Do Men Lie About Sex? Do Women?

Superb New York Times science journalist Gina Kolata addresses a question that has bothered me for years. On sex surveys men typically report having twice as many sexual partners than women do. How is that possible?  Kolata reports:

Everyone knows men are promiscuous by nature. It's part of the genetic strategy that evolved to help men spread their genes far and wide. The strategy is different for a woman, who must go through so much just to have a baby and then nurture it. She is genetically programmed to want just one man who will stick with her and help raise their children.

Surveys bear this out. In study after study and in country after country, men report more, often many more, sexual partners than women.

One survey, recently reported by the U.S. government, concluded that men had a median of seven female sex partners. Women had a median of four male sex partners. Another study, by British researchers, stated that men had 12.7 heterosexual partners in their lifetimes and women had 6.5.

But there is just one problem, mathematicians say. It is logically impossible for the mean number of partners for men to be different from the mean for women in any given population with equal numbers of heterosexual men and women….

One possibility is that the numbers could balance out if men frequently bought the services of prostitutes who are not generally included in sex surveys. However, I think it's far more likely that men and women both lie a lot about sex. Men who want to be seen as "studs" exaggerate, and women who want to avoid being regarded as "sluts" underestimate. 

Whole Kolata article here.  

Comments

[de stijl]

Sluts, like Ron Paul supporters, are more technogically savvy, thus they are less likely to have landlines, ergo they are under-represented in polls.

[Chris S.]

Yes, it's impossible for the "mean" to be different, but not the "median," which is what the U.S. study reports. Of course, for the median to be so different, some women would need to be really freaking busy racking up dozens of partners.

[tarran]

Curse you de stijl!

Water came out of my nose. It hurt!

[GILMORE]

Sans disclosure, this must be treated as a Ron Bailey-imposter post.

But ron, you dont trust the *data*??! But its DATA! Isnt that like more sciency than common sense assumptions?? I am dissapointed.

I myself lie to all pollers about everything

[Chris S.]

Oh, it's also possible that men are having lots of sex with people on their death beds, thereby racking up new partners who don't figure into this statistic.

Yup, that's it. Lots and lots of death bed sex.

[Episiarch]

Men lie about their number of partners, I'm sure, or may "fudge" the numbers by including chicks that they only got to second base with, etc.

Chicks lie about their number of partners, I'm sure, or may "fudge" the numbers by not including blowjobs, or other women, etc.

All of which means jack shit. And Jack left town.

[Dan T.]

I really don't see the point in taking a poll where you ask people about something taboo like sex or drug use where people have what you might call "social incentive to lie" and results cannot be verified at all.

[Jose Ortega y Gasset]

"One survey, recently reported by the U.S. government, concluded that men had a median of seven female sex partners. Women had a median of four male sex partners."

I take it the survey did not include Wilt Chamberlain or anyone who lived in my freshman college dorm.

[Abdul]

Wilt Chamberlain has disproportionately affected the average for all men. Take him out of the poll, and we go down to 3.7 partners on average.

[semm11]

right, but the study says median, not mean. So the results could be logically consistent and not based on subjects lying. If so, there need be only some small segment of 'hoes' out there letting the guys get a higher median without dragging the ladies median down.

I think the mean vs mean difference would be very interesting to look at.

[John in Nashville]

Other commenters have noted the distinction between the median and the mean, which the Kolata article seems to conflate. A few promiscuous women (whether paid or unpaid for their services) having sex with a great number of men would explain the difference in the median.

BTW, Wilt Chamberlain would have no more effect on the median than any other individual on the high side thereof.

[Lamar]

Perhaps the survey had separate categories for "Men", "Wussy Men" and "Girly Men."

[John in Nashville]

It is logically impossible for the mean number of partners for men to be different from the mean for women in any given population with equal numbers of heterosexual men and women.

I am ordinarily astute enough with numbers that I can count all the way to twenty-one (when I'm naked), but is that statement true when one or more of the women are virgins?

[Bee]

Perhaps men are sleeping with other men.

[Chris S.]

