A trial starts tomorrow in federal court about whether Michigan's ban on same-sex marriage is constitutional, and as the New York Times explained over the weekend, it will offer an interesting test of the best research conservatives could come up with to support their contention that gay parents are bad for children. When we take a close look at what they'll put on the stand, it shows something that I think applies to a lot of areas of the conservative movement these days: when they try to play seriously on the field of ideas, what they come up with is, frankly, pathetic.
After years of watching researchers fail to find any ill effects of children being brought up by gay people, conservatives felt like they had to do something, and here's what they did:
In meetings hosted by the Heritage Foundation in Washington in late 2010, opponents of same-sex marriage discussed the urgent need to generate new studies on family structures and children, according to recent pretrial depositions of two witnesses in the Michigan trial and other participants. One result was the marshaling of $785,000 for a large-scale study by Mark Regnerus, a meeting participant and a sociologist at the University of Texas who will testify in Michigan.
So, they throw a ridiculous amount of money at an academic who's friendly to their position but knows what he's doing, and they figure, now we'll have the support we need to bolster our case. And what is he able to come up with? The result, which has been heralded by conservatives and anti-gay organizations, is this study. I want to talk about it briefly, because if this is the best they can do, it shows what a poor position they're in.
The problem in studying the question of how having gay parents affects children is that gay people are a relatively small portion of the population, so getting a sample that includes enough of them to allow for comprehensive analysis is difficult. But Regnerus's sample, of young adults who had a parent who at one time or another had a gay relationship (keep that in mind) is just as problematic, and the way he analyzes the data makes it even more problematic. He asserts that having had such a parent is associated with a range of bad outcomes, like being depressed, being unemployed, or having a criminal record. He doesn't specify the mechanism by which these things are supposed to occur—are gay people naturally bad parents? Does exposure to gayness make you crazy? And that's where things start to break down.
The main problem with Regnerus's study is that he's dealing with such small numbers—163 respondents whose mother had a gay relationship, and 73 whose father had one—that he can identify only a tiny number whose parent had a stable relationship of the kind we're talking about when we talk about marriage equality. Only 23 percent of his children of lesbians (or 38 individuals) lived in a home where that relationship lasted over three years. And among children of men who had had a gay relationship, the figure was "less than 2 percent," which would mean either 14 individuals or less. But he doesn't ever look at these people separately, even though they're precisely the children we're supposedly concerned about when we're debating whether gay people should be allowed to get married.
So when Regnerus runs his regressions (which are only described in the text but aren't displayed in a table, which is odd), he doesn't include the length of their parents' gay relationship as a control. That means that, say, the young man who grew up in a happy home with two mothers is lumped together with the young woman whose father was living a miserable life on the down low, got caught by his wife, and then the family went through an ugly divorce. Most of the "children of gay people" in Regnerus's sample are actually children of unhappy marriages, infidelity, separation, divorce, single parenthood, or some combination thereof. So it isn't too surprising that in young adulthood, these people had more problems on average than people who came from intact heterosexual two-parent households, which is the group he's comparing them to. In fact, he doesn't say if any of the children of gay people in his sample had a similar upbringing to that comparison group—raised by an intact couple for their entire childhood—so his study tells us precisely nothing about how children like that are likely to turn out when they grow up.
Because he's an academic, Regnerus talks about his data like an academic does, with plenty of caveats about what can and can't be inferred from his results. But the point is that this study, with these glaring weaknesses that mean it doesn't at all support the argument conservatives want to make, is the best they can do.
And that's really the problem. Because there's a fundamental contradiction in the conservatives' argument. They're trying to argue that gay people shouldn't be allowed to marry—in other words, that the status quo in most states is the preferred situation—by arguing that the status quo is bad for children of gay people. But if their policy argument had a shred of coherence, what it would actually demand is that they show that allowing kids to grow up in households with legally recognized, stable two-parent gay relationships would be worse than growing up with a single parent or in a household where one of your parents is closeted. That's the choice before us, and if conservatives are going to say they're motivated by the welfare of children, that's the position they'll have to take: that unstable relationships without legal protection provide a better environment for raising kids than legally recognized marriages. And I can't imagine even they believe that.