The New York Times ran a piece last week bemoaning the lack of data about the frequency of transgender people in the general population. There have been a number of surveys with widely contradictory results. A survey from the Williams Institute seems accepted as the most authoritative word, giving the number at about 0.3%, or three per thousand. Before that survey, a figure from Johns Hopkins University was often quoted in the media of 1 in 20,000, or 0.05%, reportedly derived from the frequency of gender surgery. A study in New Zealand found as much as 1.2% of respondents transgender, with more than twice that many reporting they questioned their gender.
We shouldn’t find these discrepancies surprising. The transgender phenomenon is notoriously hard to pin down. A woman who has transitioned, taken female hormones, and had Gender Confirmation Surgery would clearly be classified as transgender. But how about the same woman, before she even began exploring her gender? When she just saw herself simply as an unhappy man?
The reality is that obtaining an accurate count of transgender people has a number of thorny challenges:
- Hard to get objective data – Survey experts know that subjective data is notoriously unreliable. When you ask people to report information about themselves, you face biases and inaccuracies. People differ in the way they evaluate their experiences. “Are you unhappy with your gender,” could mean to one person that they’ve experienced sex discrimination, while to another it may refer to crushing dysphoria. Mild dysphoria may register as a “no” on some people’s scale and a “yes” on others.
- The incidence of self-reported transgender responses is dependent on culture – More permissive cultures will have better access to information and less stigma, so people will be more likely both to recognize their symptoms and be comfortable exploring possibilities.
- Measuring transgender prevalence is not a priority – Obtaining accurate counts is expensive. It involves contacting or processing data from a large number of people. If we momentarily accept the 0.3% number, that means to find even 3 transgender people we need to examine information about 1,000 subjects, an expensive undertaking. Government is the primary source of research money, and governments by their nature shy away from controversy. Funding research on transgender people loses out to less controversial targets for research efforts.
- Hard to pin down exactly what is meant by being transgender – If you’re trying to include anyone with gender variant behavior, your wide net is going to bring in people who have one or two crossdressing experiences. Your number will be much smaller if you’re looking at people who sought transition-related health care. Many definitions of transgender defy accurate measurement because they deal with what goes on only in people’s heads. Other, more objectively measurably definitions, such as those receiving some sort of treatment, will miss people, resulting in undercounts.
The combination of these factors means that data on how many of us there are is surprisingly difficult to obtain. Asking how many people are transgender is sort of like asking how many people are athletic. Numbers will vary widely depending on how you define your terms. Are you asking how many people consider themselves athletic, how many people have actually participated in some sort of sport, how many display above average physical abilities, etc.? The answer depends heavily on exactly what you mean by “athletic” and how you ask the question.
This is not to say that obtaining accurate counts isn’t important. If we’re going to argue that the added cost to insurance companies of transition-related care is minimal, if we are going accurately to assess the cost, benefit, and impact of transgender rights laws, if we are to put effective policies in place to accommodate transgender people in the classroom, the workplace, and the community, these tasks become much harder if we don’t know how many people are involved.
The answer you’re looking for will be different depending on how you’re using the statistic. If you’re an insurance company trying to assess the cost of including transgender pharmaceutical benefits, the number of transgender people on hormones would be most important, while the number of people who occasionally crossdress would be of less interest. However, if you were a community center designing a program for people who think they might be trans, the number of people you’re interested in would be much larger.
What we really need is an effort by demographers to bridge the knowledge gap by quantifying all aspects of the transgender experience. The absolute number of transgender people is of limited usefulness, other than as a meaningless factoid to be quoted by mainstream news articles. Far more valuable would be a breakdown of how many people can be categorized in each of the many facets of the transgender experience — closeted, exploring, undergoing social transition, medically transitioning, obtaining surgery, and everywhere in between.