Box Turtle Bulletin

Box Turtle BulletinNews, analysis and fact-checking of anti-gay rhetoric
“Now you must raise your children up in a world where that union of man and box turtle is on the same legal footing as man and wife…”
This article can be found at:
Latest Posts

A new calculation of the percent of men in America who are gay

Timothy Kincaid

December 7th, 2013

In an opinion piece titled, “How Many American Men Are Gay?”, Seth Stephens-Davidowitz seeks to answer that question. (New York Times)

At least 5 percent of American men, I estimate, are predominantly attracted to men, and millions of gay men still live, to some degree, in the closet. Gay men are half as likely as straight men to acknowledge their sexuality on social networks. More than one quarter of gay men hide their sexuality from anonymous surveys. The evidence also suggests that a large number of gay men are married to women.

Stephens-Davidowitz finds his solution by looking at meta-data from Facebook, Google, and Match.com. While this may seem a less rigorous methodology than some which have been employed, there are no truly conclusive methods – or, at least, no economically feasible and non-invasive methods – for measuring the internal longings and attractions of a small percentage of the population.

While this is not by any means conclusive, his calculation adds to the growing research which suggests that greater than three percent, and less than ten percent, of men in the united states are predominantly attracted to the same sex. (I tend to favor a number closer to 4%, but it’s difficult to be certain).

Comments

POST COMMENT | COMMENT RSS 2.0

Atriokke
December 7th, 2013 | LINK

hahaha this is so true. None of my facebook accounts denotes me as a gay man.

However I wonder, how does the meta-data from google or facebook know if someone is gay if it doesnt say so on the actual data. Perhaps it’s rather extracted match.com?

occono
December 7th, 2013 | LINK

5% seems about right.

tristram
December 7th, 2013 | LINK

As I understand it, the 3% figure is based primarily on surveys that depend on self-identification. I think it significantly undercounts us. Double it and you’re right in the ballpark, probably still low if predominantly means 50+%.

Priya Lynn
December 7th, 2013 | LINK

I agree with Tristam. You can’t put much faith in surveys based on self-identification, there’s no doubt the greatly undercount the percentage of gay and bisexual men as many, perhaps most same sex attracted men are unwilling to be open about it with a stranger.

FYoung
December 8th, 2013 | LINK

An important research paper that was recently published by Katie and Lucas Coffman suggests that conventional survey methods drastically underestimate both homosexuality and homophobia/heterosexism. The Coffmans used a veiled elicitation method to correct for social desirability bias.

Unfortunately, the numbers they arrived at on the incidence of homosexuality are not credible due to unrepresentative sampling, a fact they recognize, but there is no reason to suppose the under-reporting gap they identified would not have been replicated with a representative sample.

Broadly speaking, their research shows that the incidence or homosexuality is about 60% higher, and the incidence of homophobia and heterosexism is about 70% higher, than shown in even the best surveys that use direct Q&A methods.

http://artsandsciences.osu.edu/news/gay-population%E2%80%93and-homophobia%E2%80%93may-be-underestimated

FYoung
December 8th, 2013 | LINK

PS: “…the incidence of homophobia and heterosexism is about 70% higher…” This is a crucial factor to keep in mind when planning and conducting popular initiatives to repeal anti-gay laws and constitutional amendments.

JoeHardona
December 8th, 2013 | LINK

More like 25% – if you count Evangelicals, open all the closets and include married men (congressmen too))

Priya Lynn
December 8th, 2013 | LINK

Thanks for that Fyoung. Its a major annoyance for me when people refer to surveys and claim the results show only X% of people are gay.

Jim Hlavac
December 8th, 2013 | LINK

There are 106 boys born for every 100 girls — in every human group on earth.

The 6 extra boys are gay — thus,5% are gay.

How simple.

But no, let us turn to guessing of data, and extrapolations of modern social page usage, and then too phone surveys, and every other Byzantine method that only Ptolemy could conceive, while Occam’s razor of 6 on the 100 is ignored.

