Much of the jousting over same-sex marriage, same-sex parenting and transgender issues takes place in a fact-free arena. Fortunately, there has been a pushback from academics dismayed by the lack of academic rigour. Recently we reviewed an essay published in The New Atlantis by two experts from Johns Hopkins University and yesterday a critique of Australia’s Safe Schools program by University of Sydney law professor Patrick Parkinson. Today, we report on a just-published paper by Walter Schumm, of Kansas State University about same-sex parenting.
Is having same-sex parents a disadvantage for a child? The conventional wisdom says that there is no difference from having heterosexual parents. Politicians, journalists, academics, governments, and courts seem to have reached a consensus on the issue. Questioning this orthodoxy results in accusations of irrational bias and homophobia.
This belief has taken deep root in academia. One recent study concluded that “there is a clear consensus in the social science literature” that children with same-sex parents fare just as well – or even better. And Susan Golombok, of the University of Cambridge, one of the leading researchers in family studies, said bluntly last year that “much of this debate has been founded upon myths and false assumptions about the deleterious consequences of new family forms for the children who grow up in them, rather than the findings of empirical research”.
However, Walter R. Schumm, of Kansas State University, insists that the “no differences” issue is far from settled. As he documents in his 120-page article, “A Review and Critique of Research on Same-Sex Parenting and Adoption” in the on-line first section of the journal Psychological Reports, most studies are flawed, inconclusive, or small-scale. And many troubling issues surface which never reach the media or the courts.
Schumm’s point is not that the “no difference” hypothesis is necessarily wrong, but that the current consensus leaks like a sieve. It’s not even clear what “no difference” means to a social scientist. If a parameter, such as a child’s capacity for delayed gratification, is not measured, there will be no difference – but is that scientific? If the average is the same but the extremes are very different, is there really no difference?
Here are a few highlights from this fascinating paper.
Are same-sex relationships stable? This is a key issue for social science researchers. Everyone agrees that children thrive in a stable home and suffer if their parents separate and re-partner. “Research to date indicates that gay and lesbian parents have less stable relationships than heterosexual parents,” Schumm concludes.
But he acknowledges that investigating this instability is difficult. “One major hurdle to overcome is the difficulty in obtaining random samples of large numbers of highly committed, very stable same-sex parents whose children’s outcomes can be compared to outcomes for children of heterosexual couples who would be equivalent in terms of stability, education, socioeconomic status, and numbers of children. Finding any same-sex parent couples can be a challenge.”
Some possible explanations emerge from trawling through the research. First, while having a child tends to stabilize a heterosexual couple, it may destabilize a lesbian couple, possibly because many of them have children from previous relationships. Another possibility is jealousy between the biological mother and the social co-parent, a common situation in lesbian families.
And then there is the well-known issue of monogamy. Most researchers accept that many, if not most, gay and lesbian couples are non-monogamous. In fact, some couples have more affairs after they are married than before. One expert cited by Schumm even opined that non-monogamy was beneficial for the children in the relationship.
There is also a streak of amorality in same-sex relationships. One paper comparing 170 gay and lesbian couples with 144 heterosexual couples found that none of the former believed that moral considerations should be a deterrent to breaking up. “Zero out of 170 is remarkable in social science research,” comments Schumm.
Outcomes of same-sex parenting. It is often argued that children of LGBT parents suffer no harm whatsoever. A typical line from a well-known researcher in 2006 was that “in the past 20 years, studies have consistently found that children in lesbian families are as well adjusted as children in heterosexual families,” “despite the stigmatization that comes with growing up in a homophobic culture”.
If this is true, it means that children of same-sex couples will not have higher rates of drug abuse, casual sexual behavior, or educational progress compared to the children of heterosexual parents. “It is possible that children of LGBT parents may be advantaged,” according to one study, “especially if the children are LGBT themselves”
These optimistic sentiments are echoed by lawyers and judges. It is common to read claims that there is no defensible factual foundation to reports of harm; that no legitimate research exists which proves harm; and that “the issue is so far beyond dispute that it would be irrational to hold otherwise”.
It is unwise to speak in such a categorical manner, Schumm points out; researchers have barely scratched the surface of the complexities of gay parenting.
A closer examination of the evidence, even by LGBT academics, shows that there is “some empirical evidence that children of gay, lesbian, or bisexual parents may experience life problems suggestive of difficulties in psychological adjustment (lower self-esteem, difficulties with secure attachments, substance abuse, precocious sexuality, sexually transmitted infections, criminal activity, difficulties in educational progress, need for psychotherapy, eating disorders, depression, etc.), as well as differences in gender role attitudes or behaviors”.
Is it an open-and-shut case? No, not for either side. It is only honest to acknowledge that the available evidence points both ways.
How about children adopted by same-sex parents? In this area the deficiencies of existing research are glaring. Most of the studies are very small and nearly all of them were conducted amongst well-educated, wealthy couples. Many gay couples can afford good quality child care, another factor clouding the statistics. As a result we know little about low-income gay and lesbian families. Their stories might be quite different.
“The results of most available studies can tell us little about sexual orientation and adoption for low to moderate socioeconomic status families or for families who have adopted several children,” Schumm writes. “In addition, few of the studies involved adoptions of older (e.g., teenage) children, limiting our insight into how sexual orientation might interact with age.”
Researchers have encountered unexpected difficulties. For example, what is a same-sex couple? This seems obvious, but as soon as researchers begin to examine data, they hit speed bumps.
When Schumm drilled down into one study of lesbian mothers, he found that some of them had become transmen and others had reverted to heterosexuality. Some surveys counted a mother and a daughter with the younger woman’s children as a lesbian household. It was not a straightforward question.
Children in gay and lesbian households often come from previous heterosexual relationships. In a study from 2001 of 101 gay fathers, 87 percent had been or were still in sexual relationships with women; more than 80 percent of the children were conceived in heterosexual relationships. This creates enormous methodological problems. If a child was raised for four years in a mom-and-dad household, but when she enters the statistics in her fifth year, she is living in a dad-and-dad household, is she the child of a same-sex couple or a shattered heterosexual relationship?
Some of the difficulties of being a gay or lesbian parent come from discrimination. Surprisingly, some of the discrimination comes from other gays and lesbians. A 2002 study found that “lesbian mothers … faced bias from heterosexuals as well as other lesbians and gay men. Lesbian periodical literature is replete with accounts of lesbian animosity toward lesbian mothers”.
Nice try, but try harder
Schumm is restrained in his criticism, but he points out that social scientists need to make the limitations of their studies more explicit. Sometimes they should state that their work has little value for public policy. And so he concludes with a warning about the notion of a “consensus”:
All of these concerns with the limitations of research concerning LGBT issues should raise red flags about any attempt to achieve scientific consensus prematurely, even if for a good or noble cause. If anyone is motivated to avoid a rush to judgment or a rush to consensus, it should be scientists including social scientists.
Michael Cook is editor of MercatorNet.