Evolving exactitude

A forty year old reasonably simple epistemic question.  What percentage of the population are gay or lesbian?  I first came across this in the late seventies and was intrigued by the fact that most numbers were arrived at through broad estimation rather than methodologically rigorous enumeration.  Trans and bi- were added later as categories.

The uncertainty made sense.  LGBT civil rights were the junior partner of racial civil rights - a lot of momentous legislation and shifts in social perception emerging at the end of the 1960's (the Stonewall riots were in 1969) and being absorbed by the public in the 1970s.  All, ultimately for the better but the full implication of legislation frequently preceded public cognition and certainly preceded substantial change and far preceded any firm epestimic confidence in related numbers.  

When I was in college, I think the generally accepted estimate was that some 10% of the population was LGBT with the acknowledgement that there was a material margin of error though without consensus as to how big that margin of error might be.  My supposition was that the range might be 5-15% but the probability being that it was much more likely to be on the lower end of that range.  

All major legislation and policy change requires energetic mobilization of public opinion, at least in the appearance rather than the reality.  Advocates over claim on their numbers of beneficiaries and the value to those beneficiaries and under claim the number of those who will be negatively affected or are opposed to whatever the change might be.  True, no matter what the cause.

In the past decade or two, it has seemed that the consensus estimate has fallen to a lower estimation, i.e. about 5% of the population of the population is LGBT.

But now there is a new Gallup polling looking at the numbers and with a more refined approach than before, specifically in terms of definitional exactness which has always been a challenge.

What does it mean to be gay or lesbian or bi?  Is it demonstrated acts and lifestyles?  Is it self-identification?  Is it binary (i.e. if you sleep with a member of the same sex once, are you permanently in the category of lesbian or gay?)  All fair and important clarifications.  The new survey tries to clarify with finer distinctions and yields some striking results.

From LGBT Identification Rises to 5.6% in Latest U.S. Estimate by Jeffrey M. Jones.  In 2012, 3.5% identified as LGBT, rising to 5.6 in 2020.  But the headline is not what is so interesting.  It is the definitional detail.  

More than half of those who identify as LGBT, identify as bisexual rather than gay or lesbian.  Among all adults in the nation:

Lesbian - 0.7%

Gay - 1.4%

Bisexual - 3.1%

Transgender - 0.6%

Consequently, the implication is that only (0.7+1.4) 2.1% of the adults identify as traditional Lesbian or Gay.  

There is an age and gender angle on this as well.  Women are 31% more likely to identify as LGBT as men (6.4% versus 4.9%).   By age cohort for Generation Z (1997-2002) among all adults:

Lesbian - 1.4%

Gay - 2.1%

Bisexual - 11.5%

Transgender - 1.8%

My guess is that we have transitioned from a "No Orientation Discussion" environment to an "All Orientation Discussion All the Time" environment.   That simple opening of the Overton Window across the generations has driven some material amount of perceived changes.  

If you shift the question away from particular generations and modernizing social mores to something more fundamental, the answers are probably both more meaningful and more useful.  Something like "What percentage of people have a sustained and relatively unvarying orientation over a lifetime?"  Essentially you are getting to what is real and unvarying and avoiding the topical, the exploratory, the obfuscatory, the fashionable.  

It is probably too tight a formulation but I suspect the numbers might end up looking something like:

Lesbian - 1.0%

Gay - 2.0%

Bisexual - 5.0%

Transgender - 0.8%

Just a guess and with a significant adjustment on the Transgender number owing to a recent and sustained advocacy around Transgender as a last perceived underserved civil rights frontier.  

That leaves us with the traditional binary distinction of the 1970s "What percentage of the population are gay or lesbian?"  The answer is not 10% or 5%.  It is 3.0% (1.0 + 2.0).

There is nothing bad or good about the outcome, and it is still an estimate.  But it begins to clarify what we have been talking about and how sloppy that conversation has been for so long.