Jons
December 8th, 2013 | LINK

Are you really that stupid and ignorant, Jim Havac or are you just making a point by being sarcastic? I hope for it’s the latter, otherwise it’s just really sad. Nothing sadder than idiots who think themselves more intelligent than the rest.

There are more boys born because boys have a higher mortality rate than girls during the first year, so it’s a way of compensation so that reaching adulthood there’s a 50% balance (actually in most countries, there is a slight surplus of women, like 51 – 49%).
Also, I guess by your theory there are no lesbians.

Ben in Oakland
December 8th, 2013 | LINK

Who or what is compensating before birth for people who die after birth?

Gene in L.A.
December 8th, 2013 | LINK

Jim Hlavac, it would be nice to think it’s that easy, but “Occam’s Razor” isn’t an immutable scientific law; it’s not always true. What correlation is there between the 106/100 ratio, even if it were universal, and your proposed 6% of gay men? Is there an “Occam’s Razor” number to explain gay women? It’s just not something that can be explained numerically. I’d like to believe 6%. Hell, I’d like to believe the 10% we always used to say. There’s just no way to make an accurate guess, and we may never arrive at a time when gay people will answer a survey truthfully.

Priya Lynn
December 8th, 2013 | LINK

It is a non-sequitor to say if 106 boys are born for every 100 girls that means six of the boys are gay. Occam’s razor in no way suggests six of them are gay. Jim should certainly know better.

Ed W
December 8th, 2013 | LINK

This country needs nationwide employment protection. Otherwise folks understandably stay in the closet to protect their livelihoods. The closet is way too crowed in the USA.

Jim Burroway
December 8th, 2013 | LINK

Priya is correct. The fact that there are six more boys than girls in the birth ration in NO WAY means that there are six gay boys born out of every 106. The birth ratio has been used with the sibling sex ratio in individual families to try to explain the possible effects of birth order the probability of the birth of a gay boy in some individual families (and those studies are rather tentative and somewhat speculative.) But the birth gender ratio itself has nothing whatsoever to do with the presence of gay boys/men. (Think about it — what would it say about lesbians?)

Richard Rush
December 8th, 2013 | LINK

I just assumed that Jim Hlavac’s comment was tongue-in-cheek as a parody of the nonsense that our adversaries consistently make-up about us ~ particularly the 99.9% who are religious adversaries for whom fantasy IS “reality.”

Priya Lynn
December 8th, 2013 | LINK

Richard, only Jim Hlavac knows for sure.

Jim Hlavac
December 9th, 2013 | LINK

Jons — no I am not “stupid and ignorant” at all — I have done far more number analysis than you could ever imagine — and your superficial comment here doesn’t even get to the skin of the matter.

However, you seem to be rather inept yourself – since you spelled my last name “havac” — when the correct spelling is right in front of your eyes.

I would suggest sir, that if you can’t read and then write a simple six letter name you should reconsider who is stupid …

Meanwhile, Jons, if you spell my name correctly you will find my book “The Pink Sheep of the Ninth Circle” on Kindle which takes some 40 pages to look at the numbers of gay men.

Jim Hlavac
December 9th, 2013 | LINK

I see that several more people had a problem with my statement — I cannot in a blog comment space give my full reasoning and analysis — but after careful analysis to merely dismiss my theory is ill advised.

For instance — it is true that there is a 51%-49% skew towards females in the general population — but that is weighted towards the over 70 crowd. That is, at birth there are 106-100 boys to girls — while at 20 years of age it’s roughly 102-100, and at 70 an older it is 30-100 in favor of females.

You people are taking broad aggregates and supposing that it is static throughout every age cohort.

However, if it is mathematically true that there is a 51%-49% female majority – then it must also be true that there cannot be 106-boys born for every 100 girls — for that gives us a higher male-to-female ratio. And you cannot have both a higher and lower male female ratio at the same time. Which, then, is it?

I would suggest you all think your math through more again — and go find my book on Kindle — where I had use 40 pages to look at the numbers — rather than call me an idiot. Thank you.

