Yesterday, the Washington Post’s website published a piece by W. Bradford Wilcox and Robin Fretwell Wilson in which the duo writes that “Married women are notably safer than their unmarried peers, and girls raised in a home with their married father are markedly less likely to be abused or assaulted than children living without their own father.” The authors never explicitly say that they think women should get married as a means of protecting themselves and their daughters (or potential daughters) from rape or violent assault, but that seems to be the implication. And it’s a bit crazy.
There are other, wonkier outlets better equipped to fully pull apart the numbers here, but it’s worth at least highlighting the ur-error from which all of Wilcox and Wilson’s other errors spring: mixing up correlation and causation. Anyone who has hung out with social-science researchers or perused their literature has come across this concept, and it’s pretty simple: Just because two things seem to have a numerical relationships to one another does not mean one causes the other.
Let’s say you pick a parking lot and track the number of shadows cast across it every day over the next year, as well as daily precipitation. You’ll quickly find an inverse relationship between the number of shadows cast and rainfall: the more shadows, the less rainfall. So shadows must prevent rainfall! Except obviously not: Shadows are cast when it’s sunny out, and when it’s sunny out it doesn’t tend to rain.
This is the sort of error Wilcox and Wilson are making. Taking their numbers at face value: Sure, maybe women who are married are less likely to be assaulted or raped than ones who are not, but there are a million possible reasons for this that aren’t directly tied to marriage itself. Women who are married tend to be older, for instance, than their unmarried counterparts, and older women are less likely to be targeted by rapists. Women who are married are probably less likely to be in settings in which rape by a stranger is likely to occur — the researchers themselves point this out — because most of their time is spent in a family setting with a spouse they loved enough to marry (not that this is the only kind of rape, of course — acquaintance and marital rape often don’t receive enough attention).
Wilcox and Wilson probably wouldn’t argue — not with a straight face at least — that women who want to protect themselves from rape should age, or that they should stop going out in public. But because they have a very specific idea about the importance of marriage and because they want to use the conversation about sexual assault to push that idea, they have no compunctions about making a similarly silly argument. (It’s not worth diving into the politics here, but Wilcox’s research clearly has a particular, “traditional” idea of family structure in mind.)
There is a grain of truth here when it comes to the kids part: All things being equal, the best setting for a kid is a stable family with multiple loving parents who can share the child-rearing and breadwinning duties (though Wilcox likely disagrees with the emerging social-science consensus that it doesn’t matter if those parents happen to be the same gender). But again, going back to the causation/correlation thing, parents in stable marriages are also less likely, for a wide variety of reasons, to be the sorts of people who would abuse their kids. If their marriages were accidentally annulled tomorrow, it wouldn’t suddenly make them horribly abusive parents. The factors that lead someone to be more likely to be married — education, mental health, and various other measures of stability and well-being — obviously say more about one’s propensity for violence than the presence or absence of a ring.