On Friday, I blogged about the ASA brief in the same-sex marriage cases (see here). In short, I praised the ASA brief for defending and promoting the relevance of sociology to policy debates while providing the Court with the tools to evaluate what constitutes good sociology. The ASA brief explained the scientific consensus that children of same-sex parents fare no worse than children of opposite-sex parents and the scientific methods used to arrive at that consensus. Much of the ASA brief is devoted to explaining the methodological flaws in a study by sociologist Mark Regnerus that purports to find that children of same-sex couples do not do as well as children of opposite-sex couples. The ASA brief explains how Regnerus arrived at his conclusions and why his data cannot support the claims made of them, allowing the Court to judge for itself the validity of the study.
Today I’m following up with a post on a brief submitted by Regnerus and a group of other social scientists opposed to same-sex marriage (for simplicity, I’m going to refer to this as the Regnerus brief). I’m mostly going to summarize the Regnerus brief, highlighting how it responds to the ASA brief. I’m also going to summarize the brief submitted by the American Psychological Association, which is also involved in this contest. I’ll conclude with some brief thoughts on what this all means.
The Regnerus Brief
Titled the “Brief of Social Science Professors,” the Regnerus brief was filed on January 29. It is signed by seven professors, including Regnerus.
The Regnerus brief begins by arguing that “compelling evidence shows that children benefit from the unique parenting contributions of both men and women.” This section summarizes literature on the different parenting roles that men and women play in opposite-sex couples. Much of this literature promotes the involvement of fathers in child-rearing. But as the ASA brief explains, these studies have no bearing on same-sex parenting. In fact, as noted in the ASA brief, the authors of some of these studies have objected to how their work is cited and used in opposition to same-sex marriage.
The Regnerus brief then argues that “the claim of ‘no difference’ in outcomes of children raised by gay and lesbian parents and intact biological parents is empirically undermined by significant methodological limitations.” There’s a tension in this section between reaffirming the argument that married opposite-sex couples are best for children on the one hand, and arguing that there is just not enough evidence for the success of same-sex parenting so we should be cautious on the other hand. The undercurrent of this whole section is to suggest that research on same-sex parenting is “young” as a field and politically oriented towards supporting lesbians and gays, and therefore inherently suspect.
In particular, this section of the brief explains the importance of random sampling to social science studies. The Regnerus brief specifically claims that all studies that will be included in the APA brief include small, non-random samples of same-sex parents. While this is a legitimate concern, the Regnerus brief does not discuss any of the numerous ways that research on same-sex parenting has recognized and confronted this challenge. Indeed, as I will discuss below, the APA brief explicitly describes how this challenge has been met. The ASA brief does not explicitly respond to this challenge of random samples, but the ASA brief does explain the methodological rigor and sociological standards behind studies finding no difference between children of same-sex couples and children of opposite-sex couples.
The Regnerus brief then introduces Regnerus’s study, discussing it at length. It describes it as the best national representative sample, allegedly showing clear benefits for children of opposite-sex parents. The Regnerus brief does acknowledge the controversy, explaining “One of the most frequent criticisms by supporters of same-sex marriage was that the study compared ‘apples to oranges,’ by comparing the numerous adult children of stably intact biological parents with both adult children whose mother or father left a heterosexual union for a same-sex one, and the rare scenarios in which children were raised consistently and stably in a same-sex household.” Here, we see two key themes. First, the Regnerus brief characterizes attacks on Regnerus’s research as politically motivated (indeed, later, the brief says that the volume of controversy around the study was unusual, again trying to suggest it was all just political). Second, the Regnerus brief suggests that stable same-sex relationships are simply too rare to study (a point the APA brief explicitly responded to).
The contrast between the ASA brief and the Regnerus brief on this is striking. The Regnerus brief acknowledges the controversy, but does not actually address the substance of the criticisms. It characterizes this as a legitimate academic disagreement, blown up by political backlash. The ASA brief painstakingly describes the methods used in Regnerus’s study and why he claims that they support his results. It then explains the methodological problems with the study and how Regnerus’s responses fail to deal with the criticisms of the study.
The American Psychological Association, joined by the American Medical Association and various other medical groups, filed an amicus brief that directly responds to the Regnerus brief.
