How many trans* people are there? A 2011 update incorporating new data.
Sam Winter, University of Hong
Lynn Conway, University of Michigan
Contact authors at [email protected] or [email protected]
Minorities, it has been said, don’t count until they are counted. If so then for this reason alone this question is an important one, not just for healthcare providers (who need to plan and resource services for transpeople) but also for community-based organisations concerned with broader social issues. And also, of course, for professional organisations like WPATH.
The question has a short and a long answer. The short answer is that we just don’t know how many transpeople there are, but we know it is many more than the most commonly quoted statistics imply. The long answer demands a more detailed consideration of those statistics.
Estimating numbers of transpeople is difficult. For one thing, many transpeople try to keep their trans status private; and may not be easily counted. For another, transpeople represents a broad spectrum of identity and expression. For these reasons, the figures you get on prevalence depend on who you decide to count, and how you decide to count them. What follows is a brief summary of some of the research that has been done. It does not pretend to be an exhaustive review.
Clinic-based studies of transsexuals
For various reasons prevalence researchers have tended to focus on the most easily counted sub-group of transpeople; those whose experience distress or discomfort regarding their bodies, and approach specialist gender clinics seeking gender transition counselling and healthcare.
De Cuypere et al (2007) provide a useful review of prevalence studies, as well as conducting their own. Together, these studies span 39 years. Leaving aside the one early outlier finding (Pauly, 1968, who reported a figure of 1:100,000 for transwomen and 1:400,000 for transmen, a prevalence far lower than anyone else), the Western studies (ten studies from eight countries, if one includes De Cuypere et al’s own study) yield a range from 1:11,900 to 1:45,000 for transwomen and 1:30,400 to 1:200,000 for transmen. What is interesting about these studies is that the more recent they are the higher the prevalence rates that tend to get reported. It’s worth doing a little statistical analysis here. For transwomen the correlation between prevalence rate and year of study is -0.56. For transmen it is -0.67. The downward trend is impressive. With each year that passes (at least in the developed West) it seems one has to search 629 fewer people in the general population to find a transwoman who has accessed a gender clinic, and 2591 fewer people to find a transman.
The impression that more transsexuals are attending clinic is confirmed in two recent research reports. Reed et al (2009) have examined the numbers of transpeople accessing UK clinics, and report a doubling every five or six years, including among children and adolescents. Zucker et al (2008) report what amounts to a four to fivefold increase in child and adolescent referrals to their clinic over a thirty year period (indeed, for children the increase has been around three-fold in just twelve years between 1996-2007 inclusive).
Enough of the West. What about elsewhere? Some of the clinic-based figures in the Global Southeast suggest rates as high as the more recent Western research, or even higher. In Singapore, Tsoi (1988) reported a figure of 1:2900 for transwomen, 1:8,300 for transmen. In Iran, the Director of Socially Vulnerable Groups at the State Agency for National Well-Being, who one can only imagine is relying on clinic-based figures , recently cited figures indicating a prevalence of 1:13,000 (Alizadeh, 2010).
But clinic-based figures (even the most current ones, from wherever in the world they are drawn) grossly underestimate the size of the broader transgender population. They fail to take account of ( a ) people who, though anatomically dysphoric and seeking healthcare, do not approach the clinics tapped by the statistics (perhaps seeking surgery other parts of the world); ( b ) those anatomically dysphoric transpeople who do not attend clinics at all, perhaps modifying their bodies by way of self-administered hormones (and accessing non-specialist healthcare if any at all); or ( c ) those who are socially dysphoric only, and perhaps choose not to modify their bodies at all; ( d ) those who, though gender dysphoric, live (or feel a pressure to live) cisgender lives, hoping their gender issues will abate, and possibly entering fringe therapies (at their own or others’ instigation) designed to rid them of those gender issues; or ( e) those who experience no gender dysphoria at all, but are simply cisgender people who, for whatever reason, express their gender in ways that go against or beyond gender stereotype (cisgender cross-dressers for example).
All these transgender sub-groups (and the five categories listed can only hint at the diversity of identity and expression in the trans population) add up to large numbers. Reed, drawing on his work in the UK, has suggested “for every gender variant person who has sought medical help, there may be 50 who have not yet done so”. Others, for whatever reason, never do so.
