Scholars Respond to Regnerus Study

06.20.2012, 5:04 PM

A group of eighteen social scientists, including FamilyScholars blogger Brad Wilcox, released a statement today defending Mark Regnerus’s peer-reviewed published research on the New Family Structures Study. You can find the full statement here.

A Social Scientific Response to the Regnerus Controversy

Same-sex marriage is one of the most contentious and vexing issues now facing our nation. It is perhaps in part for that reason that the new study on same-sex parenting by University of Texas sociology professor Mark Regnerus, which finds that young-adult children of parents who have had same-sex relationships are more likely to suffer from a range of emotional and social problems,[1] has been subject to such sustained and sensational criticism from dozens of media outlets, from the Huffington Post to the New Yorker to the New Republic. These outlets have alleged, respectively, that his research is “anti-gay,” “breathtakingly sloppy,” and “gets everything wrong.”

Although Regnerus’s article in Social Science Research is not without its limitations, as social scientists, we think much of the public criticism Regnerus has received is unwarranted for three reasons.

First, there are limitations with prior research on this subject that have seldom been discussed by the media. The vast majority of studies published before 2012 on this subject have relied upon small, nonrepresentative samples that do not represent children in typical gay and lesbian families in the United States. [2] By contrast, Regnerus relies on a large, random, and representative sample of more than 200 children raised by parents who have had same-sex relationships, comparing them to a random sample of more than 2,000 children raised in heterosexual families, to reach his conclusions. This is why sociology professor Paul Amato, chair of the Family section of the American Sociological Association and president of the National Council on Family Relations, wrote that the Regnerus study was “better situated than virtually all previous studies to detect differences between these [different family] groups in the population.”[3] We are disappointed that many media outlets have not done their due diligence in investigating the scientific validity of prior studies, and acknowledging the superiority of Regnerus’s sample to most previous research.

Second, Regnerus has been chided for comparing young adults from gay and lesbian families that experienced high levels of family instability to young adults from stable heterosexual married families. This is not an ideal comparison. (Indeed, Regnerus himself acknowledges this point in his article, and calls for additional research on a representative sample of planned gay and lesbian families; such families may be more stable but are very difficult to locate in the population at large.[4]) But what his critics fail to appreciate is that Regnerus chose his categories on the basis of young adults’ characterizations of their own families growing up, and the young adults whose parents had same-sex romantic relationships also happened to have high levels of instability in their families of origin. This instability may well be an artifact of the social stigma and marginalization that often faced gay and lesbian couples during the time (extending back to the 1970s, in some cases) that many of these young adults came of age. It is also worth noting that Regnerus’s findings related to instability are consistent with recent studies of gay and lesbian couples in countries such as the Netherlands and Sweden, which find similarly high patterns of instability among same-sex couples.[5] Even Judith Stacey, a prominent critic of Regnerus’s study, elsewhere acknowledges that studies suggest that lesbian “relationships may prove less durable” than heterosexual marriages.[6] Thus, Regnerus should not be faulted for drawing a random, representative sample of young-adult children of parents who have had same-sex romantic relationships and also happened to have experienced high levels of family instability growing up.

Third, another study[7] published this month in the Journal of Marriage and Family comes to conclusions that parallel those of Regnerus’s study. This study finds that “children in same-sex parent families scored lower than their peers in married, 2-biological parent households” on two academic outcomes, and that these differences can be attributed to higher levels of family instability in same-sex families, compared to intact, biological married families. This study was also based on a large, nationally representative, and random survey of school-age children; moreover, the same-sex parents in this study lived together. The parallels between the findings in this study and Regnerus’s study strongly call into question the New Republic’s claim that the Regnerus study “gets everything wrong.”

To be clear: We do not think that these new studies settle the nation’s ongoing debate about gay parenting, same-sex marriage, and the welfare of children. In fact, research on same-sex parenting based on nationally representative samples is still in its infancy. But we think that the Regnerus study, which is one of the first to rely on a large, random, and representative sample of children from parents who have experienced same-sex relationships, has helped to inform the ongoing scholarly and public conversation about same-sex families in America. Indeed, it is possible to interpret Regnerus’s findings as evidence for the need for legalized gay marriage, in order to support the social stability of such relationships. As social scientists, our hope is that more such studies will be forthcoming shortly, and that future journalistic coverage of such studies, and this contentious topic, will be more civil, thorough, and thoughtful than has been the coverage of the new study by Professor Mark Regnerus.

