If it's mainly down to biology you'd expect that the proportions involved would be pretty consistent in different places and times. On thee other hand,if it's mainly cultural you wouldn't necessarily.

But then it all gets skewed up by other factors, notably predominant social attitudes and the legal situation. If people identified as gay are liable to be persecuted that could be expected to make them a tiny bit reluctant to come out as such in response to surveys.

But the actual figures aren't what matters - either low figures or high figures could equally be used as a basis for either acceptance or persecution. Or for anything in between, for that matter.