Jim Hlavac
December 9th, 2013 | LINK

By the way, gentlemen — there have been numerous studies of marriage records of the Middle Ages through modern times — every single one shows 93% of men get married while 97% of women get married. Let us suppose that at age 21 there are indeed 100 men for every 100 women — because the six extra boys at birth all die before 21.

If the 100 marry the 100 — how then is this consistent 93%-97% difference accounted for?

Meanwhile, since the vast majority of gay men today don’t get married ever (though you will not find one statistic anywhere that purports to say this or that percentage of gay men were once married to a woman) if 5% of the men are never-married gay men (which equals 5 or 6 million gay adult men, and 2 or 3 million gay male children in America (gay children being another never counted number) then this would mean that there must be some 5 million never married women. There is no such figure ever proposed in any census data — these spinsters don’t exist. Given that there’s 2 lesbians for every 5 gay guys (oddly, the same ratio for autism, left handedness and other sex-gender disparities) it still leaves many an unmarried female simply not there in the data.

Another odd way to look at this is that according to Entrepreneur Magazine, and some state’s Dept of Economic Development analysis — it takes 5,000 people to keep a bar open. It just so happens that if you multiply the numbers of gay bars in virtually every metro area of this nation by 5,000 you get 5% of the male population of that area. It is rather strange, then, how the 5% figure keeps showing up — and that it corresponds to the 106-100 ratio at birth.

One could go on — into quite a study — which I have done. Thank you.

Priya Lynn
December 9th, 2013 | LINK

La te da Jim Hlavac. Absolutely none of that demonstrates that any number of boys are gay – your entire line of “reasoning” does not follow. I just can’t believe you cannot see the obvious.

David
December 9th, 2013 | LINK

Jim, you wrote:
“There are 106 boys born for every 100 girls — in every human group on earth.

The 6 extra boys are gay — thus,5% are gay.

How simple.

But no, let us turn to guessing of data, and extrapolations of modern social page usage, and then too phone surveys, and every other Byzantine method that only Ptolemy could conceive, while Occam’s razor of 6 on the 100 is ignored.”

But then say that we shouldn’t have commented on this, but some obscure 40 pages you wrote somewhere of Byzantine method that only Ptolemy could conceive.

Sir Andrew
December 9th, 2013 | LINK

I lean toward 8%. Unless booze is involved, in which case it’s 64%.

enough already
December 9th, 2013 | LINK

Oh, dear here we go again. I have followed the nastiness on this subject both here and on other gay blogs many times.
And each time, somebody gets torn to shreds for daring to make a statement which simply isn’t acceptable to the reigning commentators.
The 2% nonsense. The 4% nonsense. Now this one.
All that’s missing in this nasty comment thread is for somebody to throw in the ‘correlation is not causation’ card.

I do find it interesting that every single ‘we now know’ bit of nonsense is embraced by so many here and then so totally discarded not even a few months later for the next ‘now we know’ piece of ‘research’.

Pacal
December 9th, 2013 | LINK

Mr. Hlavac first it is not mathematically true that there is a 51% female v. 49% male in America at least. That is simply empirically true. I also don’t why in one posting you say it can’t be true that there is a 51% female majority if there are 106 boys born for every 100 girls. Actually it is easy just accept that boys have a higher death rate. Also there isn’t 106 boys for every 100 girls. There are 106 boys born for every 100 girls. In the same post you say that 20 there are 102 men for 100 females. So what is the problem?

In your next posting you claim that numerous studies have shown from the Middle ages to modern times that 93% of men and 97% of women have gotten married. Your assumption seems to be that the remaining % is largely of Gay people. What nonsense. In the Middle Ages and early modern times women were expected to get married and frequently coerced into doing so while men had greater options including careers in the clergy, (women had lesser opportunities to enter the clergy), that did not allow marriage. You can’t assume that most of these people were Gay. Further it is simply not true that a consistent 93% of men and 97% of women got married over such a long period of time. The marriage rate varied considerably and had much more to do with the ability to support a family than anything else. In times of economic stress the marriage rate fell and in some cases fell dramatically. Further your statement that the marriage rate was over 90% is not believable. statistics from the Middle ages are frankly dubious and questionable. however early modern England does provide some records. It appears that in the mid 17th century the proportion of people who had never married at of 40-44 years of age was almost 24%. That is rather high And that was because uncertain economic conditions i8nhibited the formation of families. as for the supposed non existence of 5 million female spinsters. Well if it is consistent that 97% of women get married than out of c. 150 million women c. 4.5 million never marry. well that is dubious. I should point out that now it appears that 49% of American adults are unmarried which is quite a change from 1970 when it was 28%. So yeah marriage rates etc., change over time.