Byron Johnson, Baylor University
Douglas Allen, Simon Fraser University
Peter Arcidiacono, Duke University
John Bartkowski, University of Texas at San Antonio
David Eggebeen, Penn State University
Michael Emerson, Rice University
Ana Cecilia Fieler, University of Pennsylvania
Alan Hawkins, Brigham Young University
William Jeynes, California State University at Long Beach
Loren Marks, Louisiana State University
Margarita Mooney, University of North Carolina at Chapel Hill
Stephen Robinson, University of North Carolina at Wilmington
Christian Smith, University of Notre Dame
Rodney Stark, Baylor University
James Stoner, Louisiana State University
Peter Uhlenberg, University of North Carolina at Chapel Hill
W. Bradford Wilcox, University of Virginia
Bradley Wright, University of Connecticut

Note: Affiliations listed for identification purposes only.

[1] Mark Regnerus. 2012. “How Different Are the Adult Children of Parents Who Have Same-Sex Relationships? Findings from the New Family Structures Study.” Social Science Research 41:4 (Forthcoming).
Loren Marks. 2012. “Same-sex Parenting and Children’s Outcomes: A Closer Examination of the American Psychological Association’s Brief on Lesbian and Gay Parenting.” Social Science Research 41:4  (Forthcoming).
Paul Amato. 2012. “The Well-being of Children with Gay and Lesbian Parents.” Social Science Research 41:4 (Forthcoming).
Regnerus. 2012. P. 766.
Gunnar Andersson et al. 2006. “The Demographics of Same-Sex Marriages in Norway and Sweden.” Demography 43: 79-98; Matthijs Kalmijn et al. 2007. “Income Dynamics in Couples and the Dissolution of Marriage and Cohabitation.” Demography 44: 159-179.
Timothy Biblarz and Judith Stacey. 2010. “How Does the Gender of Parents Matter?” Journal of Marriage and Family 72: 3-22.
Daniel Potter. 2012. “Same-Sex Parent Families and Children’s Academic Achievement.” Journal of Marriage and Family 74: 556-571.




19 Responses to “Scholars Respond to Regnerus Study”

  1. MARCO says:

    The man responsible for this group, Bryon Johnson works for the Witherspoon Institute, the folks who funded Regnerus’ study in the first place. Why wasnt full disclosure made at the beginning of the story??

    Byron Johnson is Distinguished Professor of the Social Sciences at Baylor University, Director of the Institute for Studies of Religion (ISR) as well as director of the Program on Prosocial Behavior, both at Baylor. He is a Senior Fellow at the Witherspoon Institute (Princeton)

  2. Chris says:

    Same-sex marriage is one of the most contentious and vexing issues now facing our nation. It is perhaps in part for that reason that the new study on same-sex parenting

    Huh? It was my understanding that this is not a study on same-sex parenting. It’s a study on parents who have, at any point during their child’s first 18 years, had a same-sex relationship, regardless of whether the parent’s same-sex partner played any kind of parenting role. Those are obviously two different things.

    The main problem most critics of the study seems to have is that these two categories are being conflated, and the study is being misrepresented as a “same-sex parenting” study. As I understand it, the author himself has misrepresented it in this way. It’s a shame that the scholars defending him are perpetuating the same misrepresentation.

  3. Chris says:

    Or, what Phil said. Including the swear words, which in this case are warranted.

    Regnerus could have easily compared people who responded “yes” to the question of whether their parents had ever had a same-sex relationship during their childhood to people who responded “no.” That would have been a much fairer, more logical comparison. Instead, he stacked the deck against children of parents who had had same-sex relationships by comparing them to the “gold standard.” So the study is useless in assessing the effects of having a parent who has had a same-sex relationship. There are too many other complicating factors to make this study even worth conducting in this fashion.

    It is not irrational to wonder if such a bad study was put forward to intentionally mislead people.

  4. Phil says:

    Huh? It was my understanding that this is not a study on same-sex parenting.

    The impression I get is that, since “gay marriage” has been called “same-sex marriage” for some time now, certain people have internalized that “same-sex” is a synonym for “gay.”

    Regnerus has stated that he was not, in fact, studying orientation, and acknowledged that his study was not constructed to study orientation. Nevertheless, bizarrely, he labeled the group that included individuals whose father had a same-sex relationship “Gay Father” and the group whose mother had a same-sex relationship “Lesbian Mother.”

    That, right there, is pretty shoddy. Is there anyone reading this thread, or any family scholar out there who disagrees with this statement? –> Regnerus’ choice to label those two groups “Gay Father” and “Lesbian Mother” is bizarre.

    If his study had actually intended to study same-sex parents, then, logically, he could only have included in this group individuals who, at least at some point in their childhood, were raised by their mother’s or father’s same-sex partner. This, he did not do. In fact, Regnerus includes in the group all of the young adults who may never have met their parent’s same-sex partner. He also included in the group all of the young adults who were raised exclusively by the parent who did not have a same-sex relationship.

    If you were raised by a single individual, you cannot possibly have been raised by “same-sex parents.” There is no such thing as a “same-sex parent,” the term “same-sex” isn’t really a synonym for gay, because the adjective phrase cannot be applied to a set of one; “same-sex” requires a second individual in the set who is of the, well, same-sex.

    There is no such thing as an individual “same-sex” parent.

    Which really makes this quotation from the linked statement mind-boggling:

    It is perhaps in part for that reason that the new study on same-sex parenting by University of Texas sociology professor Mark Regnerus

    It wasn’t a study of same-sex parenting! I’m not picking a nit here; it really wasn’t a study of same-sex parenting!

    David Lapp, Brad Wilcox, or other Family Scholars, I’m curious about your opinion on that. In my opinion, the flaws in the opening sentences of this statement absolutely demolish its credibility. (For example, referring to a study of “adult children of parents who have same sex relationships” as a “study of same sex parenting” is a fatal flaw; it immediately reveals that you do not understand the study that you are writing about.

    My questions to David, Brad, and other scholars are:
    1. Do you agree with me that the phrases I’ve pointed out are indeed flawed?
    2. Do you agree that these flaws hurt the credibility of the statement overall?
    3. Do you agree with me that these flaws are so significant that the statement itself does not serve as any sort of credible defense of the Regnerus study?

    I’m curious about your thought processes here as rational, unbiased scholars.

  5. admin says:

    @Phil, using profanity is not appropriate on this blog. I’ve deleted your 6:27 p.m. comment.

  6. JHW says:

    This statement attributes the instability in the family situations of these children to instability in same-sex relationships. There is no indication whatsoever in the Regnerus study–none!–that this is a correct attribution.

    There are at least two further problems with making this attribution. The first is that, since the vast majority of these children were the result of different-sex relationships, there was probably at least one different-sex relationship dissolution in these children’s lives, and we have no idea what the typical circumstances involved in that were. (Regnerus’s study notes that, of the people in the “Lesbian Mother” category, 58% reported their biological mother leaving their household at some point–a contributor to instability that does not suggest dissolution of a same-sex relationship, since that would ordinarily involve their mother’s partner leaving. There are other questions: e.g., why did so many of these children spend time in foster care? It seems very hard to believe that this has anything intrinsic to do with same-sex parenting. And why the extremely high rate of adult/caregiver sexual abuse for children with lesbian mothers? Other evidence indicates that child sexual abuse is very rare in female same-sex households, which is consistent with evidence suggesting that it is primarily perpetrated by men.)

    The second is that critics of divorce, a category that I suspect includes a great many of the signatories of this statement, often tell us that the really key problematic family transition is the dissolution of the biological married family–in other words, once the child’s mother and father have broken up, whether the custodial parent then remarries or not is much less important. Maggie Gallagher made this point in her recent IAV-sponsored discussion with John Corvino. This only highlights the oddity of emphasizing instability in same-sex relationships while neglecting to consider the family instability that may have preceded (or been brought about, depending on the chronology) by these relationships.

