservatives, of course, aren’t only in Southern red states and vice versa. His study, which breaks down and in some ways replicates Wilcox and Wolfinger’s data, shows that its conclusions, in his words, are “not supported.” When Cohen expands the data to seven categories of political affiliation, not just three, he finds the left-right conclusion unsupported, since those identifying as strong Democrats also exhibit a high percentage of respondents claiming they are in a happy marriage. He also notes that Wilcox and Wolfinger don’t specify whether the Republican “advantage” is over Democrats or just everyone else.
In her Atlantic article, Green also acknowledges that the data are specious. “Most of all, pay attention to the warning labels all over Wilcox and Wolfinger’s research: ‘More research is needed to investigate the effect of ideology and region on family stability over the life course [italics hers].'” That line? Written by Wilcox and Wolfinger.
When analyzing marriage data, the truth isn’t always obvious – just like in marriage itself.
Photo: Is this something a Republican or Democrat would use? Does it matter? AForestFrolic/Flickr