I am ordinarily astute enough with numbers that I can count all the way to twenty-one (when I'm naked), but is that statement true when one or more of the women are virgins?

Absolutely. The distribution of partners is irrelevant for the mean.

By the way, might homosexual sex account for this whole mess. That would explain quite a bit without getting into reporting biases.

[Chris S.]

Oops, beaten by bee.

[thoreau]

John-

Look at the math this way. Say there are 100 men and 100 women. On average, a man had sex with 4 women. This means that 400 distinct male/female pairings occurred. It could have been that there were 4 men who each had 100 female partners, it could be that there were 100 men who each had the same 4 partners, it could be 50 men with the same 8 partners, whatever. However it happened, 400 male/female pairings occurred.

Now look at the women: Between them, those women had to form 400 distinct female pairings. Whether 4 women each had 100 male partners, 100 women each had the same 4 men, or whatever else, between them those 400 women managed to enter into 400 distinct male/female pairings.

Which means 400 distinct male/female pairings, divided by 100 females, means an averge of 4 partners per female.

Change the numbers and the result is still the same.

[JF]

I remember a study about this a while back. The men and women involved were asked to answer on the number of their sexual partners. They were told one of three things: that their answers would be anonymous, that they wouldn't be anonymous, and that they were hooked up to a polygraph(they weren't really). In each case the men's answers remained more consistent than the women's answers.

This suggests to me that it's the women underestimating the number of partners rather than the men overestimating.

[Mo]

Bee and Chris S.
The survey specifically asked for opposite sex partners.

[relaxing]

What happens to the figures when "pairings" become more than 2, I wonder...?

[tarran]

Thoreau, that's true, but the male and female populations aren't equal.

I believe that there's something like 104 women for every 100 men.

Thus, you would expect to see a difference of a few percentage points in the mean number of partners reported by each gender.

[BG]

I am ordinarily astute enough with numbers that I can count all the way to twenty-one (when I'm naked), but is that statement true when one or more of the women are virgins?

Yes. For every male sexual partner a women had, there must be a female sexual partner a man had. So regardless of how the sexual partners are distributed, the total number of sexual partners for each population must be the same. And if the number of members of each population is also the same, then the averages must be the same.

Of course, that assumes the members of one population can only partner with the members of the other. If, for example, American guys are getting a disproportionate amount of foreign action, that could explain differences in the means between US males and US females.

Also the disparity could be explained by different standards for what counts as a "sexual partner" as well as flat out lieing.

[Reinmoose]

It's possible, I suppose, that some people had so MANY sex partners that they really couldn't remember how many they had, and thereby cause the numbers to go off. That they would go off consistently in one direction is another matter, but one to be explored regardless.

[thoreau]

tarran-

Fair enough. But that's only a few percentage points, not a 50% discrepancy.

[ChrisO]

The way to make the numbers work is that a relatively small number of women had a very large number of sex partners, thus statistically making up for the larger number of women who had fewer partners on average than the men.

Self-reporting is obviously prone to distortion, but my anecdotal experience tells me that the above scenario has some truth to it. A promiscuous heterosexual man has to work much harder to get laid than a promiscuous hetero woman. And that's not a value judgment of any sort--I don't buy into the Christian guilt bullshit.

[BG]

Thoreau, that's true, but the male and female populations aren't equal.

I believe that there's something like 104 women for every 100 men.

True, so we would expect to see a slightly higher mean for males than females. But I think if the recorded average was even 1 more sexual partner for males than females, it would be to much to be entirely explained by the different population sizes.

I guess someone could also look at death rates; if enough people had a bunch of sexual partners who are all dead now, that could throw off the averages. Although I think that would tend to favor a higher female average as women live longer, and its probably not enough of a thing to make a big difference either way.

[BG]

ChrisO

You're talking about how to make the median work right?

Also, where can I meet these hyper-sexual women?

[Garth]

I blame rape. If the much ballyhooed figure of 1 in 6 women having been raped is true, then surely the women are not including these in their numbers but the rapists surely are.

I also suspect that handjobs and blowjobs (especially during high school and even college) are added to the boys scores and not the girls.

