Paul Amato on reviewing Regnerus

I recently discussed Paul Amato’s role in the Regnerus Affair. I offered my opinion that, if Amato was a reviewer of the Regnerus article, he should not have been, mostly because he had served as a paid consultant on the study. (My long version of the Affair is here, a critique of the paper I co-authored is here.)

I regret that before writing that post I didn’t directly ask Amato if he wanted to discuss his role and whether he served as a reviewer. After the post appeared he sent me this statement, which I agreed to post. I added some more comments of my own below. (He also reminded me that he had voted for the Family Section Council resolution asking ASA to respond to the Regnerus study, which they did.)

Amato is President Elect of the National Council on Family Relations and a Distinguished Professor at Penn State University.

Thoughts on the Mark Regnerus (2012) study, by Paul Amato

One year has passed since Mark Regnerus (2012) published a highly controversial article on the children of parents who have same-sex relationships. Given that time tends to improve one’s perspective, this seems like a good time to reflect on the study and its aftermath.

My involvement

I worked for two days at the University of Texas as a consultant on the New Family Structures Study (NFSS). As I recall, seven consultants were at the meeting, along with a statistician from the survey research organization that later collected the data. I consulted primarily on sampling and measurement issues, and I was paid for two days of my time, plus travel expenses for myself (and my wife, who accompanied me). I charged for two days at my usual fee, which is $150 per hour. So I earned about $2,400. I received no further compensation after that.

About six months later, the editor of Social Science Research (SSR) asked me to review a manuscript written by Mark Regnerus. I informed the editor that I had worked as a paid consultant on the survey on which the manuscript was based. The editor said that he would like to have my views on the paper anyway, so I shared my views as honestly as I could.

This situation comes up now and then in my experience. When reviewing manuscripts for journals, I occasionally discover that I know the author and have some sort of relationship with the author or the study. In one case, for example, the author was a friend and colleague of mine, and I had read an earlier version of the paper and provided comments to the author. In this and every other case in which I have brought information like this to the editor’s attention, the editor has asked me to do the review anyway. Journal editors often have a difficult time getting reviews, and I assume they treat these reviews as one more data point. So the editor of SSR was doing what other editors do, as far as I know.

Was this particular case a conflict of interest for me? The American Sociological Association (ASA) defines a conflict of interest in the following manner:

Conflicts of interest arise when sociologists’ personal or financial interests prevent them from performing their professional work in an unbiased manner.

With respect to the Regnerus manuscript, I had no personal or financial interest in whether the paper was published. So by this definition, there was not a conflict of interest. Of course, sometimes there is the appearance of a conflict of interest. In these cases, the ASA code states:

Sociologists disclose relevant sources of financial support and relevant personal or professional relationships that may have the appearance of or potential for a conflict of interest…

As noted earlier, I disclosed to the editor that I had worked as a paid consultant on the NFSS. I also disclosed my role as a paid consultant in the commentary that I wrote for the Regnerus article, which appeared in SSR. I never attempted to hide the fact that I was part of the team that consulted on survey design.

In retrospect, I understand that providing a review was not a good idea, because one should avoid even the hint of impropriety in matters like this. At the time, however, I simply felt that I was helping the editor and being a good colleague.

Contrary to the views of some (but not all) of my colleagues, I thought the Regnerus manuscript was worth publishing. My key recommendation, however, was that the editor should publish the paper with commentaries from authors who hold a variety of perspectives, including gay and lesbian scholars who had published in this area. I believed that the Regnerus paper, accompanied by a diverse set of commentaries, could represent a useful contribution to the literature on LGBT families. Unfortunately, the editor was unable to recruit any gay or lesbian scholars to contribute commentaries, so my idea for an exchange of views fell flat. (The subsequent issue of SSR devoted to the controversy came closer to what I had envisioned.)

Almost everyone got it wrong

When the study was published, criticism from the political left was swift and harsh. Unfortunately, some commentary devolved into ad hominem attacks, accusations of fraud, and name-calling. Rather than intellectually engage the findings, the goal of some critics was to thoroughly discredit the study—and the author. While they were at it, many critics also attacked the editor, the reviewers, the consultants, those who wrote commentaries—even the survey research firm that collected the data! Anyone with any form of contact with the study became an enemy of the people.

This is unfortunate, because the political left could have benefitted from a strategic appropriation of the findings. The study involved a national sample of young adults with an LGBT parent. As the study noted, few of these young adults spent long periods of time in households with two parents of the same sex. Instead, most were born into heterosexual families that later broke up, presumably when one parent came out as gay or lesbian. Many of these youth went on to experience a variety of other family structures before reaching adulthood. One out of seven spent time in foster care. Previous research shows that instability in the family of origin increases the risk of a variety of long-term social and psychological problems for offspring. Consistent with this research, young adults in the study had modestly elevated problem profiles. It is reasonable to conclude that the elevated number of problems observed in these young adults was due to family instability rather than the sexual orientation of parents. For this reason, most observers have noted correctly that this study contributes nothing to our understanding of how children fare when raised by same-sex parents in stable households

Rather than dismiss these finding as being irrelevant, however, it’s useful to dig more deeply into the results. Why did these marriages end in divorce? More importantly, why did gays and lesbians wind up in heterosexual marriages in the first place? The explanation probably would go something like this: Like heterosexuals, many gays and lesbians wish to have families and raise children. But a generation ago, intolerance was the rule and discrimination against gays and lesbians was endemic. For many, forming heterosexual unions appeared to be the only way to achieve the dream of family and children. But these unions tended to be unstable, with problematic consequences for adults and children. Presumably, as our society becomes more accepting of LGBT families, the unfortunate circumstances of children and parents described in the Regnerus study will become less common. The freedom to marry, in particular, should increase stability in the lives of children with gay and lesbian parents.

