A pair of federal court decisions between Mother's Day and Father's Day means the Supreme Court is likely to weigh in on whether society rightly holds up both a mother and a father as the ideal for children.

Judges of the federal 9th Circuit appeals court overruled millions of Californians who voted for Proposition 8 to define marriage as one man and one woman in their state constitution. In the other case, judges of the 1st Circuit sided against an overwhelming bipartisan majority in Congress that passed the Defense of Marriage Act, which President Bill Clinton signed.

One argument propelling such judicial activism is that it makes no difference whether a child is raised by same-sex parents instead of a traditional, married mother and father.

Earlier in the Prop 8 case, for example, District Court Judge Vaughn Walker wrote: "The research supporting this conclusion is accepted beyond serious debate in the field of developmental psychology." And in 2005, a report by the American Psychological Association said "not a single study" found any significant disadvantage for children of same-sex parents.

Now, though, that "no difference" assertion is under serious challenge with the publication of two peer-reviewed studies in the July issue of the scholarly journal Social Science Research. One article, by Loren Marks of Louisiana State University, reviewed 59 studies that the APA relied on to make its claim seven years ago. He found that not one compared a large, random, representative sample of same-sex parents and their children with a large, random, representative sample of intact biological families.

Instead, the data were based mostly on small "convenience samples" drawn from easily accessible respondents. Such samples have limited use, and are completely inadequate when the goal is to generalize to the population at large - or to inform a major public policy debate such as the redefinition of marriage.

The second study published in Social Science Research overcomes these problems by using a nationally representative random sample. Called the New Family Structures Study (NFSS), this dataset is a project under the direction of Mark Regnerus at the University of Texas-Austin. NFSS surveyed nearly 3,000 young adults, ages 18-39, across eight different family structures. It looked at a total of 40 outcome areas; prior studies of children of same-sex parents evaluated only a few outcomes.

Regenerus found significant differences on 24 outcomes between children whose mother had a lesbian relationship and their peers from intact families. Among the most striking: significantly higher rates of sexual victimization, welfare dependence and unemployment.

NFSS also shows that stable same-sex households with children are extremely rare. Data collectors screened 15,000 young adults to get a representative sample of 3,000, of which 1.7 percent said a parent had one or more same-sex relationships. Only two respondents said they had lived from birth to age 18 with same-sex parents. By comparison, 58 percent of the 15,000 reported spending their entire childhood with their biological mother and father.

Even critics acknowledge that the New Family Structures Study raises the bar for quality of research on children affected by parents' same-sex relationships. Given the magnitude of the policy questions at hand, we need much more research in this area; Regnerus will open the NFSS data in the fall for analysis by other researchers.

Research on same-sex parenting seems to follow the same trajectory we saw in studies of the intact family in recent decades. At first, samples were small and methodology was inadequate. But that didn't stop the spread of popular cultural messages such as TV character Murphy Brown's single-motherhood-by-choice.

The quality of research gradually got more serious, and the results got more sobering. We had the "Dan Quayle Was Right" moment when Atlantic Monthly ran its April 1993 cover story showing Vice President Quayle's initial criticism of the sitcom's rosy picture to have been on target a year earlier. A reality show about a typical single mom would have revealed a much bleaker picture.

The two new studies suggest that we have a lot more to learn about how changing family forms impact children. They indicate Americans have good reason to stand by the tried-and-true ideal that every child should have a father and a mother. Courts have no business overruling that conclusion.

Jennifer A. Marshall is director of the DeVos Center for Religion and Civil Society at The Heritage Foundation.