Mark Regnerus conducted a study from data collected through the New Family Structures Study (NFSS), which is a large random sample (N = 2988) of American young adults ages 18-39. Regnerus examined the experiences of young adult children of parents who had been in a same-sex relationship (N = 175 raised by lesbian mothers; N = 73 raised by gay fathers) on some 40 outcome variables that dealt with emotional, social, and relational issues. He reported a number of statistically significant differences between these children when he compared them to intact, heterosexually-married biological parents.
The study is getting a lot of attention from both “sides” in the larger cultural discussions about same-sex marriage and parenting. Even the American Psychological Association (APA) has recently weighed in with an official press release in apparent response to the attention being given to the Regnerus study. (One wonders where the organization was when poorly-designed studies were being published that appeared to support what is now being touted as the “prevailing” view.)
The most frequently-cited criticism of the Regnerus study is that the researcher created various family forms (such as adult children who indicated that their mother had been in a same-sex relationship) to then compare them to intact, heterosexual, biological parents. These other family forms appear to reflect what others have referred to as mixed-orientation marriages or marriages in which one person is heterosexual and the other person is a sexual minority by virtue of attraction to the same sex (which, understandably, also has its own definitional limitations). So is this a reasonable way to represent the kinds of same-sex parents we have seen in other studies? Then: Is it a fair comparison to intact, heterosexual, biological parents?
As you can imagine, folks are lining up on both sides of that question. Those who say it is reasonable and fair are touting the differences on 25 of the 40 outcomes for children of women who had a same-sex relationship (e.g., lower levels of happiness; greater likelihood of being unfaithful in marriage or cohabitation) and 11 of 40 outcomes for children of men who had same-sex relationships, while those who say it is neither reasonable nor fair are attempting to dismiss the study as utterly irrelevant. (I haven’t seen a critic of the Regnerus study who demonstrated how common these ideal same-sex relationships and families are.) In any case, as so often happens, the truth lies somewhere in between.
It is not a meaningless study. Some of the differences noted may well be due to the dissolution of the relationship, but that does not appear to explain all of the findings. And, of course, it would be better to have intact, same-sex relationship to make comparisons, but that is the great social experiment, isn’t it? So to answer questions about various family forms (not to mention the nature and scope of social support for various family forms), it may take investigating related family forms to approximate an answer. Before you write to say that these other family forms are not related family forms at all, please note that it has been estimated that having been in a mixed orientation marriage or relationship is not unheard of. Indeed, 42% of gay and bisexual men in one study reported having been in a heterosexual marriage (yes, from a convenience sample), and an estimated 2 million sexual minorities have been in (or are in) a mixed orientation marriage.
So some folks with a vested interest can (and will) dismiss the study and rely instead on research that supports their own biases (research with much smaller and nearly hand-picked/optimal samples, unrepresentative samples, and so on). Others will make too much of the study. Neither response is particularly wise in my view. Rather, we do well to understand this study within the scope of its strengths and its limitations. We do well to conduct additional, well-designed studies that capitalize on the strengths of this study while addressing the study’s limitations–all of this is toward the end that such studies inform how we respond in the best interest of families and children, as well as the nature and scope of social support.