Finally, I suspect lots of women do not include one night stands since that person is hardly considered a partner.

[MikeT]

My money is on the women lying about how many partners they've had since this is over the course of their lives. For a man who has no moral or religious issues with having sex outside of marriage, 12 partners is a realistic figure. However, for most women to have 6 and most men to have had 12 suggests that either a lot of men are sharing part of the female population or the women are underreporting.

[BG]

If the much ballyhooed figure of 1 in 6 women having been raped is true...

Holy shit, that a pretty fucking big "if"! I've never even heard of that statistic and I watch more Law & Order SVU than I probably should.

[Franklin Harris]

There are a couple of explanations for this. First, men count every orifice as a "partner." Second, men accidentally count some partners multiple times because they don't remember their names.

[fyodor]

I wonder what else has bothered Ronald Bailey for years?? 🙂

[slugger]

Perhaps the discrepancy can be explained by people counting only achieving orgasm as sex. If the lady is left high and dry, you don't count.

[fyodor]

It's my firm belief that people are most likely to lie to themselves. Meaning that what then comes out of their mouths is technically an "untruth" rather than a "lie" because they believe it to be true. This also helps explain why people would "lie" to an anonymous survey.

[ChrisO]

Also, where can I meet these hyper-sexual women?

Having known a few in my young and single days, you're probably better off not knowing, unless you like drama and heartbreak.

[ChrisO]

You're talking about how to make the median work right?

I don't know. I went to law school so I wouldn't have to deal with all of that mathematics crap.

[miche]

Perhaps the discrepancy can be explained by people counting only achieving orgasm as sex. If the lady is left high and dry, you don't count.

My number just dropped in half.

[Jose Ortega y Gasset]

And people wonder why libertarians don't get invited to parties... and apparently don't get laid often.

[Son of a!]

She is genetically programmed to want just one man who will stick with her and help raise their children.

This is false.

If we're going to assign motives like "I'm gon' git me some" to the interests of genetic diversity, then let's do it equitably.

Women, being human, have as much self interest in genetically-diverse children as men do. While it may be true that women are more physically vulnerable during pregnancy, men have as much self interest in the survival of their progeny as women. It therefore stands to reason that men and women both want to help raise children that they know to be their own.

Men can better their odds by having a mate whose progeny are numerous and his (which he can help ensure by guarding her jealously), with numerous dalliances on the side to add diversity.

Women can better their odds by finding a mate that will raise all of her children, and by cuckolding him as often as possible.

The idea that genes program men to be promiscuous but those same genes somehow program women not to be promiscuous is silly and a little misogynistic. I recommend Promiscuity for a good look at the subject.

[Bee]

Righty-o, Mo, I did not RTFA before posting. I have remedied that.

I think the raw data is a pack 'o lies, but the fact that it's a pack 'o lies is obviously pretty interesting in itself. Why on earth are people so mentally pretzelled by sex? Is is narcissistic to lie in an anonymous suvey? Psychopathic amorality? Plain old pathetic?

[dhex]

the big unanswered question is: did this survey include real doll owners?

"I think the raw data is a pack 'o lies, but the fact that it's a pack 'o lies is obviously pretty interesting in itself."

indeed. it is fairly strange that someone would brag to an anonymous survey. or for that matter, lie to it on the downside as well.

[Rhywun]

I really don't see the point in taking a poll where you ask people about something taboo like sex or drug use where people have what you might call "social incentive to lie" and results cannot be verified at all.

Agreed. I don't trust any figures that purport to represent the incidence of homosexuality for the same reason.

[BG]

Having known a few in my young and single days, you're probably better off not knowing, unless you like drama and heartbreak.

Well, that really is a tough call. I'm not a big fan of heartbreak or drama but I really would like to get laid more.

I don't know. I went to law school so I wouldn't have to deal with all of that mathematics crap.

I asked because your theory would explain a disparity in medians but not in averages.

[Mo]

For a man who has no moral or religious issues with having sex outside of marriage, 12 partners is a realistic figure.

