Limited Interest Data Sets

In American Conservative, Rod Dreher has a Hallelujah! moment upon the discovery he may have been right all along about sexual orientation:

Wait, so you mean not everybody is “born this way”? You mean that it’s not simply nature, but also nurture? I’m so confused.

Actually I’m not confused at all. The “truth” in this matter has always been “what works to advance the cause.”

But for those who want to grapple honestly with this issue, these data from Patrick Egan show pretty clearly that the nurturing that culture provides does make a big difference.  Therefore, for communities who wish for their children to remain heterosexual, to form heterosexual marital unions, traditional families, etc., neutrality on the matter of sexuality will result in five to eight times as many people claiming homosexuality or bisexuality as would have otherwise been the case. (There have also been skyrocketing numbers of people claiming to be transgender.)

Andrew Sullivan is appalled:

But the obvious explanation for these numbers is a simple one: it is that millennials simply have much less shame about sexual orientation than older generations. Growing up in a world of legal marriage equality, that should not be surprising. Gallup suggests as much: “It’s likely that millennials are the first generation in the U.S. to grow up in an environment where social acceptance of the LGBT community markedly increased. This may be an important factor in explaining their greater willingness to identify as LGBT.” It may also, in some way, be “cool” to identify as bi in your teens or in college in a way it wasn’t before. It doesn’t tell us anything meaningful about much else — and certainly not whether they will have a gay relationship in the future.

For me, this affirms the observation that a value of a dataset is inversely proportional to the number of ways it can be interpreted. We see two here, as Rod tries to spin it to his position that heterosexuality is the only true way, but our fallen society (nurture) is corrupting us. He is making the mistake that believing all the data is true. In the absence of an admission by the surveyors that they used truth drugs on respondents, we know that’s a risky assumption. Furthermore, as I’ll elaborate on later, he assumes all the biases and factors of the data are constant over time.

Andrew sees this graph as a measurement of a societal fear, reasoning that earlier generations in which admitted homosexuals (and therefore bisexuals) were excluded from traditional society would have lower rates of such admissions. This position also has some problems, as this is obviously a snap survey and not a survey over time, by which I mean all questions and responses were collected within a few weeks to perhaps a year; in that time gap, there would not be a meaningful shift in societal norms. Therefore, the 90 year olds are operating in today’s society, not that of 70 years ago, and thus not biased by obsolete societal norms. This critique may itself be critiqued by observing that the very presence of those societal norms of 70 years ago will distort today’s data as some “true bisexuals” will have buried that attribute of their personality and answer as heterosexuals. Again, the shame they learned 70 years ago may inhibit the answers of those 70-90 year olds who know they are bisexual – even in an anonymous survey.

I’ll raise a question of my own: what is the life expectancy of a bisexual, and has it varied over the years? I strongly suspect the answer is Yes, relative to the general population it has been negatively biased but the bias has been generally reduced as years have advanced. In fact, thinking about it, the HIV epidemic impacted the bisexual community – and therefore life expectancy of key respondents. This leaves Rod high and dry, since the population bias brought on by premature deaths will bias the responses in this survey, irrespective of societally-induced answer bias. So we can interpret this dataset to reflect the premature death biases introduced by both deadly plague and now-obsolete traditional societal norms for the bisexual community. Once again, the value of this dataset is reduced because of what appears to be valid but clashing interpretations. Unless there’s more information available about the survey and how it corrected for various factors, well, this dataset is rapidly becoming worthless.

Isn’t statistical data analysis just bloody fun?

And here’s the oddest thing: I think Rod is really just riding a hobby horse into the ground. I’ll dispense with the normal lampooning and hyperbole, and simply tell my reader that, living in Minnesota most of my life, I was witness to the struggle over Minnesota Amendment 1 of 2012, the amendment to ban gay marriage. According to Wikipedia, while polls were volatile, generally the pro forces began on the high side but fought a losing battle, and the Amendment was defeated, 51.19% to 47.44%. The shift in public opinion, that movement from the blind acceptance of a dubious interpretation of a book written more than a millenia ago (and then assembled by a group of men, picking parts and rejecting others), to actually thinking about justice, just behavior, and when being inclusive is good or bad, was truly one of the more inspirational periods in Minnesota history. I only regret the margin of defeat wasn’t much greater.

Just as the relegation of women to a small number of roles within society ultimately impairs society by eliminating their genius from the inputs which make society, the suppression of other citizens because of behavior & attributes of no negative consequence also impairs society. I think we’ve made good, positive strides forward not just by defeating the Amendment, but by having that public discussion, and exhibiting the societal plasticity to recognize this old prejudice for what it was – a prejudice.

[EDIT: 10/21/2017 Fixed typo.]

Bookmark the permalink.

About Hue White

Former BBS operator; software engineer; cat lackey.

Comments are closed.