Protecting Science From Harm, And Against Harmful Science

sosThe activists are coming!  And, so they should.  A supposedly “debunkedstudy by Mark Regnerus that does not employ valid measures of lesbian couples worked its way right into a US Supreme Court case on marriage equality.

We, as sociologists, did all that we could: 1) petitioned the journal in which it was published, Social Science Research, 2) published critiques of his and Loren Marks‘s studies in the journal, 3) wrote to the media to point out the study’s flaws, 4) offered extensive methodological critiques (e.g., blogs), 5),  petitioned the leadership of the American Sociological Association (ASA) to make a public statement against the Regnerus study, 6) conducted an internal audit of the peer review process, and 7) submitted a brief to the Supreme Court as a discipline to make clear no evidence exists to worry about LGBT families.  And, there may have been other efforts of which I am unaware.

But it wasn’t enough.  Regnerus and other conservative scholars submitted their own amicus brief to the Court.  And, somehow, this one study counters all of the other studies enough that Supreme Court Justice Scalia noted:

If you redefine marriage to include same-sex couples, you must — you must permit adoption by same-sex couples, and there’s – there’s considerable disagreement among — among sociologists as to what the consequences of raising a child in a — in a single-sex family, whether that is harmful to the child or not. Some States do not — do not permit adoption by same-sex couples for that reason.

Aftershocks

The American Sociological Association released another statement thereafter to clarify that Regnerus’s study was flawed.  While imperfect, every other study suggests no evidence that children of same-gender families are worse off in terms of health, adjustment, academic performance, etc.  And, the Department of Sociology at the University of Texas-Austin held a mini-conference on LGBT families last week, featuring Gary Gates and other big names in LGBT research.  I assume this was part of the department’s effort (which started as soon as Regnerus’s study was published) to show that others in the department are doing great, pro-LGBT work.

But, it is too late.  We do not yet know the outcome of the Supreme Court cases.  And, it is unclear whether Regernus’s “debunked” study will be cited by other researchers, politicians, or in other court cases.  These are, indeed, real possibilities because his study has been “debunked,” but not retracted.  That means it still stands as a peer-reviewed, published academic article — albeit critiqued and discounted.

The lengths that these activists are going makes sense.  Though we got to the point where we felt comfortable with the “debunked” status of Regnerus’s paper, it still caused damage — on our watchDespite our intentions and efforts as a discipline, we did not do enough to prevent this study from having an impact in the fate of LGBT rights (in this case, marriage equality).  Whether it comes from religion, science, politics, education, or some other institution, threats to your rights are just that, so who wouldn’t shift into self-defense mode?

Protecting Against Harmful Science

My primary concern, which I have voiced in the discussions among sociologists, is what are we doing to prevent further harm to the community that has been affected by this study?  On our watch, a study that should never have reached publication ended up reaching the Supreme Court.  We alerted others, “watch out!”; we critiqued Regnerus’s actions, “he’s not even measuring it right!; and even issued a formal statement saying, “we’re not with this guy, he’s crazy.”  But, all while we watched Regnerus set up a very calculated assault on LGBT Americans.  Since fellow sociologists have so vehemently opposed releasing the names of the peer reviewers of the study, and do not feel compelled to push for retraction, I continue to ask, so now what?

I cannot believe I have to raise this question.  But, it seems some are more concerned about protecting science than protecting people from science.  There are general principles regarding ethical scientific practice (including discipline specific guidelines), and the universality of Institutional Review Boards to ensure researchers at universities are not causing harm to their participants.  Unfortunately, these guidelines were developed as a response to very unethical and harmful research in the past:

  • During the Holocaust, the Nazis conducted many experiments on Jews (including children)
  • The “Tuskegee syphilis experiment” (1932-1972), in which poor African American men were infected with syphilis without their knowledge nor with treatment: “The 40-year study was controversial for reasons related to ethical standards; primarily because researchers knowingly failed to treat patients appropriately after the 1940s validation of penicillin as an effective cure for the disease they were studying.”
  • Similar experiments were conducted in Guatemala from 1946-1948.  Over 80 people died as a result.
  • The use of Henrietta Lacks‘s cells without her or her family’s permission or knowledge in 1951.
  • Stanley Milgram’s 1961 psychological experiments on obedience, in which he deceived subjects into thinking they were delivering shocks (sometimes deadly) as punishment to a person completing a faux task.  Ethical concerns have been raised about the Stanford prison experiment, as well.
  • Tearoom Trade (1970) — Laud Humphreys’s study of same-sex sexual encounters in public spaces without their knowledge or consent; after observing the men, he used their license plate numbers on their cars to track down their home addresses to interview them (sometimes in front of their families).

For all of the positive things that have come from science (even from some of the awful exploitative, dangerous experiments above), science is sometimes used for evil.  Too often, marginalized communities are the targets of harmful science.  Of course, in this case, Regnerus and his colleagues did not have any direct contact with their participants; and, there is little reason to suspect that Knowledge Networks (which carried out the survey) caused any harm.