The failings of clinic-based statistics apply to children too: statistics on children attending gender clinic ignore ( a ) most of those who have already learned to tone down or conceal their gender variance, and ( b ) those whose gender variance does not cause alarm to their parents and caregivers. Neither group tends to end up at clinic for gender problems (although there is good reason to suspect that many in the first group may end up attending clinic with other problems such as depression)
Looking at transgender people more broadly.
Not surprisingly, studies which employ more inclusive methodologies (focusing not only on those who attend clinics but on other transpeople too) tend to yield higher prevalence figures. Let’s look first of all at some sample studies of childhood gender variance. There is space for three examples here, each from a different place and in a different decade. All involved items drawn from the CBCL (Child Behaviour Check List) or based upon it. Achenbach and Edelbrock (1981) noted that, among 4 to 5-year-olds in the USA, 6.0% (around 1:17) of boys and 11.8% (around 1:8) of girls were reported by their parents to behave ‘like the opposite sex’. Fifteen years later in the Netherlands Verhulst et al (1996, cited in Cohen-Kettenis and Pfafflin, 2003) reported 2.6% (1:38) of boys and 5.0% (1:20) of girls aged 4 to 11 behaved ‘like the opposite sex’, with corresponding figures of 1.4% (1:71) and 2.0% (1:50) for those wishing ‘to be opposite sex’. Finally, in a ground-breaking study in mainland China Yu (2009) notes that parents report that 0.5% (1:200) of their 6 to 12-year boys often or always ‘state the wish to be the other gender’, with a corresponding figure of 0.6% (1:167) for girls.
The Achenbach and Edelbrock (1981) study referred to above reported prevalence rates substantially lower for 12-13 year-olds than for 4-5 year-olds, both in regard to cross-gender behaviour and cross-gender identity. It is pretty clear from this and other studies that the prevalence of gender variant behaviour reduces as children grow older. Nobody knows why this happens. It is likely that as children grow older they become better (and perhaps more harshly) gender-socialised. In many cultures this may be particularly true for boys. Notwithstanding, it is clear that by adulthood substantial numbers report substantial discomfort or distress with their assigned gender. It is at this point that we turn to some of the adult studies that attempt to step out of the clinic.
Veale (2008), in a study of applications to change sex markers on New Zealand passports, has reported a transsexualism prevalence rate of 1:6364 in that country. This is a prevalence rate far higher than any of the figures yielded by clinic-based studies.
Even higher figures are evident when researchers simply ask people how they identify, or attempt to measure gender dysphoria. Consider three recent studies. In the USA Gates (2011) has recently reported that 0.3% (1:333) of adults may identify as transgender. Across the globe in Taiwan, Lai et al (2010) have reported that, among university students, 2.2% (1:45) of birth assigned females (and 0.7% (1:143) of birth assigned males) ‘very often’ wished they were the other sex.
Three very recent studies (still in preparation) support the view that there are transpeople than is commonly thought. In the Philippines, Cruz and Rogando-Sasot (in preparation) report that around 1:357 of birth-assigned males aged 15-24 desire to be female, and do so because they feel that they are female. In Mainland China Chi (in preparation) has found that, among 18-24 old university students, 0.8% (1.125) of those who are birth-assigned males (whose sex/gender as indicated on their ID card is male) report that the ‘sex/gender I feel in my heart’ is male (with another 0.4% indicating that their perceived gender was ‘other’). Among birth-assigned females, 2.9% (1:34) indicated they perceived their gender as male, with another 1.3% indicating ‘other’.
Finally, in an international study spanning university undergraduates in seven societies (USA, UK, Hong Kong, Malaysia, Philippines, Singapore and Thailand ) Winter et al (in preparation) have found that 0.9% (1:111) and 2.4% (1:42) of birth-assigned males and females respectively reported that they ‘all the time’ wished they had been born as the other sex.
