The Newest Gender Identity Scandal
On at least a hundred different occasions, I've been asked whether I'm related to Jerry Springer. I found this amusing – the first 10 or 15 times.
Once, a cashier insisted that I must be Jerry's brother. I was annoyed, but not deeply. It barely registered.
Though we all experience mistaken identity from time to time, it's hard to imagine the distress of transgender and nonbinary people, who've spent years being viewed by everyone (not just random cashiers) in a way that clashes with how they view themselves.
In this newsletter I'll be discussing a recent controversy over the nature of the trans experience.
This a colorful, albeit painful story. First, a research article sparks public outcry. The authors, some say, are deeply biased against the trans community. The publisher launches an investigation and, yesterday, retracts the article. During the investigation, others proclaim that the journal is censoring free speech.
Woven into this story are illustrations of some of the most common ways that statistics are misused.
The controversy
On March 29, the Archives of Sexual Behavior published an article on the experiences of children and young adults with gender dysphoria.
"Gender dysphoria" refers to the distress that a person feels when the gender assigned to them (typically at birth) doesn't match their gender identity. Geena is born and raised as a girl, for instance, but comes to view herself as male.
People with gender dysphoria often identify as transgender, but there are exceptions. (Geena might consider himself male but suffer in silence. Or, he might feel that he has no gender, or multiple genders, or that he embodies a third gender.) At the same time, nearly all transgender people describe experiences with gender dysphoria.
What made the article controversial is its claim that gender dysphoria isn't real. Individuals who report it simply have mental health problems. Peer feedback causes them to decide that the root of their problems is having been raised as the wrong gender.
As you can imagine, public backlash was swift. Following receipt of an open letter of concern on May 5, Archives of Sexual Behavior added a note on May 10 that the publisher was investigating methodological concerns about the study. On May 16, Archives removed a portion of this article, and yesterday, the journal formally retracted it.
While the publisher's investigation was underway, another open letter, signed by over 2,000 people, argued that the the journal acted inappropriately in response to public pressure and has begun to censor free speech.
At this point you might be wondering where I stand on this controversy. The title of this newsletter refers to a scandal. Who do I consider scandalous?
–Is it the allegedly transphobic authors of the study, who used pseudo-science to claim that trans people are mentally disturbed and not actually trans?
–Or is it the allegedly uber-woke journal that caved to public pressure and is censoring scientific expression?
Well, do I have to choose?
Jerry Springer would not have chosen. He would've let representatives of each side duke it out (perhaps literally). Indeed, on Twitter and in blogs, the ranting has already begun. Now that the paper is retracted, you can expect to hear more. (Conservative news outlets such as the Wall Street Journal and National Review have covered this story, but the retraction will surely prompt commentary from other perspectives.)
The conversation thus far has been not helpful. It does nothing to alleviate the distress of young people who experience gender dysphoria. It also does nothing to advance current debates over the rights of transgender people. There's a lot at stake. For example, you might believe that trans youth should have access to gender-affirming surgery, or you might believe that such options should be banned (as they are now, or will be soon, in many states). Either way, you probably agree that that the issues are vital.
I'll be discussing the authors' biases as well as some problematic moves on the journal's part. But for me, the bigger scandal is that people on both sides of the controversy have begun to scream rather than considering the data. This is the Jerry Springer approach, and it's not good for anyone.
I'm biased of course, but I prefer the Ken Springer approach to what Jerry Springers do. I believe that data is always useful. You have to decide whether or not to trust it, and if you do, you have to decide what it tells you.
At the end, I will point out that the Ken Springer approach (obviously I didn't invent it) can be problematic too. Impartiality represents an ideological stance too, and it has weaknesses.
"Rapid Onset Gender Dysphoria"
Experts agree that the number of teens and young adults in North America and Europe expressing gender dysphoria has spiked during the past decade.
The most common explanation for this trend is increasing acceptance of non-traditional gender identities. In other words, the number of people with gender dysphoria hasn't changed. It's just more socially acceptable now to talk about it openly, seek support, and express one's identity as trans, nonbinary, etc.
A less commonly held view is that gender dysphoria is actually becoming more prevalent. A small number of academics, clinicians, and parents subscribe to a highly controversial version of this idea known as Rapid Onset Gender Dysphoria (ROGD) theory, which was first proposed five years ago.