However, I argue that we have an obligation to ensure that harm is not caused in the activities that come after research is conducted: how the research is used and for what purposes.  Some argue that, even when studies are carried out for good, we owe it to our participants to give something in return — immediate and tangible, not just “thanks for advancing science!” — for opening up about their experiences, backgrounds, thoughts, opinions, and feelings.

So, now what are we doing to protect this marginalized community that has been further harmed by science?  What can we do?  Below are some things that have been suggested, and my thoughts on them.

Speaking Out, In General

It is important that we speak out about this scandal, in general.  Unfortunately, it feels as though some sociologists feel they have done all that they could and just want this to go away already.

But, who speaks for us?  I may be wrong, but many of those — “some sociologists” — do not appear to either be LGBT themselves nor do they study LGBT communities (I’m including here bloggers and those who have left comments).  So, maybe it is simple to walk away from this when you can return home to your legally-recognized spouse after a day’s work.  Unfortunately, it appears that the sociology bloggers at orgtheory and scatterplot are serving as The Voice for the entire discipline, and the LGBT activists are in direct dialogue with them.  I wonder what LGBT sociologists and sociologist of sexualities have to say about this scandal, and whether they feel that we have done enough.

I worry, as I have before: who gets to speak?  The subfield of sexualities in sociology is relatively new and disproportionately young.  We must tread lightly.  And, it is likely that many have remained silent on this issue because they are soon to be or are currently on the job market; or, they are on the tenure-track; or, even with tenure, they are at the margins of their department and the discipline as a whole.  Or, just like other fields, maybe some sexualities scholars see their work as irrelevant to activism.  And, even for those of us who do pursue activism, we risk professional consequences.  But, even those who are not explicitly involved in activism may be the target of political witch hunts or other external threats, or lack of support from the academy to do our research.

silenced

Retract It Already

The retraction of published studies is more common than I realized.  But, it looks like there is no movement to retract the Regnerus study.  There is a lot of shadiness, omission of important details, and conflict of interest sprinkled throughout this entire scandal.  But, within conservative standards of “when to retract,” Regnerus’s study is safe.  It was the peer review process that is problematic.  Specifically:

[T]he paper was submitted for publication 20 days before the end of the data collection, and 23 days before the data were delivered to the University of Texas! That’s fast.

There must be some post-hoc excuse Regnerus or the journal could give to clear this up.

That is in addition to the serious methodological problems that the reviewers should have caught.  That is more than enough for some to call for the study’s retraction.  Okay, so, since this is not Regnerus’s fault, per se (short of questionable political motivations and funding sources), retract the study and then invite him to go through the peer review process again — this time with different reviewers who are not his colleagues.

“Out The Reviewers!”

LGBT activist John M. Becker has moved forward in demanding records from Social Science Research, namely to out the reviewers of the Regnerus study.  Some of my fellow sociologists have been talking about this — I’m sure informally, but in this case publicly on blogs.  Some have taken issue with Becker’s efforts, suggesting that it subverts the sanctity of the peer review system for academic publishing; to reveal the identities of anonymous reviewers is a threat to the entire scientific enterprise.   Oh, and does it get ugly when sociologists and activists go head to head.  But, understandably, when outside forces threaten science (e.g., forced oversight, taking away funding), we necessarily lash out in self-defense.

But, I wonder what would happen if we did reveal the names of those scholars who reviewed Regnerus’s study.  Recently, while reading one article about the source of whites’ attitudes toward race-based attitudes, I noticed that the reviewers were explicitly named, right on the first page:

Editor’s note: The reviewers were Lawrence Bobo, Warren E. Miller, David O. Sears, and Susan Welch (p.723).

I decided to search Google for “editor’s note: the reviewers” to see if this was a fluke.  I came across two other journals that have, or at least used to, explicitly name the reviewers of a published article, Teaching Sociology and Sociological Inquiry.  In the case of the former, I thought maybe as it has become more popular, and moved toward publishing more empirically-based articles, the editorial board might have dropped this practice along the way.  But, even a recent article, by sociologist Janice McCabe, dawns the editor’s note, naming each reviewer.  It looks as though Sociological Inquiry published the names of authors just for a few years in the early 1990s.  These are not the top journals of the discipline, but this discovery leaves me wondering what the harm would be to reveal the names of the publishers in this instance — in this case in which the peer review system was abused, misused, or underused (depending on your perspective).

This is not a question of whether sociology or any other academic discipline should maintain anonymous peer review for publishing.  While imperfect, it strengthens science and minimizes (some) concerns about bias.  If anything, I see room to strengthen the peer-review system further.  And, let’s set aside the potential harms of the overwhelming pressure to publish for jobs, tenure, promotion, etc. as well.  The question here is what harm would be caused to the peer review system, or even the entire scientific enterprise, if the reviewers of this one “debunked” study were revealed?

That some journals have revealed the names of reviewers — including articles that are ethically and politically sound — leads me to suggest that the sky will not fall if Becker is successful in his demand for the SSR records.  Science will still exist the following day.  But, I do agree that this may not actually get us any further in squashing Regnerus’s study or the harm caused by it.

