The claim that there are no differences in outcomes for children living in same-sex households arises from how scholars collect, analyze, and present data to support a politically expedient conclusion, not from what the data tend to reveal at face value. Nearly five years ago, I began to question the scholarly consensus that there were “no differences” between same-sex and opposite-sex households with children.
It was nearly five years ago that I first received data back from the research firm that had carried out the New Family Structures Study (NFSS) protocol. Shortly thereafter, I began to question the scholarly consensus that there were “no differences” between same-sex and opposite-sex households with children. My skepticism didn’t sit well with the guild.
Social network analysts subsequently assessed patterns of “citation networks” in the same-sex parenting literature and concluded that there is indeed a consensus out there that claims there are “no differences.” I see it. I just think the foundation for the consensus is more slipshod than rock-solid. It is the result of early but methodologically limited evaluations that formed a politically expedient narrative. It is not the product of many rigorous, sustained examinations of high-quality data over time, across countries, and using different measurement strategies and analytic approaches.
Some liken the debate over the science of same-sex parenting to that around climate change. They argue that the science is so extensive, the published studies so numerous, and the conclusions so overwhelmingly unidirectional that only scientists acting in bad faith would object. But that’s not where this field of study is. How can we possibly come to a legitimate consensus about the short and long-term effects of a practice (childrearing) of a tiny minority about whom generalizable data of sufficient size for valid comparative analysis was not available until the past decade? The answer, of course, is that you cannot—or rather should not—yet.
What we have, rather, is a politicized consensus generated by lots of small studies of tiny, non-representative samples misinterpreted as applying to the entire population of same-sex parents. It is—by comparison to climate change—like saying that since the surface temperatures in Taiwan, Togo, and Texas have inched up a degree or two that the entire globe must have as well. Scientists know better than to declare such a thing based on limited evidence.
In fact, I am unaware of any other domain of science in which scholars have so little high-quality data to answer comparatively new research questions and yet are so quick to declare those questions answered and done with. We don’t do that in any other field or with any other question. What ought to be an empirical matter—an important one, no doubt—has instead turned into a moral test of fealty.
Where’s the Data?
A popular misreading of that analysis of citation networks is that there are over 19,000 quality studies of same-sex households with children, and that not one of them has detected problems. The authors themselves note that is not the case. Their study began with what amounts to over 19,000 scholarly “hits” on an online Boolean word search. That is, there were 19,000 publications on the ISI Web of Science that used the word “gay” or “lesbian” or “homosexual” or “same-sex” and “parent” somewhere in the title or abstract. Far fewer were actually assessed in their analysis (as they pointed out). Indeed, anyone who pays careful attention to this domain knows that there aren’t nearly as many studies as is commonly believed—indeed, well under 100 that are worth discussing—and fewer than ten datasets that can appropriately yield generalizable information about the population of same-sex households with children.
If I were up against 19,000 quality studies that disagreed with me, I would have long ago gladly capitulated. But that’s not the scientific reality. Instead, two datasets—each of which has significant problems here—are responsible for the majority of studies upon which this “consensus” is based.
First, twenty-six peer-reviewed studies have emerged from just one data source: the National Longitudinal Lesbian Family Study. It repeatedly surveyed seventy-eight children of mostly white, educated, well-off lesbians who were recruited to participate. It may not be a stretch to suggest that these are the seventy-eight most influential children in modern American social and legal history. For nearly twenty-five years now, the scholarly community (and the media) has been treated to studies of this unique cluster of kids, and the general public has been left with the impression that they represent the whole population of children of American lesbian mothers. Moreover, it stretches the imagination to suggest that these seventy-eight respondents, now followed for nearly twenty-eight years, are still valid, reliable sources of information. When the adolescent children of lesbian parents are being intermittently interviewed for a study whose results have proven quite politically expedient—and almost always covered favorably by the mainstream media—it is prudent for scholars to be skeptical.