Every year since 2013, Gallup has been releasing new data on the size and composition of the LGBT+ identifying population in the U.S. Measuring gender and sexual identities on this scale presents a host of challenges. My sense is that the figures that result are bound to be conservative estimates of the size of the LGBT+ population in the U.S. I aways chart Gallup’s data because I often make use of the figures in some of my courses and find it more engaging to look at figures than tables. So, I thought I’d share those here for anyone interested.
One important place to start with the figure above is asking what it might actually mean. I am not convinced that the story told should only be understood as one in which there are simply “more” LGBT+ people than there used to be – which might strike some as a recommended interpretation of the figure above. Some important scholarship does suggest a radical shift, of the “more than there used to be” variety. For instance, Danya Lagos’s recent article in the American Journal of Sociology, “Has There Been a Transgender Tipping Point? Gender Identification Differences in U.S. Cohorts Born between 1935 and 2001” (HERE), documents cohort effects illustrating shifts in the likelihood that individuals identify as transgender or gender non-conforming in later birth cohorts. Importantly, this work was only about gender, not sexual demographic shifts. But it’s an interesting question and debate. Do these shifts demonstrate radical change in the experiences of gender and sexualities over time, or are more people simply adopting identity labels than used to in the past? I’m not convinced it’s an answerable question, but I’m interested in the debate.
Despite this, as of 2022, Gallup documented that 7.2% of the U.S. population identifies as LGBT. This is a remarkable shift from the 3.5% they documented when starting this data collection process in 2012. There have been shifts in the way in which the question has been asked of respondents. And no doubt, this will frustrating demographers trained in the “if you want to measure change, don’t change the measure” school of demographic analysis. I’ve always found that mantra frustrating when used to justify continued use of questions and measures that might have worked in earlier waves of data collection, but become out of touch over time. Identities like race, gender, sexuality, and more are so interesting precisely because they transform – sometimes radically – over time. So too does a population’s understanding of those shifts. In 2012, during the first phase of data collection by Gallup, they did not trust cisgender heterosexual people to understand questions about their gender and sexual identities. So, the question offered was a “yes/no” “Do you identify as lesbian, gay, bisexual, or transgender?” But now, Gallup asks respondents to check all that apply.
And the menu of options has changed over time as well, invariably playing “catch-up” with the ways these shifts are happening on the ground. For instance, in the 2021 data (see HERE), respondents could check all that apply, but they were not offered two options respondents were offered in 2022 (“asexual” and “queer”) – both of these responses would have been captured by “other LGBT” in the 2021 data. Below, you can see that “queer” identities are claimed by about 1.2% of those claiming LGBT+ identities; “asexual” is claimed by 1.3% of those claiming LGBT+ identities.
As you can see below, it remains true that “bisexual” accounts for the lion’s share of shifts in LGBT identification (something I’ve previously written about with D’Lane Compton HERE as well as Mignon Moore HERE/HERE). It is also true, however, that this question is of the “check all that apply” variety. So, someone can select all, none, or some of the identities in the figure below which is why the proportions don’t all add to 100%. Gallup did not supply these data, but the proportions suggest that few people select more than one identity. Thus, we don’t know, for instance, what share of those identifying as “queer” here also identify as “gay” or “lesbian” or “transgender.” But, the proportions suggest that a majority of respondents selected one identity.
In the 2022 report (on 2021 data), Gallup provided a gender breakdown for how people of different genders responded to these identity options questions. This is something I have been interested in for a long time. But this year, Gallup did not report the data by gender. I’ll be writing them to see if they’d be willing to share those figures and update this post if I receive the data and able able to share those results as well.
Gallup still reports the data by generation labels (Gen Z, Millennials, etc.). But, as sociologists are largely moving away from considering generation labels demographically meaningful, I am reporting the data simply by age groups below. As was true in the 2021 data (HERE), it is still true that “bisexual” accounts for largest share of LGBT+ identities among 18-57 year olds. But “gay” accounts for the largest share of LGBT+ identities among those 58 and older. “Other LGBT” remains most represented among the youngest cohort still as is also true of “transgender” identification. And “gay” still accounts for a larger share of LGBT+ identities than “lesbian” for each age group depicted below (this may be in part due to the fact that people selecting “lesbian” may also select “gay” – though, as I mention above, I think this may account for a smaller group of respondents than we might think).
Finally, Gallup shared figures on the proportion of LGBT+ adults claiming different LGBT identities offered on their most recent survey. So, while the above figure reports on proportions of all American adults in the different age ranges claiming specific LGBT+ identities, the figure below charts the proportions of LGBT+ identifying American adults claiming specific LGBT+ identities on Gallup’s survey. You can see that roughly two thirds of 18-41 year olds identifying as LGBT+ identify as “bisexual.” That’s a huge proportion of LGBT+ identifying people and continues to recommend that we invest more time and resources in investigating bisexuals.
Another interesting fact that jumped out at me from the figure below is that the largest share of respondents identifying as “lesbian” are now 58-76 years old. About one in four LGBT+ identifying adults in the same in that age range identify as “lesbian.” The figures are much lower in all of the other age groups. It’s also really interesting that “other LGBT” accounts for the largest proportions of LGBT identifying respondents in both the youngest AND oldest age groups here – accounting for the least share of respondents in the middle age groups.
Finally, the share of LGBT+ respondents identifying as “gay” is highest among the oldest groups listed below. Almost half of all LGBT+ adults over the age of 76 identify as “gay,” while only 17% of those in the youngest age groups here selected that identity. There are a couple things that might be going on. Part of what is going on is that there are simply so many young people identifying as “bisexual” that are simply driving the proportions of all other LGBT+ identities down. But it means that the composition of the LGBT+ community is really interestingly stratified by age in ways that probably deserves a lot more attention.
Not much here is really “new.” But it’s always interesting to see the new figures as they offer one way of tracking shifts and transformations in gender and sexualities over time. Soon, I hope to be working again with D’Lane Compton and Mignon Moore examining shifts in “bisexual” identities using different databases. And, again we did not get to see Gallup’s data reported by state – something they used to report in the earliest years of the survey. Fun to see the newest figures. But I continue to wish they shared more or made their data more easily accessible so that we can answer some of the questions posed above (among others).
Inequality by (Interior) Design 