Fight Fire With Fire: More Research!

As Fabio Rojas suggested in response to my plea to do something to take this study down, another possibility is to simply beat Regnerus at his own game.  Do more, better research.  Indeed, sociologists Andrew Perrin, Philip Cohen, and Neal Caren have done just that in a forthcoming article in Journal of Gay and Lesbian Mental Health — even using the New Family Structures Study data. (Of course, they find that Regnerus’s conclusions were bogus and methodologically flawed.)  I do hope, however, that awareness of their new study spreads, as JGLMH is a psychiatry journal and has a so-so impact factor.   But, Perrin makes clear that this journal was chosen because of the speedy turn around, and it actually sent out a call for papers to address the Regnerus scandal.

As Michael Bader notes, this scandal has sparked even better work, and maybe science will be even stronger in the first place.  But, shouldn’t we be getting it right the first time?  Isn’t that what peer review is for?  Sure, with time, maybe we will set the record straight.  But, for now, the damage has been done for LGBT people.  With so much that we have yet to study about LGBT families, it also warrants asking whether we should be worried about having to spend time, energy, and resource on redoing research.

Other Suggestions

Fabio also suggested:

  • [Realize] that that history is on our side. Increasingly, public opinion polls show greater and greater majorities favor LBGT equality. So if we are winning already, I wouldn’t go and ruin one of academia’s most valuable assets – blind review.
  • [R]elentlessly critique garbage and draw attention to the body of research.
  • I would engage the other side with sincerity and fervor. I would show people how to maintain the high ground.

In other words, don’t worry, keep blogging, and be the bigger person.  As gay people, my partner and I still cannot get married, not in the state in which we currently live nor the one we are moving to this summer.  I am pretty worried about the outcome of the Supreme Court case.  And, I am worried how easily this one study breezed through the peer review process, to publication, to press, to the courts.  Shouldn’t more sociologists be worried about this, too?  And, I am not sure what to say about maintaining “the high ground”.  It seems, for the oppressed, playing nice and playing by the rules does little to protect your rights being debated and denied on a daily basis — and my colleagues seem less concerned with my well-being as a human than with the well-being of science.

A Final Plea

“You don’t know what the heck you’re talking about!”  Exactly.  I am just days away from receiving my PhD, and have little experience publishing and providing reviews for journals compared to the sociologists at the fore of these debates.  What do I know?

That is a problem, in my opinion.  A systemic problem.  With a few research scandals going on these days, I am surprised that my colleagues and I are not in dialogue about science and research ethics.  In fact, all that I recall is one week in my research methods course devoted to ethics.  We read ASA’s code of ethics, Van Maanen’s (1983) “The Moral Fix: On the Ethics of Fieldwork,” Allen’s (1997) “Spies Like Us: When Sociologists Deceive their Subjects,” and Simonds’s (2001) “Talking with Strangers: A Researcher’s Tale.”  

I read Tearoom Trade for another course, though we did not discuss Humprhey’s unethical methods.  My knowledge of the Milgram experiment comes from a brief coverage of ethics in my undergraduate psychology and sociology methods courses.  And, much of my knowledge about eugenics, the Tuskegee experiments, and other exploitative practices on communities of color comes from my knowledge of Black history rather than science.

In speaking with other LGBT sociologists, I know that I am not alone in my anger, disappointment, and frustration — and, my ignorance about what I can do.  This is partly due to our relative lack of power, as a subfield in general (soc of sexualities) and as individuals (pre-tenure).  But, it is also due to our lack of access to memories of prior scandals of this sort.  For example, while I did read Richard Udry’s “Biological Limits of Gender Construction” (ASR 2000), and even Barbara Risman’s and otherscritical responses in a class, we never talked about the broader context.  What happened after the article and the responses were published?

seminar

Why don’t we talk about these types of events in our graduate courses?  Why does our training on research ethics only cover the stages leading to submitting an article for publication, ignoring ethical and professional practices that follow publication?  In general, I think we could benefit from a bit more reflection on science as an institution.  It would be nice (I would even say crucial) to discuss the contexts behind published articles and books.  A sociology of sociology, if you will.  Why are the authors in certain journals overwhelmingly women, while the top sociology journals are about two-thirds men authors, and the most male-dominated journals are on methods and mathematics?  Why are broken barriers in publishing somehow undermined as “affirmative action in publishing” or “trendy, but not really important” (yes, I have heard scholars say this).

If anything, I ask that we stop trying to make this scandal go away in hopes that history will stop repeating itself.  Just 12 years after the scandal surrounding Udry’s study, we are faced with a similar problem.  And, my generation of sociologists barely knows about it.  How can we learn from the mistakes of our discipline if we are not teaching new members about them — what happened and how we resolved it?  C’mon colleagues — we have got to do better, for the future of our discipline, but also for society as a whole.

UPDATE (05/02/13):  And, now we have an example of the potential impact Regnerus’s study can have outside of the courtroom: the everyday harassment of LGBT people.