  7. JeffreyRO5 says:

    The best take-away from this study is that same-sex couples need to be able to marry, so that they can provide much-needed stability to their children, and mixed-orientation marriages can be avoided. And those conclusions are dubious. Regnerus’ shows that instability results in poor outcomes for children; a same-sex affair may result when a straight person marries a gay person. Perhaps an affair is more likely, since there’s a basic mismatch in plumbing. He’s right, based not just on his research, but on common sense. We knew it before, but now we have solid research to support the need for legal same-sex marriage.

    I am a former senior researcher at one of the most prominent research universities in the country and if I may, this study is ill-suited for its purpose.

    1. It is neither a large nor random sample. Only 200 people with a parent who had a same-sex (and adulterous) affair, as opposed to people actually raised in same-sex parent households.
    2. Hmm….unstable marriage because one of the spouses had a sane-sex affair)….hmmm….bad outcomes for children…..hmm….let’s blame it on the gayness of the affair, not the affair itself! Please! Compare children whose adulterous parents had straight affairs to children whose adulterous parents had gay affairs, maybe, but this setup? How does “being raised by happily married straight parents” get compared to “raised by mixed-orientation parents where one of the parents had an affair”? This looks too close to NOM’s exposed tactic of trying to get the children of gay parents to rat out their parents.
    3. He should be clubbed over the head for drawing the conclusion that gay parenting is to blame for whatever troubles these children have. Unstable marriages, likely due to one spouse having an affair, are the more likely culprit. It’s not that the spouse had a same-sex affair, although that creates the issue of whether the marriage can be salvaged at all; it’s that a spouse had an affair.

    “We do not think that these new studies settle the nation’s ongoing debate about gay parenting, same-sex marriage, and the welfare of children”

    What on-going debate about gay parenting are you referring to? Same-sex, or gay, parenting is legal in all 50 states, and there is no effort to change that, that I’m aware of, and I read about this stuff a lot. Are you trying to start a debate about same-sex parenting? Is that why you support this flawed study’s conclusions about same-sex parenting?

    “has helped to inform the ongoing scholarly and public conversation about same-sex families in America.”

    Except that Regnerus didn’t study same-sex families, he studied different-sex families where a spouse had a same-sex affair.

    “Indeed, it is possible to interpret Regnerus’s findings as evidence for the need for legalized gay marriage, in order to support the social stability of such relationships.”

    Even that’s a stretch but I can see where you buy credibility by acknowledging the adversarial position.

    “As social scientists, our hope is that more such studies will be forthcoming shortly”

    More flawed, dishonest studies? Really? What purpose is served by propaganda cum research?

    I think this is a “let’s put on a study!” attempt to add secular gravity and respectability to the religion-fueled anti-same-sex marriage campaign, and it might backfire. Most people will see this for what it is: propaganda. It didn’t work for creationism when that was repackaged as “intelligent design” and peddled as science, with an institute and a bevy of “scientists” supporting it. It’s still not taught in schools, not reputable ones at least. I doubt a scientific makeover will work for opposition to same-sex marriage, either.

    Like all good propaganda, this study isn’t there to persuade but to rationalize. “Folks, your suspicions about gays being defective (as humans, as Christians, and now, as parents) were spot-on! And now we’ve got the proof!”

  8. Phil says:

    Thanks for the heads up! I shall attempt to rewrote the comment without the profanity. For the record, in case anyone is joining this discussion, my use of profanity was not directed at anyone here, or at human beings. It was for rhetorical effect.

  9. Philip Cohen says:

    By a quick check, 9 of the 18 signers are Witherspoon fellows or Templeton Foundation grant recipients. Among the others, one has co-authored with Maggie Gallagher, one testified against SSM in Iowa, and another wrote the review accompanying Regnerus. A little disclosure woulda helped put the statement in perspective.

  10. Box Turtle Bulletin » Regnerus’ Defenders Miss the Point

    A really excellent blog post by Rob Tisinai responding to the letter.

  11. David Lapp says:

    Phil and Chris,

    I agree with your critique that it’s not completely fair to call it a study of same-sex parenting for the reasons that you outline. And I agree that it would have been helpful to report the differences between only those kids who ever lived with the parent while the mom/dad was in a same-sex relationship.

