Perfect news:fluff ratio = 0.0

Plenty of other bloggers have pointed out how the Mail website these days exists solely as an excuse to post pictures of scantily clad women to boost clicks, and today, their semi-nude supermodel output intersects with their science output: Beauty summed up: To tell if a woman’s really attractive, it’s all in the figures (possibly NSFW, assuming your workplace would frown on gratuitous images of Kate Moss in her underwear).

Mathematics may not sound sexy, but the right measurements determine sexual attraction within milliseconds for men, it has been revealed.

New Zealand anthropologist Barnaby Dixson studied what the sexes found attractive in a partner across cultures and over history using a method of eye tracking.

The usual suspects of personality, breast size and weight apparently do not figure.

Instead, Dixson found the same formula for what men favoured in women came up almost every single time: a waist-to-hip ratio of 0.7.

Why did 0.7 come up “almost every single time”?* Well, because there were only two options.

The eye-tracking experiment (Eye-Tracking of Men’s Preferences for Waist-to-Hip Ratio and Breast Size of Women, 2009 – link behind Athens/paywall, again NSFW) used six computer manipulated images of a nude woman (the same model each time, with her body adjusted in Photoshop), described as “small, medium and large breasted”, with waist-to-hip ratios (WHR) of 0.7 and 0.9. These were the only two used in this experiment.

The claim that he’s studied attractiveness across cultures is fair enough – he’s previously investigated how attractive Chinese, Papua New Guinean, Cameroonian, American and New Zealander people found women with a wider range of WHR (though the Papua New Guinean study had a questionable methodology; he used before and after photos from cosmetic surgery rather than manipulated images) – but the results aren’t as clear cut as this makes out. Chinese and American men preferred a slimmer WHR of 0.6, Cameroonian men 0.8 – only New Zealanders found 0.7 the most attractive. True, 0.7 appears be some sort of median here, but it certainly didn’t come up “almost every single time”. How they tested what men found attractive historically, I can’t find out – presumably it involved looking at the waist to hip ratios of women in paintings.

Personality did not come into this study or any other, for the obvious reason that personality is untestable and unqualifiable. The subjects were simply shown, without context, a photo of an anonymous nude woman.

Breast size did come into this study, and it actually turned out that (unsurprisingly), it did have an effect on attractiveness score.

Finally, it’s rather disingenuous to say weight did not figure. OK, true, the subjects were never told the woman’s weight. However, when you’re talking about an experiment that involves looking at the amount of fat on a woman’s body, the underlying message is about nothing but body weight.

Oh, and while eye tracking was used in the experiment, that was only in a separate part of the study to do with breast size. The only part that had to do with the waist-to-hip ratio was when, at the end, men were asked to score the image on a 6 point scale. The Papua New Guinean paper mentions offhandedly that “changes in BMI were not related to men’s judgments of attractiveness”, but later says that they had to control for BMI for the results to be statistically significant.

Sure enough, men did on average rate the slimmer/curvier image as having a higher attractiveness. That doesn’t prove anything much however beyond the fact that on this particular woman, the “configuration” of body parts which scored highest was “medium breasts”, with the lower waist-hip ratio adding roughly one extra point to the score (~4.3 for WHR of 0.7, ~3.3 for WHR 0.9). We don’t know how the exact arrangement of scores that led to this, however – did everyone consistently rate the WHR 0.7 image higher, or did a few anomalous high votes drag it up? Did a few anomalous low votes drag the WHR 0.9 image down? Was there a wide spread for both images?

What would be a more accurate way of phrasing this result might be “On average, men tend to consider a slimmer waist more attractive than a broader one, but only while taking other factors into account.” Of course, that wouldn’t be especially newsworthy, and the Daily Mail has a quota of Jessica Alba pictures to post, damnit!

And Dixson said other factors weren’t taken into account by men when they saw a woman with that waist-to-hip ratio.

Except according to his very own paper, they are – there’s a graph (fig. 2 – NSFW due to embedded images) demonstrating how no matter what the woman’s waist-to-hip ratio was, men took breast size into account.

Dixson, who explained his finding on Australian television also said men were wasting their time pumping iron in the gym as women invariably preferred a leaner, less muscle-bound physique.

Either Dixson or the Mail needs a dictionary. “Invariably” does not mean what you think it means. If this was true, the world would be filled with single heterosexual bodybuilders unable to find even one partner. Even the basic claim behind it isn’t true either – according to his paper “Human physique and sexual attractiveness in men and women: a New Zealand-U.S. comparative study“, “women in both countries rated mesomorphic (muscular) and average male somatotypes as most attractive”.

For once, I don’t think this is the papers misrepresenting a study. In this case, I think the misrepresentation comes from the scientists themselves wanting a bit of media attention. And of course, they’ve got it.

* Also, “0.7” is not a formula but a constant. No, Daily Mail, it does not make you sound smarter, it makes you sound stupid.

  1. #1 by wickedday on Thursday, 26th August 2010 - 14:20 GMT+0100

    Um, also, what possible scientific value does measuring attraction have? It’s so heavily dependent on culture, time period and personal quirks that concocting an average tells you absolutely zip about the proclivities of any given person. Averages: often stunningly unhelpful! Like the old joke about how your average human being has one testicle, one breast and slightly less than two legs. (Theoretically true, but completely useless for e.g. trying to accurately describe the human race.)

    Also another testament to how nobody in the media seems to have heard of socialisation. When it’s basically taboo to admit to an attraction to people outside the current beauty norms, people a) self-delude, and b) lie. Noooo, it’s all about SCIENCE!

    And surely pretty crap science at that. Asking people whether they prefer X over Y, when X and Y are two arbitrary values on a rather wide spectrum, is going to get you next to no useful data, surely?

  2. #2 by wickedday on Thursday, 26th August 2010 - 14:25 GMT+0100

    Also, fuck yeah heterocentrism! If they’re genuinely interested in measuring what people who are attracted to women are attracted to, why aren’t they including lesbians & bisexual women in the survey group?

    • #3 by atomicspin on Thursday, 26th August 2010 - 16:52 GMT+0100

      Because homosexuality and bisexuality basically destroy the entire theory behind evolutionary psychology; if this was entirely biological, and not due to socialisation like you say, lesbian and bisexual women wouldn’t express any particular interest in so-called “fertility indicators”. In fact, a study by someone else found – again, on average – that lesbian/bisexual women also expressed a preference for a figure with a hip-waist ratio of 0.7, meaning that is almost certainly a cultural preference, not an evolved one.

      Evo-psychologists like Dixson tend to ignore the gay elephant in the room and hope people don’t notice.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: