I never implied all traditional marriages were perfect, just that a traditional marriage is a healthier environment to bring up children than a comparable queer marriage.
III. WEIGHING THE EVIDENCE
How should legal thinkers and decision-makers evaluate such competing claims about family structure and child well-being both allegedly grounded in social science evidence?
Numerous reviews of the literature on sexual orientation and parenting have been conducted.12 At least three such reviews have pointed to the serious scientific limitations of the social science literature on gay parenting.13
Perhaps the most thorough review was prepared by Steven Nock, a sociologist at the University of Virginia who was asked to review several hundred studies as an expert witness for the Attorney General of Canada. Nock concluded:
Through this analysis I draw my conclusions that 1) all of the articles I reviewed contained at least one fatal flaw of design or execution; and 2) not a single one of those studies was conducted according to general accepted standards of scientific research.14
Design flaws researchers have found in these studies include very basic limitations:
No nationally representative sample. Even scholars enthusiastic about unisex parenting, such as Stacey and Biblarz, acknowledge that "there are no studies of child development based on random, representative samples of [same-sex couple] families."15
Limited outcome measures. Many of the outcomes measured by the research are unrelated to standard measures of child well-being used by family sociologists (perhaps because most of the researchers are developmental psychologists, not sociologists).
Reliance on maternal reports. Many studies rely on a mother's report of her parenting skills and abilities, rather than objective measures of child outcomes.
No long-term studies. All of the studies conducted to date focus on static or short-term measures of child development. Few or none follow children of unisex parents to adulthood.
But perhaps the most serious methodological critique of these studies, at least with reference to the family structure debate, is this:
The vast majority of these studies compare single lesbian mothers to single heterosexual mothers. As sociologist Charlotte Patterson, a leading researcher on gay and lesbian parenting, recently summed up, "[M]ost studies have compared children in divorced lesbian mother-headed families with children in divorced heterosexual mother-headed families."16
Most of the gay parenting literature thus compares children in some fatherless families to children in other fatherless family forms. The results may be relevant for some legal policy debates (such as custody disputes) but, in our opinion, they are not designed to shed light on family structure per se, and cannot credibly be used to contradict the current weight of social science: family structure matters, and the family structure that is most protective a child well-being is the intact, married biological family.
Children do best when raised by their own married mother and father.