The APA brief begins with a brief section on “The Scientific Evidence Presented in this Brief.” The brief explains that “In drawing conclusions, Amici rely on the best empirical research available, focusing on general patterns rather than any single study. Before citing a study herein, Amici have critically evaluated its methodology, including the reliability and validity of the measures and tests it employed, and the quality of its data-collection procedures and statistical analyses.” Like the ASA brief, the APA brief stands for disciplinary standards and methodological rigor. I like that the APA brief is even more explicit and thorough in this regard.
The APA brief then presents the consensus that homosexuality is normal and that lesbians and gays form stable relationships at equal rates as heterosexuals. Read in response to the Regnerus brief, this asserts that not only did Regnerus fail to study stable same-sex relationships, but these relationships are in fact just as common as stable opposite-sex relationships.
Like the ASA brief, the APA brief thoroughly documents a scientific consensus that children of same-sex couples are similar to children of opposite-sex couples (in fact, it is probably a little more thorough on this than the ASA brief, given that much of this research is housed in psychology). Throughout this discussion, the APA brief stresses the cumulative nature of the evidence, also mentioning that the body of research includes three recent studies that use national probability studies. Read in response to the Regnerus brief, it is clear that this review of the literature is geared to respond to the contention that studies of same-sex parenting are inconclusive, young, or non-representative.
The APA brief then explicitly turns to the contention that these studies are invalid because they do not use random samples. I quote at some length:
“Scientific research is a cumulative process. Empirical studies inevitably have limitations. Simply because a particular study’s methodology has imperfections or its results warrant qualifications does not mean that the entire study should be dismissed. Rather, it should be evaluated within the context of the cumulative relevant research, recognizing that some studies’ strengths can offset other studies’ corresponding limitations.”
“Amici who challenge all empirical findings in this area because some studies used small nonprobability samples ignore the fact that many findings from those studies have been replicated in national probability samples. They also fail to acknowledge that studies with nonprobability samples can answer important scientific questions, especially when they include appropriate comparison groups.”
The APA brief explains that research on same-sex parents has addressed sampling challenges by using matched samples and representative probability samples. The brief stresses that, “regardless of the sampling method used, greater confidence can be placed in findings that have been replicated by others using different samples.”
The APA brief also discusses the criticism of Regnerus’s study, but only for a couple paragraphs. I’m guessing the interplay between the ASA brief and the APA brief was at least partially planned.
Given my discussion of how the ASA brief did not privilege marriage, I do also want to note that the final section of the APA brief argues that “denying the status of marriage to same-sex couples stigmatizes them.” As I noted in my blog post on the ASA brief, this move is common amongst advocates of same-sex marriage: children need stability, marriage promotes stability, therefore same-sex couples should marry. While I don’t necessarily fault the APA for this move, given that this is a brief in a same-sex marriage case, the comparison does reinforce my earlier conclusion about how the ASA brief carefully avoided privileging marriage.
What Does This All Mean?
In order to fully understand these briefs, we need to read them against each other and understand how they are using law as a site to contest the meaning of sexuality and family. The Regnerus brief was filed first because of different court filing deadlines for both sides. It clearly predicted the APA brief but did not seem to predict the ASA brief.
The Regnerus brief presents a facade of engaging with the APA brief. It pretends to counter research on positive outcomes of children of same-sex parents, but it doesn’t really engage with the evidence in any substantive manner. For example, the Regnerus brief describes the challenges faced in locating same-sex parents in a national random sample. The brief claims this problem is fatal to all research on same-sex parenting. But it does not discuss ways that this challenge is addressed (matched sampling, probability sampling, comparing successive results). Instead, the strategy seems to be to give conservative justices something to latch on to. In an extensive review of the scientific evidence, the trial court in the Prop 8 case concluded that the scientific evidence conclusively supports positive outcomes for children of same-sex parents. Courts are clearly equipped to do this sort of judging. Thus, the Regnerus strategy seems to be to claim that there are unanswered questions, instead of engaging in actual intellectual debate. Indeed, this seems to be the point of accusations of political bias in the research on same-sex parents.
The strength of the ASA brief is even clearer when it is read against the Regnerus brief. The ASA brief spent a lot of time defining the standards for strong sociological research and how those standards were used in developing the consensus on children of same-sex parents. Instead of just responding tit-for-tat to the Regnerus brief, the ASA brief provided the Court with the tools to evaluate all the claims for itself.
This back and forth underscores my concluding point from my previous post. The ASA brief was not submitted in a vacuum. In the absence of the ASA brief, the Regnerus brief would claim the mantle of the voice of sociology. In this context, the ASA brief is a model of the role the ASA should strive to play in policy settings.