Many adolescents and adults clearly go beyond wishing. This is nowhere so obvious as some of the countries of Southeast and East Asia, where apparently large numbers self-identify and express gender in ways that cross or go beyond gender boundaries. Depending on their geographical location, personal circumstances and individual needs, many do not seek professional gender transition healthcare. The vast majority identify as a gender other than that assigned (with their assigned sex) to them at birth. Based on small -scale empirical work in Thailand, Winter (2002) has estimated that around 0.6% (1.167) of birth-assigned males there may be transwomen (often called ‘kathoey’ in Thailand, but nowadays increasingly called ‘phuying kham phet’). In Malaysia, it is estimated that ‘mak nyah’ transwomen range in numbers from 10,000 (cited by Teh,2002) to 50,000 (Jamaludin, 2001). Based on a population of around 10 million birth assigned males aged 15+ (Nationmaster website data), that would yield a prevalence rate of between 1:1000 to 1:200. In India, conservative estimates for ‘hijra’ number them between 500,000 (Nanda, 2000), and 1,000,000 (Sitapati, 2009) . Based on a population of around 414 million birth assigned males aged 15+ (Nationmaster), the prevalence rate would be around 1:818 to 1:414. In Pakistan one estimate, again conservative, numbers ‘hijra’ at 400,000 (Curran, 2009). Based on a population of around 57 million birth assigned males aged 15+ (Nationmaster), the prevalence rate would be around 1:143. Interestingly, the latest Indian census (along with one in neighbouring Nepal) will allow citizens to identify themselves as 'third gender'. Pakistan has plans to allow ‘third sex’ individuals to register with local authorities. It'll be interesting to see what numbers these three countries come up with.
Finally, it is worth noting indications that in some locations, transpeople may be even more prevalent than in the research we have been discussing. Wikan (1991), engaged in anthropological research in Oman, reported that around 1:60 (around 1.7%) of the birth-assigned males in the town of Sohar were transgender ‘Xanith’. A newspaper report by Taylor (2011) lists 50 ‘sistagirls’ in the indigenous population of 2500 on the Tiwi Islands (a figure that implies, based on an estimate of 1250 birth-assigned males of all ages) a prevalence rate of 4% (!:25)!
Seen in the light of all this, Olyslager and Conway’s (2007) suggestion that ‘the lower bound on the prevalence of transsexualism is at least 1:500 (0.2%), and possibly higher’, at the time seen as a radical estimate, begins to look very credible indeed. Indeed, a figure such as the one provided by Gates (around 1:300) may be closer to reality.
A concluding thought
Let’s end with a thought experiment, using the recent Gates (2011) research indicating that 0.3% (1:333) of US adults may identify as transgender. This figure, quite similar to the earlier Conway and Olyslager research, looks very much like a ‘Goldilocks’ statistic, neither too high nor too low, well within the wide range we earlier looked at. So it is worth pausing for a moment, setting aside the anxieties we rightly have about generalising figures from one nation to others, and spending a little time thinking about how many transpeople that would imply worldwide. With 5.1 billion people aged 15+ worldwide (US Census Bureau, estimates for mid 2011) that implies 15,324,000 transpeople on this planet, many more than the (still) commonly cited clinic-based estimates would suggest, and around equal to the populations of countries like Kazakhstan, Equador and Cambodia (ranked 64th to 66th out of 196 countries for population). Viewing the numbers in this way puts a new perspective on things. It looks like there are a lot of transpeople out there. True, quite a few have a need for gender transition healthcare. But others don’t. The vast majority suffer life-long discrimination, sometimes with a cumulative effect on physical and mental health and well-being. On health grounds alone these people would benefit from rights advocacy.
Achenbach, T. M., & Edelbrock, C. (1981). Behavioural problems and competences reported by parents of normal and disturbed children aged 4 to 16. Monographs Of the Society for Research in Child Development, 46, 1-82.
Alizadeh,H. (2010). Iran: transgender people no longer classified as “mentally sick”. (translation of Fars news agency report). Available on the website of the global forum on MSM and HIV (accessed 8th June 2011 at www.msmgf.org/index.cfm/id/11/aid/691
Chi,X.L. (in preparation). Sex education of adolescents: a comprehensive study on sex attitude, sex moral value and desire of sex education among college students in China.
