Note #1: I wrote this post a few days ago and pre-posted it for today. By chance, The Chronicle of Higher Education carried an opinion piece, “An Academic Auto-da-Fé,” by sociologist Christian Smith of Notre Dame that came out today. He captures many of the same points, and makes them well.
Note #2: At the time I wrote this post, I was under the impression that “Scott Rose” was a nom de plume. Mr. Rose informed me otherwise (see comment) and this post has been changed to use the name Scott Rose throughout.
Mark Regnerus has been accused of research ethics violations; most such accusations accuse researchers of trimming their data or making it up entirely, like the two psychologists recently tagged by Uri Simonsohn.Regnerus is facing something different.
Regnerus, of the University of Texas at Austin is facing a complaint made to the university president that his research is unethical because it is “a study designed so as to be guaranteed to make gay people look bad, through means plainly fraudulent and defamatory.” In a blog, the complainant Scott Rose elaborated, saying that “the main political aim of the NOM-leaders-funded, invalid Regnerus study is to demonize gay people in an election year.”
Focus on children’s outcomes
The Austin Statesman summarized the research findings at the heart of the debate:
“Bucking the consensus of the past decade of scholarship that the sexual orientation of parents does not negatively affect children in consequential ways, Regnerus found that adults with gay parents tended to report lower levels of success in economic and romantic pursuits and struggled more with mental health issues.”
Since the publication of his research in the peer-reviewed journal Social Science Research, Regnerus experienced a lot of media attention, followed by a firestorm in the blogosphere. As the Austin Statesman reported, Regnerus’ findings differed from those of publicized earlier studies – and were perceived to be ideologically biased. That accusation is routinely tossed back and forth when research findings are relevant to contemporary political debates. NARTH publishes similar claims that the research of a leading pro-same-sex couples psychologist, Charlotte Patterson, is unscientific and biased. They have not yet filed a complaint with the University of Virginia.
What is research bias?
Rose points to a sentence in an interview Regnerus gave to an alumni magazine for his alma mater, where he said, “I believe that if your faith matters, it should inform what you teach and what you research.” This is a pretty standard statement – people choose to conduct research on topics that are, in their view, important and interesting. Franz de Waal studies primate empathy because he believes it offers “lesson’s for a kinder society” according to his book title. Gordon Hodson was willing to grapple with three hot-button topics at once - intelligence, racism, and political conservatism – because he thinks it’s important to understand the foundation of racism. In short, the things that researchers think matter “inform” their choice of question and topic.
Research methodology, on the other hand, are the practices of a trained mind, the standardized procedures, the approaches to gathering and analyzing data that maintain the standards of science: objectivity. A researcher asks a question because she is interested in the answer – but is willing to accept whatever answer arrives.
No perfect research
The research endeavor is filled with choices and trade-offs: in-depth interviews of a smaller number or surveys of a larger number? Which questions to ask? Which statistics to use? Which group(s) to compare with which other groups? Researchers frequently criticize each other – even criticize the shortcomings of their own work – for the choices they made. Regnerus’ study has shortcomings. He identifies some. Even his supporters name others. But they do not amount to falsification or intentional mis-reporting of results. They point to the need for better data that would produce a larger sample of people from stable same-sex families.
When researchers criticize each other’s work, they often replicate or extend the research, using different methods, to see if the results change. Whether the answer is yes or no, the second study is as important as the first: it may confirm a pattern, or call it into question, hinting that the underlying relationship is more complex than initially thought. This is the method of science for dealing with possibly incorrect results.
A number of excellent social scientists wrote a response to the critics of Regnerus’ research. Their three primary points deal with the adequacy of the research design – that the previous studies also had methodology problems, that the groups compared (children of gay or heterosexual parents) differed on other characteristics too, and that another study published the same month had similar conclusions. At the very end of their statement, these social scientists pointed out that the findings could be used to make a liberal political argument – that children suffered poorer outcomes because their parents’ relationship was stigmatized rather than as a direct result of that relationship. In short: the research does not have a direct correspondence to one political view.
How to respond to Regnerus’ research?
William Saletan, writing for Slate, points to the best response to Regnerus’ research: get into the data. Instead of accepting the ten-second sound bite version, he read and dissected the study, noting another factor – the amount of family stability – that seemed to be a main player. He calls for more research and more analysis – and for researchers of all ideologies to use these data to ask deeper and better questions.
The blogosphere, on the other hand, is full of blood. Scott Rose and others in the pro-gay-marriage movement write calls to action with a 3:1 ratio of inflammatory adjectives to names and nouns. Some from the pro-traditional-marriage movement write with equal fervor for policies. Both deal primarily with the sound bite and less with the intricacies of the relationships shown in the data. These hyped up posts from both sides only rev up their core constituents for political battle: they do nothing to move our understanding forward.
The UT-Austin said it would decide on the complaint after a review of Regnerus’ work, but within the next 60 days. UT-Austin has a highly-ranked prestigious sociology department. I certainly hope that their decision leaves room for researchers to continue to take up difficult questions with the best data they can get. I hope Regnerus is an AAUP member; he may need some academic freedom protection.
- Scholars defend findings of difficulties among same-sex parented children (catholicnewsagency.com)
- U. of Texas Investigating Mark Regnerus for Scientific Misconduct Over Flawed, Anti-Gay Parenting ‘Study’ (towleroad.com)
- UT-Austin scrutinizes ethics of controversial same-sex parenting study (insidehighered.com)
- More social scientists stepping forward to defend Regnerus study (winteryknight.wordpress.com)
- 200 Scholars Sign Critique of Regnerus Paper (familyscholars.org)
- Same-sex parenting study hits courts (salon.com)
- Regnerus, and Our Enlightened Public Discourse (familyscholars.org)