Your comment that today the vast majority of Gay men don’t get married is an assumption that is likely not true. AS for 2 Lesbians for every 5 y male again not set in stone. The spinsters do it seems exist.

AS for counting the number of Gay bars etc. You can’t be serious. That is simply nonsensical except in the most general sense. I note that you forget that a lot of Gay men don’t go to bars.

Pacal
December 9th, 2013 | LINK

Mr. Hlavac, there is nothing strange about 106 boys being born for every 100 girls, and for us to have 51% women and 49% men. All that is required is a different death rate. Since you mention 102 males for 100 females at age 20 you seem to accept that. As for your consistent figures of 93% of men getting married and 97% of women. Aside from the fact that men traditionally have had much greater opportunities to pursue careers that don’t involve families, (Clergy for example), whereas women were pressed into getting married your percentages are dubious. The marriage rate has varied over time Historical records of this type for the Middle ages are dubious. we do have reasonably accurate records for early modern England and in the mid 17th century 24% of adults aged 40-44 had never married. AS for the spinsters. Even using your own figure of 97% of women never getting married you get 4.5 million women in the US never marrying. Historically middle age spinsters have in fact been “common”. Your spouting nonsense.

Pacal
December 9th, 2013 | LINK

enough already – And your point?

Priya Lynn
December 9th, 2013 | LINK

Correlation is not causation.

enough already
December 10th, 2013 | LINK

My point, Pacal, is that these discussions are absurd.
It is desirable for the social-sciences to attempt to apply the rigor of the natural sciences to their work.
The attacks upon those who question these results (and Boxturtle has seen several really nasty back and forths on this over the last few years in the comment sections) are, however, absurd.
I’ve seen the entrenched group here assign a probability approaching 1 to everything from 2% to 4% and now 5% with nothing but disdain for anyone who dared to question the shifting foundations of wet sand underlying the research behind each statement.

That’s my point, dear Pacal – Hlavac may or may not have any basis for his attempt to establish a causal relationship between male/female birth ratios. I’ve no opinion on the matter.

You and all the others being nasty to him have, however, nothing in the natural sciences to back up your attacks. You’re working from the same artificial color and artificial flavor basis as he is, but you are pretending that he simply can’t be right while you, of course, are applying ‘real’ science to your well-considered conclusions.
Until the next flavor of the week ratios come in. At which point you all will jump on them as ‘proof-positive’ how many of us gay men there really are.

Priya Lynn
December 10th, 2013 | LINK

Enough, when you start out with absurd assumptions like “If 106 boys are born for ever 100 girls then 6 boys must be gay” you simply CANNOT be right. It doesn’t matter how sophisticated and detailed your statistical analysis is when it starts out with an irrational assumption it can’t be right – garbage in, garbage out.

You don’t need to be a rocket scientist to know this and given that neither you nor Jim Hlavac know it I don’t need to spend any time reading further details from you to know you have no idea what you’re talking about.

Priya Lynn
December 10th, 2013 | LINK

I should have said “Enough, when you start out with absurd assumptions like “If 106 boys are born for ever 100 girls then 6 boys must be gay” you simply CANNOT be right except by random chance.

And you sure don’t need anything in the natural sciences to back up that truth.

David
December 10th, 2013 | LINK

Pacal, you wrote:

“You and all the others being nasty to him have, however, nothing in the natural sciences to back up your attacks. You’re working from the same artificial color and artificial flavor basis as he is, but you are pretending that he simply can’t be right while you, of course, are applying ‘real’ science to your well-considered conclusions.
Until the next flavor of the week ratios come in. At which point you all will jump on them as ‘proof-positive’ how many of us gay men there really are.”