Curran, B. (2009). Pakistan to recognise eunuchs The Nation (Abu Dhabi, 30th June, 2009), archived at http://groups.yahoo.com/group/transgendernews/message/39030. Accessed 9th June 2011.
Cohen-Kettenis, P., & Pfafflin, F. (2003). Transgenderism and Intersexuality in Childhood and Adolescence:Making Choices. Thousand Oaks: Sage.
Cruz, C. J. P., & Rogando-Sasot, S. (in preparation). I prefer to be the opposite sex: prevalence of cross-gender desires and male-to-female transsexualism among Filipino youth.
De Cuypere, G., Van Hemelrijck, M., Michel, A., Carael, B., Heylens, G., Rubens, R., et al. (2007). Prevalence and demography of transsexualism in Belgium. European Psychiatry, 22(3), 137-141.
Gates, G. (2011). How many people are LGBT? Los Angeles: UCLA School of Law, Williams Institute. Retrieved 8 June 2011, from http://www3.law.ucla.edu/williamsinstitute/pdf/How-many-people-are-LGBT-Final.pdf
Jamaludin, F. (2001, Jan 21st). Transsexuals: declare us as women. Originally published in The Star Newspaper. Archived at http://ai.eecs.umich.edu/people/conway/TS/MalaysianTS.html (accessed 8th June 2011)
Nanda, S. (2000). Hijra and Sadhin: neither man nor woman in India. In S. Nanda (Ed.), Gender diversity: cross-cultural variations. Long Grove, Illinois: Waveland Press.
Nationmaster populations statistics. www.nationmaster.com (accessed 7th June 2011)
Olyslager, F., & Conway, L. (2007). On the calculation of the prevalence of transsexualism. Paper presented at the 20th WPATH Symposium, Chicago.
Reed, B., Rhodes, S., Schofield, P., & Wylie, K. (2009). Gender variance in the UK: prevalence, incidence, growth and geographic distribution. Accessed 8th June 2011 at www.gires.org.uk/assets/Medpro-Assets/GenderVarianceUK-report.pdf
Sitapati, V. (2009, March 14). Sex-change for free. IndiaExpress.com, from www.indianexpress.com/news/sexchange-for-free/434285/0
Taylor, A. (2011, January 9, 2011). At home on the Tiwi Islands but doing it tough: meet Crystal Love and her Sistagirls. Sydney Morning Herald, from http://www.smh.com.au/lifestyle/lifematters.at-home-on-the-tiwi-islands-but-doing-it-tough-meet-crystal-love-and-her-sistagirls-20110108-19/9s.html
Teh, Y. K. (2002). The Mak Nyahs: Male to Female Transsexuals in Malaysia. Singapore: Eastern Universities Press.
Tsoi, W. F. (1988). The prevalence of transsexualism in Singapore. Acta Psychiatrica Scandinavica, 78(4), 501-504.
US Census Bureau, International Data Base. www.census.org (accessed 15th May 2011).
Veale, J. (2008). Prevalence of transsexualism among New Zealand passport holders. Australian & New Zealand Journal of Psychiatry, 42(10), 887-889.
Verhulst, F. C., Van der Ende, J., & Koot, H. M. (1996). Handleiding voor de CBCL/4-18. Rotterdam: Afdeling Kinder- en Jeugdpsychiatrie, AZ/Erasmus Universiteit.
Wikan, U. (1991). Behind the veil in Arabia: women in Oman. Chicago: University of Chicago Press.
Winter, S. (2002). Counting kathoey Retrieved 22nd October, 2008, from http://web.hku.hk/~sjwinter/TransgenderASIA/paper_counting_kathoey.htm
Winter, S., Chalungsooth, P., Teh, Y. K., Rojanalert, N., Maneerat, K., W., W. Y., et al.(in preparation). Gender disaffection: an international study.
Yu Lu (2009) Gender-related behavior, gender identity, and psychological adjustment in Chinese children (HKU Faculty of Education, PhD thesis)
Zucker, K., Bradley, S., Owen-Anderson, A., Kibblewhite, S., & Cantor, J. (2008). Is gender identity disorder in adolescents coming out of the closet? Journal of Sex and Marital Therapy(34), 287-290.