ROGD theory holds that young people with mental health issues are led to believe they have gender dysphoria through exposure to others who experience it, or at least talk about it supportively, via social media as well as in-person interactions. In the Appendix I provide more details on this theory.
A simple distinction can thus be made between people who recognize that gender dysphoria is an authentic experience, versus ROGD proponents who think that gender dysphoria merely reflects social influence. To keep things simple, I'll refer to these as the two sides of the controversy.
Again, there's a lot at stake, because ROGD proponents seem to universally support what others would call anti-trans attitudes and policies, such as banning gender-affirming care to minors.
The new study
The newly retracted study, which appeared on March 29 in the respected peer-reviewed journal Archives of Sexual Behavior, was authored by Suzanna Diaz, the leader of an organization called Parents of ROGD Kids, and Dr. J. Michael Bailey, a professor of psychology at Northwestern University and well-known figure in research on sexual orientation. As Dr. Bailey explains, he heard Ms. Diaz describing a Parents of ROGD Kids survey at a conference and invited to help her publish the findings.
For this study, surveys were administered to 1,655 parents who believed that their adolescent or young adult children (age range 11 to 21) have ROGD. Survey questions primarily concerned the children's experiences. In other words, it's a study about the children; parents were merely the source of data.
The authors are quite clear that they weren't studying a representative sample of young people with gender dysphoria. Survey participants already assumed that their children had been victimized by a social construct. This is evident from the recruitment email sent to parents, which begins as follows:
As I'll explain later, one of the journal's earliest responses to public backlash was to retract the Supplemental Materials section, which includes the survey. Currently that survey is not publicly available. I queried Dr. Bailey and he kindly emailed it to me.
Main findings
Many of the study findings pertain to social transitioning, which the survey defines as follows:
Parents were asked when their children first experienced mental health issues, when they transitioned socially, and how the transition affected them. A key finding was that "mental health deteriorated considerably after social transition."
The data provide no support for this conclusion.
The first problem here is that the conclusion is timeless and simplistic. Jay transitions, Jay's mental health deteriorates. End of story.
As the name suggests, transitioning is part of an extended process. Mental health changes over time, and following transition all of those changes count, whether they occur during the early weeks and months, the early years, or decades down the road.
No doubt some of those changes are negative. Social transitioning is by definition a public act. The first steps are fraught with uncertainties, rejection from others (actual as well as anticipated), and new challenges. But there are victories too. Do the positives outweigh the negatives? If so, when? What do trans people say about their experiences at different points in time? Do they arrive at a settled perspective? The authors don't address these issues.
In fact, they were unable to address them, because the survey questions ask parents how their children were doing "after" transitioning. The term is never defined. Parents could be thinking of the day after or the entire year after (assuming parents thought they could identify a specific point in time when transitioning began).
A second problem is that the parents sampled are among the least qualified people on the planet to give an accurate account of the connection between transitioning and their children's mental health.
Remember that these parents believe their children are victims of a false narrative. They desperately wish that their children would just come to their senses and de-transition. Mary's dad just wants Mary to be his girl again. Naturally, parents such as these would be inclined toward a negative view of transitioning. Since the questions merely referenced the time period "after" that process, a parent could easily focus on the most difficult times.
Finally, there's a statistical problem. Parents were asked directly whether their child had ever experienced a "mental health issue". Separately, they were asked whether their child had ever received a formal diagnosis of a "mental health issue". For each question, 0 points were given for a "no" response, and 1 point was given for a "yes" response. The authors then added the two scores together. Why? Here's what they claimed:
"the summed composite should have greater reliability than either of its component items. We refer to the composite as “mental health issues,” and higher scores indicated more problems with mental health."
The statement about reliability is false. The composite variable makes little sense. And, a higher score does not necessarily indicate more problems. Simple example (based on an actual case): Bree is a transgender boy whose mother thinks that he has mental health issues, simply because Bree is transgender. During middle school and high school, Bree's mother takes him to psychologist after psychologist until, on the sixth try, she finds one who provides a diagnosis. In this study, Bree would receive the highest possible score in spite of probably meriting the lowest one.