What? We said that his comment was rubbish. None of us ever said that we know the real number and that we had a methodology that was correct. His statement was rubbish. I have no idea what he wrote in his 40 pages, but that is not what anyone was responding to. I have no idea what the accurate percentage is, but Hlavac’s original statement was ridiculous.

Priya Lynn
December 10th, 2013 | LINK

David, you’re quoting the commentor “Enough already”, not Pacal.

And I second your comment, I never suggested any particular number was the real percentage of gay men, I was just pointing out the logical fallacies in Jim Hlavac’s comments.

Ben In Oakland
December 10th, 2013 | LINK

I Have degrees in both the social sciences and math, though I opted out of both and became a professional photographer for 30 years.

One of the problems in the social sciences, as well as one of the benefits, is the attempt to make social data conform to the same rigidity that the physical sciences have. As one progresses from the bottom of the scale– economics– to the top of the scale– physics and chemistry– one finds that the reliability, conformity, and predictability of the data increases. The lower one goes on the scale, the more one finds that the human factor– call it free will, if you like– consumes a greater and greater percentage of the variation.

My own honors thesis, employing fairly sophisticated statistical analyses, got me a magna cum laude, but also was a pile of data crap. Fully half of my data was simply random. Calling Half of what was left “organized” could only be called somewhat wishful thinking.

Nevertheless– Magna cum laude.

One of the biggest problems in social data is the desire to make it conform to mathematical models when it simply doesn’t and isn’t likely to. You do a regression analysis of a shotgun blast of data on a barn wall, simply drawing a straight line through the scattered data and calling it linear, when all you’ve done is minimize the errors in calling the date linear. It’s the best straight line through a data set that isn’t a straight line.

I have no idea– nor does anyone else– about what the true percentage of gay people are. Or even how one might determine it, given the Ted Haggards, Larry Craigs, and Michael Glatzes of the world. The 10% often floated around is a misinterpretation of the kinsey figures from 65 years ago. It might be true, but is not justified as true based upon where it was pulled from.

Having read literally hundreds and hundreds of books on the subject of homosexuality– history, anthropology, psychology, sociology, zoology, theology, biology, among others– I’ve come to the conclusion that most people are actually bi, with a pronounced tendency towards one end of the polarity or the other. Whether that’s due to psychological or social forces, I have no idea. Nor does anyone else.

Despite two different hetero experiences 35 and 40 years ago, in which I performed like quite the hetero stud, I am nevertheless a homosexual man without the slightest desire for any further heterosexual experience. My brother in law had one or two homosexual experiences when he was young, but realized it was not for him, as he so nicely put it.

I have no problem with the 4% figure. I doubt we’ll never have an answer, because it will be impossible to know. we do know that, or at least can suppose, that more people are likely to come out in an accepting society, so the 70% homosexual person is less likely to rely on his 30% heterosexuality to get by– or bi, as the case may be. For myself, I’m 99% homosexual, and that 1% heterosexual is unlikely never to see the light of day again.

Pacal
December 10th, 2013 | LINK

enough could you drop the condescension. its quite boring. I also suggest you lose the mind reading. I made absolutely no comment about what percentage of the population gay people are. I do not have a clue about that. My point was that several of Hlavac’s assertions were dubious and the one that 93% of men and 97% of women have been getting married since the Middle Ages is not just dubious but demonstrably wrong. So is his implied assumption that most of those who don’t marry are Gay.

As for nothing to back up our “attacks”. Well there are the English marriage records which I referred to which undermine his 93% and 97% assertion. To say nothing of the fact that there are indeed millions of spinsters in the US despite his assertions. And marriage rates do change over time.

Leave A Comment

All comments reflect the opinions of commenters only. They are not necessarily those of anyone associated with Box Turtle Bulletin. Comments are subject to our Comments Policy.

(Required)
(Required, never shared)

PLEASE NOTE: All comments are subject to our Comments Policy.