    Mark Regnerus addresses this critique in his recent exchange with Will Saletan at Slate. See here:

    In that article, he tells us the results of the 81 young adults who lived with their mom while in a same-sex relationship “at least a good share of a year or more.” And he reports that out of 40 outcomes, after controls, there are still 18 outcomes in which there are significant differences compared to the kids from bio, intact families.

    With all that said, let’s not forget that of all the kids who reported their mom was ever in a same-sex relationship, 91 percent of them ever lived with their mom while she was in the same-sex relationship. (And that’s the group with the most troublesome outcomes.)

    Also, let’s not pretend that a parent’s relationship — even if the child is not living with that parent — is of no consequence to many children growing up. Of course a mom or dad’s relationship — even if from a distance — often profoundly affects a child’s development. Regnerus seems to understand this by virtue of the fact that one of the eight family structures he looks at is young adults whose parents divorced after the young adult was age 18.

    In that light, including kids who NEVER lived with their parent while he/she was in a same-sex relationship makes more sense, and it suggests some consistency to his method. (It would appear that he’s not arbitrarily stacking the deck.)

    All that said — and I’m not a social scientist, so I’m willing to stand corrected! — to an average follower of the debate like me, it seems like it would have been helpful to only examine the kids who ever lived with their parent while he/she was in a same-sex relationship.

  12. David Lapp says:

    Philip Cohen –

    You’re a social scientist. Before you say anything publicly about same-sex issues, do you make it a habit of disclosing which foundations you’ve ever received money from or been affiliated with (and whether those foundations are liberal or conservative), any critical things you’ve ever written about Maggie Gallagher or Robby George, and if you have ever spoken publicly in favor of or against SSM?

  13. Jeffrey says:

    But isn’t it important to know who or what is motivating an organized defense of controversial research? What the bias of the defenders is?

  14. JHW says:

    David Lapp: Thanks for that link. I find this part of Regnerus’s response particularly striking:

    How about those 81 cases wherein respondents reported living with mom and her partner for at least a good share of a year or more? While hardly an ideal sample size, the story there changes little from that which appeared in the study (in which my analytic sample size was 163, and included LMs who lived with their mother and her partner and LMs who did not): 20 out of 40 outcomes reveal simple statistically-significant differences with the “intact bio family,” and 18 after controls. This dip from 24 statistically-significant differences after controls (the number in the study) to 18 differences here appears largely a result of declining statistical power, not genuinely disappearing distinctions.

    In fact, the two LM groups (those raising kids while living with their partner in the home, and those raising kids with a relationship partner who didn’t live with them) look pretty similar to each other on most counts. Those young adults who reported LM-and-partner togetherness also reported greater current unemployment, public assistance, and expressed greater insecurity about their family of origin, though they appear to report slightly less depression than those respondents whose parent’s same-sex relationship was never “residential.”

    What Regnerus is telling us here, if I understand him correctly (he’s not entirely precise in his description), is that, if you break the “lesbian mother” group into (a) respondents who reported living with their mother and her same-sex partner for “a good share of a year or more” and (b) respondents who did not, the group in category (a) did slightly better than the group in category (b), but was not all that different. He appears to think that this somehow nullifies the criticism of his study. But actually, the opposite is true. What this comparison actually shows is that the presence of a same-sex partner in the home made little difference; in other words, whatever the cause for the differences between people in the LM group and people in the “intact biological” group, the cause does not appear to have had anything to do with same-sex parenting.

    What might tell us something is if he matched the 81 respondents reporting cohabitation with their mother and her same-sex partner with respondents in the LM category who reported cohabitation with their father and his different-sex partner, post-dissolution of the original different-sex relationship. This would be a problematic comparison too, because you’d have to deal with overlap, because the cohabitation patterns of same-sex and different-sex relationships are probably different, and because custody determinations are endogenous. But it would be a comparison more likely to shed some light on same-sex parenting than this comparison.

  15. Philip Cohen says:

    David – the short answer is, “Yes,” since all that stuff is on my blog and CV.

    Of course not everyone sees that every time I make a public comment. But if I were to comment on a particular person’s work I would want to disclose my institutional affiliation with that person (e.g., being a fellow at the same institute, or having funding from the same foundation).

    It’s a question of what’s relevant and helpful, not just what’s ethical. Ethically, I may be covered by the transparency of my public record. But to put my comments in perspective I would want to disclose that stuff if it was relevant to the specific case.

