New polling shows a significant growth in the share of adults, especially, young adults who identify as Lesbian, Gay, Bisexual and Transgender (LGBT).
Most of the growth in the rate of LGBT identification appears to be due to younger people identifying that way, rather than from older people changing their LGBT identification.
The most plausible interpretation of the data (not just from the survey, but from other sources), in my view, is that there is definitely a biological predisposition (not necessarily genetic, but probably present at birth) to be lesbian, to be gay, to be transgender, and possibly also to be bisexual, although the mechanism may not be physical gender neutral (there are cases of people whose physical gender is ambiguous, but this is rare) with the mechanism causing a physical female to be lesbian, bisexual or transgender differing from the mechanism causing a physical male to be gay, bisexual or transgender. There may be a "spectrum" of gender identity and gender attraction, but I suspect that it is a very bimodal one.
Notably, some of the most striking early transgender identification cases in the literature involve children who were physically gender ambiguous at birth who were surgically modified in infancy, for example, to appear female despite an ambiguous but male leaning physical gender ambiguity at birth, who psychologically were fully male from a very young age onward.
But the factors that cause someone who has a biological predisposition to be LGBT (or someone who is biologically predisposed to be straight and cis-gender) to identify in the manner that they do is strongly culturally driven, in part, due to have these constructs of defined culturally, and this is what is causing the generation differences observed.
For example, with respect to transgender identification, the 0.2-0.3% identification rate for people older than Millennials, probably almost entirely reflects a biological predisposition, and to some extent is probably an undercount of people biologically predisposed to be transgender because the vast majority of people in the older generations were not aware of the possibility that someone could be transgender, rather than gay or lesbian.
Indeed, the line between being gay and lesbian and being transgender is not entirely black and white. Among people who identify as gay, some live their lives in a very conventional cis-gendered way, while others welcome effeminate touches and tendencies in their lives and may also cross-dress at times. Likewise, among people who identify as lesbian, some tend towards a "butch" identity, while others have a strongly cis-gendered "femme" identity. An effeminate gay man, or a butch lesbian, to some extent blurs the line between being cis-gender gay or lesbian, and being "straight" transgender.
The jump in the total share of people identifying as gay, lesbian or transgender from 0.8% among Traditionalists to 2.1% in Generation X, probably largely reflects a decline in repression and denial of these biological predispositions.
Some of the rise in this percentage from Generation X to Millennials (4.0%) and Generation Z (5.3%) may reflect the same thing, but my intuition is that the gay, lesbian and bisexual percentage growth between Generation X and the subsequent generations mostly reflects a tendency for people who are not fully stereotypically straight cis-gender, even if that is their predominant biological predisposition, to identify as gay, lesbian or transgender. So, for example, a physical male who loves pink, enjoys reading romance novels, prefers to fine arts to sports, and isn't freaked out by touching another man, might identify as gay or transgender despite not having such a strong biological predisposition towards either identity.
The surge in bisexual identification from 0.3% among Traditionalists to 1.8% in Generation X to 11.5% in Generation X almost surely also involves a similar phenomena.
Past data had suggested that among people older than my age, that most men identifying as bisexual came to identify after a transitional period of identifying as bisexual as gay, while a bisexual gender identity was stable among women (some of whom were arguably lesbian women at a fundamental level who tolerate sexual activity with men).
Now, the rise in bisexual identification probably largely reflects a decline in the strength of the taboo against intimate same sex physical touching, rather than a change in primarily biological predisposition to be sexually attracted to someone of the same physical gender.
If you are primarily biologically predisposed towards sexual attraction towards someone of the other physical gender (i.e. you are "straight"), but don't feel a strong taboo against same sex intimate contact, identifying a "bisexual" in a hypothetical sense is a much less difficult thing to square with your daily reality than identifying as gay, lesbian or transgender.
This cultural shift is different than lying (which is also present in survey data, but rare and which there is no reason to think that younger generations are more prone to than older ones). It is really more a case of redefinition of terms from the narrow sense predominant in older generations, to a broader sense predominant among younger people. Thus, the term "bisexual" appears to have been redefined from an older sense of lacking a predominant gender preference in sexual attraction, to a newer sense of lacking a strong aversion of same sex sexuality.
See also a 2013 post at this blog on a similar topic.