In sum, there are no grounds for drawing the simplistic conclusion that children's mental health "deteriorated considerably" after transitioning.
Most of the findings and interpretations in this study reveal bias. I want to mention just one more, because it's especially egregious. I'll let the authors speak for themselves here:
"Youths with a history of mental health issues were especially likely to have taken steps to socially and medically transition...The finding is concerning because youth with mental health issues may be especially likely to lack judgment necessary to make these important, and in the case of medical transition permanent, decisions. "
Here you see the interface between statistics and deeply ingrained bias.
As I mentioned earlier, the operational definition of "mental health issues" is statistically flawed, so nothing can be concluded here about the influence of these issues. (If you haven't measured the weight accurately, you can't guess how difficult it will be to lift.)
Even so, the second part of the quotation is revealing. The authors are saying, in effect, that children were mentally impaired to the point of making an unsuitable decision.
The most common mental health symptoms the children had experienced were anxiety and depression, according to their parents. Although these symptoms can impair judgment about anything when severe enough, the survey didn't ask about severity.
It's condescending and false to assume that experiences such anxiety and depression might cause someone to lack judgment about whether to transition. Tim was born a boy, Tim feels like a girl, and now Tim decides to change pronouns and ask others to treat her like a girl. Why would Tim have been lacking in judgment owing to, say, anxiety about the prospect of beginning high school?
I'm not even taking much of an ideological stance here. Even if you have an anti-trans perspective and you think Tim made the wrong decision, why would you assume some degree of anxiety or depression impaired Tim's judgment?
Here's a completely opposite yet at least as valid way to interpret the data: The fact that youths with mental health issues were especially likely to have taken steps to transition is a cause for celebration. It's a good thing, because it shows they were actively trying to help themselves.
Earlier I suggested that we shouldn't trust how the authors linked mental health issues to transitioning. My point here is simply that even if you accept that association, you can think of it as a positive thing rather than as a source of concern.
I emailed Dr. Bailey about this last week, asking whether one could argue that steps toward transitioning represent proactive, problem-focused coping strategies. Here is his reply:
"I suppose that’s another possible interpretation, although I doubt it’s correct. (Parents believed that mental health issues preceded gender-related issues by about 4 years, on average.) We definitely need better data, and I will be collecting some."
I appreciate his acknowledgment that we need better data. He was not responsible for the survey design, and he and his co-author were quite open about potential biases in their study. But I'm not persuaded by his response.
Even if parents were accurate about timing, a 4-year delay between mental health issues and transitioning does not rule out the possibility that transitioning is a pro-active way of dealing with those issues. After all, if you feel depressed, it may take time to realize that your assigned gender is an issue, then try out different strategies, then decide transitioning might be needed, and then actually make plans and carry them out. Meanwhile, you've got schoolwork, family obligations, social activities, and the rest of your life to contend with. Besides, younger kids face more barriers to transitioning than older kids do.
(Do parents even know when their children began to transition? Social transitioning is not an all-or-nothing event. Perhaps the kids were "out" with their friends and social media communities long before their parents realized.)
The coups de grâce
If you have any lingering doubts about whether the researchers were ideologically biased and used wretched data to advance an ideological agenda, here are two questions for you:
1. Why was this study conducted?
The Introduction to the study is coherently written. A number of important issues are covered. But it does not tell you exactly what the study intends to contribute to the literature, over and above reconfirming a few things we already know. For example, the authors want to know whether social transitioning is more common than medical transitioning. They found that it is. (How loud can you say the word "duh"? Of course changes in pronouns and hairstyle are more common than medical interventions, given how time-consuming, expensive, painful, and permanent the latter tend to be. Not to mention the fact that if you're a minor, medical interventions might not be a legal option for you.)
Explaining why a study is conducted – i.e., what it hopes to contribute to the scientific literature – is one of the fundamentals that doctoral students learn and most journals require.
All I can glean (and this is mostly from the Discussion) is that the authors would like to persuade us that ROGD exists, and that it's harmful to children and young adults.
In fact, the 2018 study that first proposed the existence of ROGD, like this one, was grounded in an extremely biased, non-representative sample of parents. The one study in which children themselves were the source of the data provides evidence against the existence of ROGD. (See Appendix for details.)
2. Was there harmful intent?
Perhaps this will invite a lawsuit, but yes, I see evidence of harmful intent. Specifically, the survey asks parents to provide the names and addresses of the clinics, as well as the individual health care providers, clinicians, and other specialists who offered their children psychological support, hormone therapy, and/or surgery.
I can understand why the survey asks, for example, whether the gender specialist who spoke with the child was a doctor versus a social worker. But to ask for the names and addresses of specific clinics and people? This could serve no useful purpose unless the intent is to attack or otherwise expose those clinics and people. This feels like a witch hunt.
Journal response
Following public backlash, Archives of Sexual Behavior added a note on May 10 that the publisher was investigating methodological concerns about the study. (The publisher happens to be Springer Nature, but, as with Jerry, I have no family ties that I'm aware of.) On May 16, Archives removed the Supplemental Materials from the article, and yesterday, the journal formally retracted the entire thing.
I accused the study authors of being profoundly biased because there's only two of them. They're responsible for their work.
As for Archives, I can't accuse or defend the journal from bias, because it's unclear how their decisions were made. According to the May 10 note, Springer (the publisher) initiated the investigation, but it's unclear who at Springer made the decisions and to what extent the Archives editorial leadership was involved.
In any case, the journal handled this situation badly. Among other things, it does seem to have caved inappropriately to public pressure.
Yesterday, Archives announced that the paper is officially retracted, because informed consent had not been obtained from participants.
What happened is that Susanna Diaz, the leader of Parents of ROGD Kids, had conducted an informal survey. Dr. Bailey learned about the survey at a conference, and offered to help publish the findings in a peer-reviewed journal.
Anyone can conduct a survey. Ms. Diaz did not need permission from any source in order to do so. However, according to federal law, survey findings cannot be treated as "research" (e.g., published in a scientific journal) unless the researchers first obtain permission from something called an Institutional Review Board.
I was the Institutional Review Board chair at my university for nine years, as well as a committee member for three years prior to that. I'll try not to geek out here on the details, but I make no promises.
A federal code governs the IRB decision-making process. Generally speaking, the IRB cannot approve a survey study unless informed consent is obtained from participants. People need a rough idea, in advance, about things like the purpose of the survey, the topics covered, the extent to which responses will be kept confidential, and so on.
Federal law also allows for exceptions. For instance, data that has already been collected can be published if the researcher obtains it in aggregate form and no individual participant could be identified. In such cases, IRB review may not be required.
Springer (the publisher) has a policy stating that journals such as Archives can makes case-by-case decisions about whether or not to publish data such as that.
Here's what the journal did wrong:
1. It's clear from the article that the data was collected without IRB consent. It's impossible to miss this – e.g., the authors state that the survey was administered to a community of concerned parents, and that scientific publication was not originally planned.
–If the journal editor felt that informed consent (and IRB approval) was necessary, the paper should not have been published in the first place.
–If the journal editor felt that informed consent (and IRB approval) was not necessary, the paper should not have been retracted.
2. On May 16, the journal retracted the Supplemental Materials section "due to the lack of documented consent from study participants".
–The Supplemental Materials section included the survey itself. There was no reason to retract that, as it's simply a list of questions.
–At the same time, on May 16, the journal failed to retract Table 1. This table is now unavailable, but I have access to it in the original version of the paper I downloaded. The table contains statements made by individual study participants. Highly personal statements, in some cases.
If the purpose of removing the Supplemental Materials section was to protect study participants, the journal should've removed Table 1 too. This was a shocking oversight.
The broader issue
Given how shoddily Archives handled the retraction process, it's understandable that some observers question their motives.
Specifically, people unhappy with the retraction process, including the 2,000+ signees of an open letter, suspect that Archives caved to public pressure and used the informed consent issue as a smokescreen.
Perhaps they're right. It certainly looks that way. But, again, we can't be sure, because it's unclear who made the decisions and what they were thinking.
Meanwhile, although people who signed that letter see a case of scientific censorship carried out in fealty to political correctness, many of those individuals happen to be prominent anti-trans activists. (Jordan Peterson is one of the main signatories, for instance.) So, what exactly are their motives? In some cases less than pure, I suspect.
The whole incident feels very much like a Jerry Springer episode.
Conclusion
I've tried to remain impartial, but it's abundantly clear that gender dysphoria is a genuine experience. Trans and nonbinary people are not simply deluded about what they're feeling.
I've argued that we always need data, and it may seem like I'm leading to the conclusion that Archives should not have retracted the study. Rather, the study should've been published, so that we can all see what a piece of crap it is, methodologically speaking, and why the interpretations are deeply biased.
In other words, it may seem like I'm one of those liberals who holds that the best way of dealing with bigoted people is to let them speak up, so that their biases can be exposed.
In some situations I feel that way. But a mainstream, peer-reviewed journal gives authors instant and sometimes dangerous credibility. I'm not so sure this study should've been published. Here's why:
The main purpose of an IRB is to protect the well-being of research participants, and one principle of IRB review is that a study should not be approved unless the potential benefits outweigh the risks.
As an IRB chair, I would've asked my committee not to approve this study, because the methodology is so awful it's hard to imagine potential benefits. But there are risks, particularly for how the findings might impact people with gender dysphoria.
In light of those risks, as a reviewer or journal editor I would not have considered the study publishable without IRB approval.
In effect, I'm suggesting that ethical considerations should've blocked all avenues to publication.
Let me bring in a different example for a moment.
Social scientists have a long and sordid history of racial/ethnic group comparisons on variables such as moral character. This kind of research can no longer be conducted – though fringe theories still percolate – but in the past, liberal academics would sometimes say, oh, go ahead, do the research. Of course racial/ethnic groups won't differ on something like that. No need to fear the data.
We do need to fear data when the methodology is bad, or the statistics are misapplied, and the results end up in a mainstream journal.
Comparing racial/ethnic groups on moral character is sort of like asking whether people with "innie" belly buttons differ in moral character from people with "outies". There's no scientific justification for even asking the question. You only run the risk of finding statistical differences that would be misleading.
Analogously, why conduct a study premised on the assumption that gender dysphoria isn't genuine? There's no good evidence supporting that assumption. Why raise the possibility of that transitioning is harmful, in some cartoonishly simple sense, when there's no reason to suspect that? You only risk finding effects that would stigmatize trans people, or lead others to question the authenticity of their experiences.
In sum, freedom of expression is as precious in academia as anywhere else, but not all research questions are equally justifiable.
Thanks for reading!
Appendix: ROGD and ROGD theory
ROGD theory, now five years old, would not be controversial if it merely claimed that young people with gender dysphoria experience it for many years, suffer in silence, then discover others with similar experiences, thereby gaining a supportive community as well as language to express how they feel.
What makes ROGD theory controversial is that it denies that these young people were experiencing gender dysphoria in the first place. According to the theory, they were experiencing stress, mental health issues, etc., and when they found communities of people discussing gender dysphoria, they quickly came to believe that they themselves had been experiencing it too. The underlying mechanism is a form of social influence called "social contagion."
It's obvious why ROGD theory is controversial. Ultimately it denies that gender dysphoria is what it is. Proponents of the theory hold that a transgender person is actually just a person with mental health issues who has been misled by some community to believe that they're transgender.
To be clear, ROGD does not currently exist apart from ROGD theory.
Gender dysphoria is both a formal diagnosis (as laid out in the DSM-V) as well as an informal term used to sum up various kinds of distress arising from the mismatch between assigned gender and how one personally feels. Simply put, "gender dysphoria" is a broad and ideologically neutral term used to describe a kind of distress.
In contrast, "ROGD" is neither broad nor ideologically neutral. The "RO", or rapid onset, is meant to convey the idea that some people who experience mental health issues don't consider their problems to be particularly gender-oriented, but then, suddenly, a community of people persuades them that they have gender dysphoria.
The retracted study discussed in this newsletter is too weak, methodologically speaking, to constitute evidence for (or against) ROGD. The original 2018 studypositing its existence only claimed that it represents one pathway in the development of gender dysphoria, a nuance that has been lost on some proponents of ROGD theory. In any case, both the retracted study and the original study relied on reports from clearly biased parents. Since then, at least one newer study, from 2022, examining youth themselves, failed to find evidence for a rapid onset form of gender dysphoria.