In the Iowa court case that legalized same-sex marriage in that state, some "experts" offered testimony that same-sex marriage is bad for children. A group of social science organizations and scholars, myself included, signed an amicus brief arguing that evidence should be excluded because it lacked scientific merit. In the decision, the Iowa Supreme Court cited counter-evidence showing the same-sex couples can be good parents.
Research on the risks or benefits to children of being raised by same-sex couples has been hotly debated. The bulk of existing evidence supports the conclusion that children raised in stable married-couple families have better outcomes on average than those raised in other family contexts. That much is not really controversial. And it is interesting and important to study these questions. But there are at least three problems with using that research to support the argument against same-sex marriage.
1. There are no U.S.-based studies comparing children raised by opposite-sex and same-sex married couples, because there are no legally-married same-sex couples to use in the comparison. The few years of legal marriage we have in a few states hasn't yet yielded enough information to study this systematically.
The Iowa decision declared that "the ultimate disadvantage [from having the right to marriage denied] expressed in the testimony of the plaintiffs is the inability to obtain for themselves and for their children the personal and public affirmation that accompanies marriage." If the benefits of marriage even partly result from that "affirmation," then no current study can test whether differences in child outcomes between same-sex and opposite-sex married couples are caused by family structure. Maybe the benefits, if they are real, result from the families' membership in a commonly valued and valuable social institution.
2. The same people who want to ban gay marriage, and who sex up the research to support their ideological arguments, also insist that the government not collect data on same-sex married couples.
The U.S. Census Bureau, principally, is forbidden from counting same-sex married couples as married -- even when they are legally married in their home states -- by the Defense of Marriage Act. In 2010, as in previous decades, the decennial Census will once again unmarry those same-sex couples who report themselves as married, and count them instead as "unmarried partners." The same rules apply to the other surveys collected by the Federal government. This denial of state definitions of marriage undermines scientific efforts to study family structure and its consequences. (Of course, no one has to prove they are married for the Census -- the concern over whether self-reported marriages are really legal is suspiciously sudden.)
3. Finally, when a condition is advantageous -- that is, it yields benefits compared to being in some other condition -- that also means it contributes to inequality. That's because not everyone gets to experience it. If children of married couples are more likely to finish high school than those who grow up with single mothers, for example, then there is inequality between those two groups. One policy approach to that inequality is to make the condition more common - for example, encourage or coerce people to marry (or discourage people from having children when single).
Another approach is to improve outcomes for people in the disadvantaged group. For example, because resource scarcity is a big part of the problem for single-parent families, we could support a public school system that educates all children effectively, or provide income support to poor families.
Which is the better approach? When the difference is a matter of civil and human rights -- as it is in the case of legal marriage for all couples -- then I favor the second approach. Rather than prevent same-sex couples from having or rearing children, let's find a way to support same-sex parents, and others, to help equalize the chances of success for all children.