    Thus, the fact that I received money from the Russell Sage Foundation to do a Census report is not especially relevant to my comments on marriage equality – but if you think it’s important the information is on my CV.

    I also think funding sources and personal affiliations are more important to disclose than personal viewpoints expressed in other forums. I’m not saying they were wrong not to disclose all this in their statement – since it is discoverable – but I think it’s relevant information that could help others put their statement in context, which is why I made the comment.

  16. JHW says:

    Another comment on the stability of same-sex couple households, since that seems to be one key underlying issue in this discussion. In the Slate response to Saletan David Lapp links to above, Regnerus also says this:

    We had only two cases in which mom and her partner were together for 18 years. We’ve got only six cases where mom and her partner were reported to have stayed together for 10 or more years, and 18 cases for five years.

    I assume he is getting this from the calendar he asked the respondents to fill out. But the calendar, which you can find here (p. 12), only asks about living arrangements. You could get from this that the respondent lived with their biological mother and her partner for 18 years, or 10 years, or five years, but you can’t get relationship length. Prior to the respondent turning 18, you would be able to detect a disruption of cohabitation between the biological mother and her same-sex partner, if the biological mother was still raising the child (one of the most striking findings of this study is still the high rate of biological mother exit), but there is no direct evidence about relationship length.

    This really tells us very little about household stability among same-sex partners generally. Not only is the evidence indirect, but it is also probably non-representative, and there is no way to know how serious these relationships were, or how much involvement in parenting the same-sex partners had. If we want to know about planned same-sex parenting, we should be looking at relationship duration and outcome in planned same-sex couple families; the National Longitudinal Lesbian Family Study did, and found that (for relationships for the most part existing prior to formal recognition) rates of separation were somewhat higher than typical of different-sex married couples raising same-age children, but child outcomes were not worse. (Somewhat oddly, no significant difference was found even when adolescents whose mothers had separated were compared to adolescents whose mothers had not, which the authors suggest could have something to do with the prevalence of shared custody and amicable break-up among the female same-sex couples in the study.) And if we want to know about marriage, we should be looking at relationship stability among same-sex couples with formal legal recognition; with the single exception of the study from Sweden and Norway social conservatives are always citing, all the other evidence I know of indicates that same-sex couples with such formal recognition break up at comparable or lower rates than different-sex married couples. (One study that directly compared same-sex couples in civil unions to same-sex couples not in civil unions found that those in civil unions did not break up at significantly higher rates than different-sex married couples, while same-sex couples not in civil unions did; see p. 109. There is actually a substantial body of evidence, going at least back to Blumstein and Schwartz’s work a few decades ago, that while casual cohabitation appears to be more common among same-sex relationships than different-sex relationships, same-sex relationships with strong indicia of commitment, be it seeking formal recognition or just staying together for a long time, tend to be fairly stable. And contrary to stereotype, there are a whole lot of those relationships; see, e.g., this study, pp. 584-585.)

  17. Comrade Svilova says:

    Still, living with a mother and her female partner for ‘a year or more’ is not the same as being raised in a planned, lesbian-headed household from birth.

  18. olterigo says:

    Barry, thank you for the very informative link.

    Especially, once you look in the comments of the “StraightGrandmother” there. Quite informative. It looks like the studies these “researchers” cite pretty much rely on the same kind of characterization of “same-sex households” or families. Even to the point of possibly classifying a woman raising her child in the house with her own mother (the child’s grandmother) as a same-sex family…

    But, I guess, if they don’t see a problem with Regnerus’s slight of hand (and even engage in it themselves with their constant sliding to “same-sex families”), why should it bother them that the studies they rely on also make use of the same slight of hand?

  19. Matt N says:

    I think a major problem here is that the 2005 critique by Regnerus of the APA’s policies focused very intently on unintentional (at least, I believe so) flaws in existing studies of same-sex couples and LGB(T?) parents. I think it’s great if Regnerus wants to create better research that more directly tries to isolate differences between children more and less exposed to gay, lesbian, and other non-heteronormative parent(s). That said, doing so while replicating many of the flaws he originally complained about makes it seem like he cared more about the findings than the studies’ rigor.