In short, findings from the Regnerus study can be interpreted as strong evidence in support of same-sex marriage. The American Psychological Association and ASA research briefs emphasized the fact that almost all prior studies found no differences between children with heterosexual parents and children with gay or lesbian parents. The “no difference” perspective suggests that children will not be harmed by same-sex marriage. The lesson from the Regnerus study, however, is that children thrive on family stability, including children with gay and lesbian parents. We know that marriage tends to stabilize relationships, yet same-sex marriage is not allowed in most states. Given that children benefit from the stability provided by marriage, it is unfair and unkind to deny children the right to live with married parents. In contrast to the “no difference” perspective, a “family stability” perspective implies that we need to change our laws NOW to protect and benefit children.

If the political left missed an opportunity by failing to understand the full implications of the Regnerus study, the political right made even more serious blunders. Many conservative observers have cited the Regnerus study as if it provided evidence that being raised by gay or lesbian parents is harmful to children. This claim is disingenuous, because the study found no such thing. A noteworthy example came from Regnerus himself, who signed an amicus brief to the Supreme Court citing his study as evidence against same-sex marriage. This is curious because on page 766 in his 2012 article, Regnerus stated that his study was not intended to either affirm or undermine the legal right to same-sex marriage. And on page 768 of his response to the commentaries in the same issue, he stated that his data should not be used to press any political program. Given these cautious early statements, it is exasperating to see Regnerus later cite his own study as evidence against same-sex marriage.

Concluding thoughts

Many observers have argued that the Regnerus study should never have been published. It is important, however, to focus on what the study actually showed, and not on what people claim that it showed or wanted it to show. The study showed that family instability is not good for children, and many children with gay and lesbian parents, a generation ago, experienced a lot of family instability. It is not difficult to see how the personal problems of these families were affected by the restrictive social milieus in which they lived.

Since the Regnerus study was published, studies by Potter (2012) and Allen, Pakuluk, and Price (2013) have shown associations between having same-sex parents and child problems. Like the Regnerus paper, both of these articles survived the peer review process and, in fact, were published in top-tier social science journals. Rather than try to discredit these studies (and any future studies that may show similar results), it is better to examine the findings carefully and figure out what is going on. In fact, both studies are entirely consistent with the family stability perspective described earlier.

In conclusion, the political left discredited the Regnerus study without fully considering its findings, and the political right used the study disingenuously to further their political goals. Few people have focused thoughtfully on what the data actually show and what we can learn from the study. The controversy over the Regnerus study provides a sobering illustration of what can go wrong when ideology distorts social research.

References

Allen, Douglas W., Catherine Pakaluk, and Joseph Price. (2013). “Nontraditional Families and Childhood Progress through School: A Comment on Rosenfeld.” Demography 50:955-961.

Amato, Paul R. (2012). “The Well-Being of Children with Gay and Lesbian Parents.” Social Science Research 44:771-774.

American Sociological Association. (undated). American Sociological Association Code of Ethics (http://www.asanet.org/images/asa/docs/pdf/CodeofEthics.pdf)

Potter, Daniel. (2012). “Same-Sex Parent Families and Children’s Academic Achievement.” Journal of Marriage and Family 74:556-571.

Regnerus, Mark (2012). “How different are the adult children of parents who have same-sex relationships? Findings from the New Family Structures Study.” Social Science Research 41:752-770.

Regnerus, Mark (2012). “Response to Paul Amato, David Eggebeen, and Cynthia Osborne. Social Science Research 41:786-787.

Philip’s followup comments

I don’t think it was a gross ethical violation for Amato to review the paper, in which material or ideological gain lay behind his decision to do the review and led him to recommend publishing the paper against his better professional judgment. Rather, he thought it was a reasonable paper and offered that opinion when asked — which is unsurprising given his involvement in the project. So my real disagreement with Amato is over the value of the paper. I think it’s a worthless paper, done wrong, and only advanced because of the author’s ideological attachment to its results; that it accidentally helps reveal something true about family instability does not make it worth publishing.

Being a consultant on a small project like the NFSS is not like being a consultant on a giant project like the National Longitudinal Study of Adolescent Health. Consulting on Add Health, which has yielded thousands of publications and involved dozens of experts at many agencies, does not disqualify a person from reviewing a study using that data. In contrast, the NFSS has so far yielded one paper, solo-authored by the PI, who personally invited the several consultants. With only a few people involved, being one of them matters more.

Anyway, beyond the appearance of conflict, the problem with Amato serving as a reviewer is that it did not provide an outside perspective to the editor, James Wright. How could Amato’s review help Wright make the decision? Getting the input of a consultant on the project might help an editor shape a revision or build a special issue, but given Amato’s involvement his endorsement should not have counted as part of the peer review process. When Amato revealed his role, Wright should have declined his review.

Taking for granted the unethical behavior of Regnerus, and Brad Wilcox, on whose behalf Regnerus acted, the real failure here is by Wright. Instead of seriously reviewing the paper, he essentially whispered into an echo chamber of backers and consultants, “We should publish this, right?”

I believe the paper should be retracted because the conclusions are demonstrably wrong, because the author lied in the paper about the involvement of the institute that funded it, and because the peer review process was compromised by conflicts of interest. As long as this remains uncorrected, and James Wright remains editor, the integrity of the journal is indelibly tarnished.

While Wright is editor, I will no longer review for or submit to Social Science Research. I hope others will join me in that decision.

Comments that rehash well-known opinions or make person attacks will be shortened or deleted.