August 18th, 2007
Regular readers of this blog will know I regularly rant on about dodgy sex surveys and why they don’t make sense. Now it seems a mathematician has gone further and argues that sex surveys simply don’t add up.
A report from the International Herald and Tribune cites mathematician David Gale who argues the discrepancy of men reporting more partners than women simply doesn’t add up – and therefore renders sex surveys invalid.
With respect to the professor this isn’t something that’s news to sex researchers – and it doesn’t mean we shouldn’t run sex surveys. I’m glad the health and science pages of the Herald and Tribune wanted to cover this fascinating area, but I’m disappointed they didn’t really seem to know much about sex surveys or sex research as it means the feature’s both outdated and flawed.
A number of factors often explain gender discrepancies in surveys. We know that men often feel a need to exaggerate their sexual experiences due to the pressures of masculinity. They are also far more likely to overestimate rather than underestimate answers. Women, on the other hand know they’re generally judged negatively for reporting sexual experiences and so will downplay their activities. Men are far less likely to admit having sexual problems, while women are far more likely to share their concerns or questions.
There’s also the problem of what constitutes ‘sex’. Although it may seem obvious people define and interpret it very differently so kosher sex surveys always define it and ask about a range of activities people might enjoy to find out more about what we do in (or outside) the bedroom.
There are a number of limitations to using surveys in any research, not just research on sex. That includes people misunderstanding questions, over or underemphasising answers, or completing surveys inaccurately. You may find out how much or how many from surveys, but you won’t get the more detailed answers about how people feel. That said sex surveys can be useful as they’re easier to answer for some people than having to speak to a researcher about their experiences.
Within this feature there’s an assumption that men who respond in sex surveys are the partners of the women in such surveys and that often isn’t the case. Where you do survey couples you tend to find much greater concordance in answers. In all likelihood men do have more partners during a lifetime because culturally it’s more acceptable for them to do so. The issue is whether they have significantly more partners and although you may find in research it looks like men report more encounters than women it may not be the case that they’re having significantly more sexual relationships.
The flaw in the do gender-differences-in-surveys-add-up argument is that people apply cultural models to a mathematical problem. By that I mean they assume that sexual relationships are monogamous. They don’t account for people having more than one partner because of cheating, an open relationship or having a threesome, foursome or moresome.
You can see this error played out within the Herald and Tribune piece right from their intro which makes massive sweeping evolutionary statements: “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”.
For the record everyone doesn’t know men are promiscuous by nature (some men may have multiple partners but you can’t always attribute that to nature or call it promiscuity). This reasoning assumes the only reason women have sex is to bear kids and the only reason men do it is to get women pregnant. It denies lesbian, gay and bi experiences and it also denies the reason most people have sex most of the time – which is for pleasure (hopefully). It also excludes forced sexual experience.
Sex surveys for the most part measure attitudes, experiences and behaviours. You can apply an evolutionary explanation for the data you find, but you can’t present all sexual behaviour as being evolutionarily based. In fact the good thing about sex surveys is they show how our sexual attitudes, experiences and behaviours change or remain consistent which is evidence that our sexual relationships are driven by something far more complex than simple evolutionary theory.
Sex researchers account for gender differences for the reasons given above, and we often use a wide range of questions to assess what’s going on rather than making a competition between the sexes – it’s only when research gets to the media that this happens. What’s more interesting about sex surveys is you frequently find more similarities between genders than differences, although again this message never makes its way to the media.
There are problems with sex surveys currently. They’re overused as a PR tool – a means to promote products that are nothing to do with sex. They’re overused within media and journalists give far too much importance to statistics. And they’re misunderstood as a method. They can tell us some things about sex, but not everything – and they’re not the only method in a sex researchers toolkit.
So I’m afraid I have to disagree on this occasion and say that there are discrepancies in sex surveys – but not all sex surveys show massive gender inequalities, and not all sex surveys show men having more experiences than women. The problem with sex surveys is not they don’t work mathematically, but that most people working in media simply don’t understand how they work and what they do.
If you would like to know more about sex surveys here’s a useful beginners guide – Sex Surveys 101.Tweet