A set of plaintiffs in Michigan who are challenging the state’s ban on same-sex marriage and same-sex adoption have filed a brief asking the court to disallow the testimony of Mark Regnerus as an expert in the case, citing the many methodological problems in his famous study of gay parenting (which wasn’t really about gay parenting at all). This is what is known as a Daubert challenge to the relevance and reliability of an ostensible expert in a case.
Daubert suggested four non-exclusive criteria “against which to measure the validity of the underlying principles and methods which undergird an expert’s opinion:  whether the technique or theory is capable of being tested;  whether it has been published and reviewed by peers in the relevant technical community;  the potential or known rate of error yielded by the methodology; and  whether the principle or theory has been generally accepted or shunned by the community of experts in the field.” The Sixth Circuit has identified several “red flags” that “caution against certifying an expert,” including (1) reliance on anecdotal evidence, (2) improper extrapolation, (3) failure to consider other possible causes, (4) lack of testing, and (5) subjectivity…
Specifically, Regnerus’s expected testimony is irrelevant because (A) it relies on misleading comparisons between children continuously raised by married parents and children who struggled with family instability, and (B) any purported association between parental same-sex romantic behavior and family instability is immaterial to
Defendants’ justification for the MMA. As a result of these flaws, Regnerus’s testimony lacks a “valid scientific connection” to the “pertinent inquiry.”…
Regnerus’s expected testimony is based on the New Family Structures Study (“NFSS”). However, the results of the NFSS are irrelevant and misleading. The study divides survey respondents into categories that are designed to show dissimilar results between children of heterosexual parents and children of parents who engaged in same-sex romantic behavior. The NFSS divides respondents with heterosexual parents into six different groups, distilling those respondents who “lived in [an] intact biological family . . . from [years] 0 to 18, and [whose] parents are still married” into a specific group labeled “IBF.” Regnerus refers to stably-coupled heterosexual households, like those in the IBF group, as the “gold standard.” The majority of NFSS respondents with heterosexual parents do not fall into this IBF group and are instead classified into one of the remaining five heterosexual-parent groups, such as “stepfamily” or “single parent” if their parents “were either never married or else divorced.” The NSFF takes a different approach to respondents who reported parental same-sex romantic behavior. It groups them into only two categories: children of “lesbian mothers” (“LM”) and “gay fathers” (“GF”). All six of the heterosexual parent categories are collapsed into these two groups, allowing the study to obscure the effects of divorce or similar factors behind a gay or lesbian label.Despite the fact that the IBF group is designed to contain only the “gold standard” of heterosexual family arrangements, Regnerus repeatedly uses it as the baseline to which he compares the LM and GF groups. By engaging in this “apples-to-oranges” comparison, Regnerus conflates the effects of divorce, adoption, and single parenthood with the status of same-sex couples. As a result, he fails to address the relevant issue at hand — whether the MMA is justified because “only heterosexual marriages can provide children with the appropriate gender role-modeling required for healthy psychological development.” If the NFSS made an “apples-to-apples” comparison between children
of stable heterosexual couples and children of stable gay and lesbian couples, exactly two respondents would fall into the gay and lesbian analogue of the IBF group. The remaining respondents in the gay and lesbian parent groups struggled with factors like divorce that would disqualify the child of heterosexual parents from membership in the IBF group. Regnerus himself concedes that the two respondents raised by lesbian parents since birth are well-adjusted and compare favorably to the other respondents. Thus, in the only two instances where Regnerus could offer comparisons potentially relevant to the question of whether gay and lesbian couples parent as well as heterosexual couples, his study supports Plaintiffs’ claim that gay and lesbian parents can raise happy and healthy children.
I’ll be very curious to see if the judge disallows Regnerus’ testimony. The case here is pretty much unassailable, but judges are still reluctant to prevent a witness from testifying if one side calls them. But the motion asks the judge either to disallow the testimony or to give it no weight for all the reasons stated in the brief, so even if he allows Regnerus to testify, that doesn’t mean he’ll consider that testimony relevant or reliable. You can read the whole brief here.