MikeT,

Oh yeah because men, never, ever exaggerate the number or quality of women they've been with. It's the rule of 3 for guys and girls. Add 3 to the girls, subtract 3 from the guys.

[Chris S.]

Oh yeah because men, never, ever exaggerate the number or quality of women they've been with. It's the rule of 3 for guys and girls. Add 3 to the girls, subtract 3 from the guys.

Christ, that's brings me to -2. That hurts.

[fyodor]

dhex and bee,

You two either didn't read my 1:41 post or, worse yet, didn't appreciate it!!

[Stevo Darkly]

I've thought of yet another possibility that would explain the variance in the reporting of opposite-sex partners -- and it would not require deliberate lying by men or women.

Suppose a large proportion of the heterosexual encounters reported by men include blowjobs received from what they thought were women, but were actually, unbeknownst to them, men.

That would inflate the heterosexual encounters reported by men, without actually increasing the number of men-women sexual encounters.

[dhex]

yeah i mean fyo i tend to believe everyone knows what the score when it comes to what they're doing (or not). even a really extreme case like a ted haggard - i can buy that he'd basically go on binges and totally give up trying to control himself for meth punctuated 24 hour bursts or whatever, but that he didn't know he was doing what he was doing is pretty fucking incredible as far as claims go.

on the other hand, self-fibbing can be a pretty strong drive for some so i can see it carrying over. perhaps it's part of keeping their story straight, i know not.

[Jeff Smith]

The thing that struck me about this is what a useless article it is. That is consistent with the general pattern that everytime the NYT (or its shared progeny, the IHT) writes about something I know about, they get it wrong. In this case, it is not so much wrong as meaningless.

As already noted, the article confuses means and medians (something one might not expect from a "superb" science writer). It does not even mention the care that is taken in the better of these surveys (e.g. the one that underlies the books by Laumann and Michael) to encourage truthful revelation. It does not raise the issue of prostitutes (who were in the population for that survey and likely are for others). It does not discuss the care that goes into defining what constitutes as "sex" in the course of the survey or how that definition is tranmitted to the respondent. All of these are interesting issues and would make for an interesting science article. None appear here; indeed, the article seems to encourage the reader to ignore all survey data, which is not a good plan, even in sensitive areas like sex research.

The Laumann and Michael books are called "The Social Organization of Sexuality" and "Sex in America". One has lots of footnotes, the other does not.

I await the article that explains to my why people worship, let alone read, the NYT.

[lunchstealer]

Wait, I've accounted for the whole shift.

The male number is 12.7, and the female number is 6.5, so we've got to account for a 6-partner deficit? It's clear that all those men are counting "Rosy Palms and her Five Sisters" as sexual partners. Take out 'batin and you've only got a 0.2 partner deficit to account for, and that's just sampling error and the aforementioned female/male ratio of 1.04.

[Garth]

Lunchstealer,

you mean Mother Thumb and her four lovely daughters don't you?

[Syd]

One survey, recently reported by the U.S. government, concluded that men had a median of seven female sex partners. Women had a median of four male sex partners.

No problem. They didn't specify that the partners had to be human.

[Syd]

Or men in the military have more intercourse with other nations.

[Joe R.]

Kolata either doesn't understand the difference between median and mean, or she deliberately chooses to confuse the two. Either way, it prevents her from being a "[s]uperb...science journalist".

[the innominate one]

The idea that genes program men to be promiscuous but those same genes somehow program women not to be promiscuous is silly and a little misogynistic.

Your overall assertions about women's sexual strategies could be right, but not necessarily for this reason. A particular gene doesn't have to have the same effect in both sexes, thanks to epigenetic interactions. Not to mention, some genes are usually only found in men (e.g. testis determining factor, found on the Y chromosome).

@ lunchstealer @ 4:00 pm:

maybe men are counting each hand as a separate partner.

all:

this underscores the importance of knowing the variance as well as the average

[Clemsonuee]

12.7 for guys? I have a friend who did that in the last 2 years; and I live in the Bible Belt of the Bible Belt.

Of course I even him out by being a virgin; but still given that I seem to be a distinct minority I would have guessed a higher number.