Given that we can only label someone as "gay" once they come out, you can replace all instances of "gay" with "people who have admitted to themselves and others that they are gay".
It's not surprising that children of gay parents would be more likely to accept their own sexuality (whatever that might be). Given that people default to "straight", you're going to get more openly gay/bi children in open minded families.
There's no need to invoke "genes".
I'm sure a little bit of work would also show that straight liberals, hippies, or people that list themselves as "pro-diversity" on a questionnaire are also much more likely to have children that are "gay". Certainly, just looking at grad school: liklihood of grad students being gay/bisexual (more the latter) is waaaay higher than you see elsewhere. I assume it's because they're more open to questioning their sexuality than most.
Note: Also worth noting that Schumann is trying to argue for a strong non-
genetic component to sexuality. He's arguing that upbringing plays a role.
: Borderline Ridiculousness.
just read the rest of the article. Regardless of what the truth turns out to be Schumann's research is a meta-analysis that includes, among other things, reviews of texts in gay parenting books.
NYT article, linked by OPHis study is a meta-analysis of existing work. First, Schumm extrapolated data from 10 books on gay parenting; Cameron, for what it's worth, had only looked at three, and offered no statistical analysis in his paper. Schumm skewed his data so that only self-identified gay and lesbian children would be labeled as such.
This is important because sometimes Schumm would come across a passage of children of gay parents who said they were "adamant about not declaring their sexual orientation at all." These people would be labeled straight, even though the passage's implication was that they were gay.
Schumm concluded that children of lesbian parents identified themselves as gay 31 percent of the time; children of gay men had gay children 19 percent of the time, and children of a lesbian mother and gay father had at least one gay child 25 percent of the time.
Meta-analyses have a host of problems. But from the minimal description in this article his sounds like it is strongly
biased. You don't need a PhD to see huge problems with trying to extrapolate real world sexual orientation figures from passages in parenting books, which are selected to illustrate various issues, not be statistically representative of all families.
I wouldn't put too much stock in this particular study without a lot more info.Meat
dittoFinally, Schumm looked at the existing academic studies, the ones used to pillory Cameron's work. In all there are 26 such studies. Schumm ran the numbers from them and concluded that, surprisingly, 20 percent of the kids of gay parents were gay themselves. When children only 17 or older were included in the analysis, 28 percent were gay.
^^^These numbers are more likely to be legitimately representative. But, if nothing else, still subject to huge self-reporting biases.