Flawed Study of "Gay Parenting": Roundup of Recent Media Cov
How and why a recent study on "gay parenting" should be viewed with caution.
Posted Jun 29, 2012
Reading the news lately? If so, you have probably seen coverage of a new study that claims to find negative outcomes for children of lesbian and gay parents. The paper that makes this claim is titled, “How different are the adult children of parents who have same-sex relationships? Findings from the New Family Structures study” and is authored by Mark Regnerus. This article, and several articles surrounding it, have received quite a bit of press and attention. I will not use this space to rehash the many criticisms that have been levied at the study; rather, I will briefly summarize the major critiques that have been made and to point the interested reader to some excellent articles that discuss these critiques in greater detail.
But first, a brief description of the study: Regnerus used Knowledge Networks to sample approximately 3000 participants 18-39 year olds (i.e., individuals who were born between about 1981-1994). Of these, 175 had mothers who at one point had had same-sex relationships, and 73 of had fathers who at one point had had same-sex relationships. Based on his analyses, Regnerus claimed that adults raised by parents who had same-sex relationships did worse socially, economically, and psychologically as compared to adults raised in intact, biological, heterosexual parent families. He further concluded that "the scholarly and popular consensus that there are no notable differences between the children who grew up with a mother or father in a same-sex relationship and those whose (heterosexual) mother and father were and are still married is a fiction.”
Critique #1: Definition of Same Sex Parents
Participants were asked, “From when you were born until age 18 (or until you left home to be on your own), did either of your parents ever have a romantic relationship with someone of the same sex?” Respondents who answered “yes” were classified as children of "Lesbian Mothers" or "Gay Fathers" (based on who they recalled as having a same-sex relationship) and compared to respondents from “Intact Biological Families,” as well as to respondents from adopted, divorced, single-parent, and step-family environments.
Thus, as sociologist Andrew Perrin, Box Turtle Bulletin blogger Jim Burroway, and others have pointed out, the article’s single biggest weakness is this definition of same-sex parents. Participants were categorized as having same-sex parents if their parent had ever had a same-sex romantic relationship while the respondent was under 18. This definition of “same-sex couples” is fundamentally flawed, in that (as closer examination of the paper reveals), almost none of the respondents were actually raised in a same-sex parent household. (In fact, only two children of “gay parents” in the survey were raised by the same same-sex couple their entire lives, and only 8 were raised by a same-sex couple who were together over 8 years.) Yet the paper repeatedly refers to children of “gay men” and “lesbians” when in fact, these children are individuals who, as adults, knew about and recalled that their parents had at some point had a romantic relationship with someone of the same sex.
Regnerus then compared this group of individuals, on various outcomes, to individuals who grew up in stable homes with married, biological, heterosexual parents. As William Saletan (writing for Slate) concludes, the findings that Regnerus produced should not surprise us, because this isn’t a study of gay couples who decided to have children. Regnerus used a “loaded classification system” that produced “predictable results.”
Critique #2: Categories are Not Mutually Exclusive
Related to the above critique is the fact that the categories that Regnerus used are not mutually exclusive, but yet were treated as such for purposes of analyzing the data. Namely, participants who fall into the “lesbian mother” and “gay father” categories may quite plausibly also be in any other category as well (e.g., they may also be from divorced, single parent, or intact biological families), and, indeed most of them almost certainly were. Thus, as Andrew Perrin points out, in his excellent critique of the study, “the analytical strategy amounts to manufacturing a virtual world that doesn’t fit the data at all, then analyzing that world. Why do it this way? Well, ‘gay and lesbian parents, as well as heterosexual adoptive parents, can be challenging to identify and locate.’ [Regnerus’s] article goes to great lengths to explain how hard they worked to find the necessary respondents. But trying really hard doesn’t help when the conceptual category is fatally flawed. The parents discussed aren’t, by and large, gay and lesbian parents!”
There are numerous other critiques and problems that have been raised in regards to this study.
First, as Perrin points out, one of the outcome measures, respondent’s sexual orientation, is dichotomized as “100% heterosexual (vs. anything else)”. This is problematic and may be designed to find large differences between groups, since children from nontraditional families may be more likely at least to question their sexuality (based on prior research).
Second, many scholars have drawn attention to the unusually fast timeline by which the journal, Social Science Research, operated in publishing Regnerus’s paper. As Philip Cohen, a sociologist and family scholar, notes, “According to the data listed at the top of the article, it was submitted, revised, and accepted within 6 weeks. That’s fast, but that doesn’t make it wrong, necessarily.” However, Cohen goes on to point out that “according to the study design document online at the University of Texas and the article history dates listed by the journal Social Science Research ... the paper was submitted for publication 20 days before the end of the data collection, and 23 days before the data were delivered to the University of Texas.” All of this raises serious questions about the unusual process by which Regnerus’s study was prepared, analyzed, and published.