Thursday, June 21, 2012

Confirmation Bias and Same Sex Parenting

Are you familiar with the term? A confirmation bias is the tendency that people have to look for, or only attend to information that confirms what they already believe. Everyone does it, and it's almost impossible to escape. It's also exactly what just about everyone does when it comes to politics and other similarly touchy subjects, as Jeff Jacoby writes about here:

Assuming Marks is right about the weakness of the findings on which the APA's verdict was based, how many advocates of same-sex marriage or adoption by gay and lesbian parents will consider changing their view? How many would back away from their support for gay marriage in the light of anything social science might say? I'd estimate the number at, roughly, zero. Conversely, suppose Marks's paper had demonstrated that the APA's declaration was even more firmly supported than previously realized. How many principled opponents of gay marriage would change their minds? My estimate would stay at zero.

In the same issue of Social Science Research, University of Texas sociologist Mark Regnerus publishes the results of a large national study, based on interviews with a random sample of 15,000 young adults (aged 18 to 39) about their families, upbringing, and life experiences. Regnerus's bottom line: Children raised by their biological mother and father in stable families tended to turn out better than those whose parents had been in same-sex relationships. Even after controlling for age, race, gender, as well as subjective factors, such as being bullied as a youth, the findings were stark. Children raised by one or more gay parents, Regnerus wrote in an essay on Slate, "were more apt to report being unemployed, less healthy, more depressed." They were also more likely to have experienced infidelity, trouble with the law, and sexual victimization.

Regnerus's methodology has been sharply disparaged. Even some scholars who oppose same-sex marriage have underscored its weaknesses. Regnerus himself acknowledges that outcomes might be very different for kids being raised by same-sex parents today, "in an era that is more accepting and supportive of gay and lesbian couples." And he stresses that sexual orientation has "nothing to do with the ability to be a good, effective parent."

But even if his methodology were unassailable, would it change the larger debate over homosexuality and same-sex marriage? If you believe legalizing gay marriage is a matter of fundamental fairness, no scholarly study is likely to turn you around. And if you regard same-sex marriage as inherently immoral or absurd, a shelf of scientific journals touting its benefits won't convince you otherwise.

We like to think of ourselves as rational creatures with a healthy respect for facts and logic and science. And yet when it comes to the most controversial questions of public policy -- gun ownership, abortion, church-state separation, waterboarding, illegal immigration, you name it -- does anybody start with the data and only then decide where to stand? Most of us move in the other direction.

A lot of people have gone to great lengths to disparage the study, as Jacoby notes, but after reading his assertions in the article itself, he makes an honest effort to qualify his findings, suggesting that they are still very limited and may have different outcomes if they were to be redone today.


Mona Charen notes:


The studies on children raised by homosexual parents that predated Regnerus’s work suffered from a number of flaws. They tended to be examinations of “mostly white, well-educated, lesbian parents” living in metropolitan areas. They were often based on parental reports of childhood outcomes, and were composed of people who had been recruited at lesbian bookstores and other contact points — skewing the sample in favor of those eager to make a point. Not all of the studies were marred by such flaws, but nearly all were small, and thus lacked, in Regnerus’s words, “enough statistical power to detect meaningful differences should they exist.”

Regnerus’s study, the New Family Structures Study (NFSS), interviewed 15,000 adults aged 18–39, and asked dozens of questions about their lives, including whether their mother or father had ever been involved in a same-sex relationship. Among those whose parents had been involved in same-sex relationships, the outcomes were significantly worse than for children raised by married mothers and fathers. Even after controlling for factors such as age, race, gender, or the gay-friendliness of the state in which they lived, those raised in homes with one (or more) gay parents reported that they experienced more depression, ill health, unemployment, infidelity, drug use, trouble with the law, sexual partners, sexual victimization, and unhappy childhood memories.

 Anyway, these findings are interesting, but only insofar as a person can successfully interpret them in a subjective manner.

No comments: