Share on facebook
Share on twitter
Share on email

Why psychology should be for “all” of us

Mental health diagnostic tests overlook gender diversity, but a new gender category would change that
Share on facebook
Share on twitter
Share on email

Across times and cultures, gendered bias has affected our design choices in many ways. In the past, many Western structures have prioritized a masculine norm: candidates for clinical trials were mostly or exclusively men, surveys of public opinion only consulted men, and athletic competitions excluded women. 

Despite the continuation of some of these structures, we’ve witnessed a positive shift away from this masculine norm: women have their own categories in sports, masculine clinical trials are at least now recognized as failing women, and the public sphere generally includes women and holds a vestige of support for trans and nonbinary people. Our methods of categorizing gender to better design competitions, public systems, and mental health treatment has evolved, and largely for the better. 

That said, I’m not convinced we’ve found the right method for gender inclusion yet, especially when it comes to mental health. Here’s why.

In 2018 and May 2021, I went to see a psychologist. As is common in the first few sessions, I took a few diagnostic tests — one of which was the Personality Assessment Inventory™ (PAI™). The PAI™ asks 344 questions and measures 11 important clinical constructs. The test asked me to select my age, ethnicity, and education, as well as indicate whether I was “male or female.” 

After that, I responded to the run-of-the-mill anxiety and depression screening questions, along with seemingly non-sequitur ones about matters of personal taste, such as sports preferences. The test then synthesized my answers and graphed them for my clinician, who described my results to me.

Anyone who takes this test will be confronted with a score that estimates how likely you are to have a mental illness like depression when compared against standards of mental health. These standards consider factors such as your age, ethnicity, education level, and gender — meaning that two people taking the test who differ in gender will be compared against different standards. The parent company of this test — PAR, Inc. — describes “specific normative data based on age and gender” as a key strength in their diagnostic accuracy. 

But I was shocked by my results. I was compared to both the general population and to postsecondary students — a population riddled with stress and mental health issues. What ranked as unhealthy behaviour for the general population appeared exemplary compared to my peers. Results that might have otherwise scraped the diagnosability line now disappeared from its sight, plummeting below the average. At the very least, I was thankful to see my general population norm group score and postsecondary student population score displayed side-by-side; otherwise there would be no way to know how bad I had it as a postsecondary student. 

But again, I was only being compared within my selected “male or female” category and I was blind to the norm of the group that had chosen the other category. This separation establishes different norms for differently gendered individuals, as it assumes different patterns of mental illness and resilience. 

The PAI™ isn’t alone in this. Historically, tests like The Behaviour Rating Inventory of Executive Function, the Millon Clinical Multiaxial Inventory, and the Schedule of Nonadaptive and Adaptive Personality (SNAP) use or have used normative gender or sex categories to differentiate between test-takers. 

However, for just as long as gender or sex-norm groups have been used, researchers have also debated whether relying on a unisex norm group would be better. The SNAP manual wrote that “the long-held assumption that gendered norms provide a more valid basis for assessment is being challenged.” That was in 1993.

How we display data and design gender and sex categories matters. This is because our data, and the methods we use to interpret them, impact how we structure our world to accurately reflect our population. If diagnostic tests for depression and anxiety continue to divide people by gender, then we will continue to believe that mental health conditions affect those people in different ways — when, in fact, anyone can experience them and share the same symptoms of someone gendered differently than they are. 

With that in mind, I propose that we should begin using the (un)gendered category ‘all.’ This category could serve as a unisex norm, which would enable us to reflect the full scale of diversity in gender and sex, and understand them as spectrums. 

The function of ‘all’

If we acknowledge that mental illness are not caused by inherent gender or sex differences — and especially if we acknowledge mental illness are socially shaped — it makes little sense not to compare men and women together. 

Regardless of whether we believe that differences between men’s brains and women’s brains are inherent or socialized, we know that our brains are indelibly shaped by environmental stressors. So, when we treat brains as ‘male’ or ‘female,’ we create differences between them. This becomes a self-sustaining, self-justifying system. In other words, the more we assess people and conduct research with differential gender or sex categorization at the outset, the more we entrench these normative gender or sex categories. 

Discussions started by feminists and men’s rights activists often cite society as the cause of a harmful mental state. The premise for their claims is about the same: victimhood on account of being assigned a gender without consent and being raised on a pre-existing social script to perform gender in a constricting way. 

For those who aren’t familiar with gendered socialization in its subtler forms, a 1999 article popularized by Caroline Paul’s 2016 TED Talk serves as a good case study. It found that toddlers were taught very differently about risk depending on their gender. 

Researchers observed parental supervision of playtime on a “fire-house type pole” and a “free-play episode.” In these episodes, they found that parents gave sons “more directives, fewer explanations, and more questions communicating information about how to perform the pole task than daughters.”

They also found that parents provided more “spontaneous physical assistance” to their daughters than their sons, even though the report found that “there were no sex differences in children’s playground skills or their abilities to complete the task independently.” Given the fact that prepubescent children exhibit little to no sex-related strength differences, these results suggest an unfair bias in the way individuals are socialized throughout childhood just because of their gender.

If our goal is to create a society where no one suffers undue hardship because of gender or sex, then we must establish where socialization ends and nature begins. Only by doing so can we understand whether or not our gendered categories are justified, and to what extent they are useful. 

Most importantly, we would have a clearer picture of the degree to which mental illnesses are exacerbated and encouraged by gender roles, and where the average ought to be for everyone, regardless of sex or gender. 

Using ‘all’ as a gender category helps us in this endeavour, as the category reveals how much our behaviours are influenced by gender norms. ‘All’ is the best way to categorize how people feel when they allow themselves to be, without punishing gender deviancy.   

The efficacy of unisex scoring

Some are also concerned that women and men express disorders and neurodivergence too differently to be scored by the same metric.

While it is true that many of the diagnostic criteria for ADHD and autism are biased against women, that bias primarily exists because, historically, the parameters for diagnosis were modelled after primarily male clinical groups. In other words, the shortcomings of these criteria to accurately diagnose both men and women stem from men-biased metrics, not from the use of a true ‘all’ category. 

ADHD and autism don’t look very different across gender categories, either. Granted, women with autism tend to ‘mask’ more — that is, to hide autistic traits and behaviours to better fit in with neurotypical people. But, even this likely has more to do with gendered socialization — specifically the idealization of ‘quiet and shy women’ — rather than inherent sex or gender differences. 

Moreover, studies conducted on gender and sex bias have found that women were more likely to be misdiagnosed with borderline, dependent, and histrionic personality disorders when using gender-normed scores. It’s important to note that the word ‘histrionic’ itself comes from a word with sexist origins. On the other hand, when unisex norms were applied, they were less likely to be misdiagnosed. 

Ultimately, claims that an ‘all’ category would lead to misdiagnosis disregard the fact that diagnostic tests are not the be-all-end-all of assessment. Such claims ignore the client interviews, observer responses, and triangulations of scores between multiple diagnostic tests, which are all integral to the diagnosis process. 

Some still argue that an ‘all’ category is unfair or will lead to misdiagnosis because it doesn’t account for the real, biological differences between men and women. These detractors believe the categories are inherently incomparable. But I can make the same argument about comparing individuals within the same binary sex and gender category and the widespread misdiagnoses that causes.

For example, let’s take a 2015 article in the Current Behavioural Neuroscience Reports titled “Is Impulsivity a Male Rather Than A Female Trait?” Immediately, the title promises the reader a blanket conclusion, but upon closer evaluation, the study does not provide any evidence for any blanket statements. Rather, the researchers found that candidates who identified as ‘female’ were on average less impulsive than ‘males’ during the fertile stages of their menstrual cycle. Meanwhile, adolescent ‘females’ exhibited more risky and impulsive behaviours than adolescent ‘males.’ 

From this, you may come away with the idea that women are less impulsive than men. However, you would be assuming that all women have active menstrual cycles — when, in fact, even among cisgender women, many do not. If the average age of menopause is 51, and the expected age of menstruation onset is 11 to 14, then a significant number of women must be menopausal or premenstrual. 

According to the World Bank, only 42.04 per cent of the world’s women population — the term used in the World Bank report — is between 20–49, the ages at which we can confidently expect women to be menstruating.

Hence, the study immediately neglects more than half of the cis women population — and that doesn’t even account for trans women, who may experience estrogen hormone cycles but no menstruation, trans men and nonbinary people who still menstruate, and cis women who use birth control that affects their cycle. 

The reality of sex and gender diversity

There is so much variation within binary categories. While there are real reasons that we use ‘male’ and ‘female’ as categories in many concrete medical contexts, the combined sex and gender experiences of the majority of people don’t fall uniformly into either category — which is why it is necessary for us to account for gender diversity if we want to better understand and reflect our population.

If binary gender categories aren’t even accurate for cis people, then it’s not hard to see how they could be even less accurate if you’re not cis. Trans and nonbinary people become factored into gender binary categories as outliers rather than reliable members of any particular group. I experienced this myself in these tests. 

After I took my diagnostic test, I received an email titled: “Apologies — no non-binary option on normative scales so your scores are compared to a [binary gender] norm group.”

It’s important to have trans-norm groups to understand the common stressors within this community. However, once these tests establish the normative baseline for mental health among trans people — a group that suffers from disproportionate amounts of mental illnesses — we have to ask what measurement of mental health we should be aiming for. 

Including ‘all’ as a category promises to be beneficial even when applied beyond the realm of diagnostic psychology from the treatment of mental health disorders to the organization of sports to every other corner of life. My vision for ‘all’ as a gendered category is that it will become as legitimate and universal as ‘male’ or ‘female’ or ‘man’ or ‘woman’ in every context; anytime there is a discussion of gender or sex, ‘all’ should be included.

© 2020 Varsity Publications Inc — All rights reserved