May 12th, 2009
The papers are abuzz today with the story that women with higher levels of ‘emotional intelligence’ (EI) have better sex lives.
And here we had a classic example of journalists clearly not consulting with an original study and/or misunderstanding a press release. Many papers mistakenly interpreted the story as brainy-women-have-better-sex.
Here are a few gems…
Women with higher intelligence have a better sex life – Hindu Times
Smarter girls have far better sex lives – The Sun
Bad luck bimbos: Intelligent women have better sex, study reveals – The Daily Mail
Perhaps these duff interpretations could be attributed to the Press Association’s pickup, where they also completely missed the point of the study.
So what did the study actually measure?
The research was a postal survey of female twins, which asked a wide range of questions about health, life and relationships. Among those questions were two asking about orgasm during intercourse and masturbation, and a separate questionnaire measuring Emotional Intelligence (using the Trait Emotional Intelligence Questionnaire – Short Form). A 30-item questionnaire that focuses on emotions, wellbeing, motivation and self control. You can see a sample copy here. (This includes details of scoring and researcher contact information if you wished to use this tool in your own work – please note commercial use of this tool is prohibited).
The researchers compared the women’s reports of their orgasmic frequency with their scores on the emotional intelligence questionnaire. They found women who reported low frequency of orgasm also reported lower levels of emotional intelligence. Researchers concluded emotional intelligence is linked to orgasmic frequency and suggested that measuring EI could be a useful tool when addressing female psychosexual problems. You can access the paper here (not open access I’m afraid).
As you can see this is not a study that measured IQ or linked general intelligence with sexual functioning. So nothing like a lot of the press coverage, then.
What are the problems with the research?
At first glance it may seem there’s very little wrong with the study. After all it’s fairly intuitive to link low emotional intelligence with orgasmic problems. But look closer at the research and questions emerge about how orgasm was measured, and whether the study truly did find that women with sexual problems had lower levels of EI.
The paper opens by claiming up to 30% of women ‘suffer from female orgasmic disorder’ (difficulty in experiencing orgasm). This claim is supported by mentioning the 2000 consensus conference which created the guidelines for identifying and treating female sexual dysfunction. Yet this event has been documented by the British Medical Journal (and many others) as problematic, as it was based around the construction of ‘female sexual dysfunction’ by drug companies and practitioner/academic affiliates.
Most recent research estimates the prevalence of female sexual problems as nearer to 16%. That research is also drug-company funded and so there are some critics who argue the 16% is still a fairly high estimate. However, the idea that 30% of women globally have a clinical problem that prevents them from having an orgasm is certainly questionable – yet this literature has not apparently been consulted in the current study.
While the authors do note in the introduction that psychosexual problems can have a variety of causes, the remainder of the paper does not focus on these multiple factors. Instead it solely focuses on EI and orgasmic frequency.
The press coverage has talked at length of this being an impressive study of over 2000 women. But when you read the paper you realise the response rate to the study was actually only 24% of the overall sample (in fact 8418 women were sent the survey but only 2035 responded to both the orgasm and EI questions).
This raises questions about the representativeness of the participants. The sample was a volunteer database of female twins which does allow for interesting analysis, but also means participants are not entirely representative of the UK population. Such a low response rate also makes it likely those who did complete both sets of questions probably weren’t representative of the study group originally approached. The paper does not make it clear how those who did respond compared with the non-responders, which you usually expect to see when research is reported.
The authors acknowledge their response rate was “relatively low compared with other medical surveys but respectable compared with other sex surveys”. I don’t agree. I think 24% is a very low response rate whatever the study. Most sex studies do not have such low response rates – or those that do are not always published. This low response rate should give us pause for thought over whether we can rely on the study to predict the behaviour of women generally.
The biggest problem with this study, however, is about how ‘female orgasmic disorder’ was constructed and measured. The researchers asked women to report on two questions relating to orgasm. ‘Overall, how frequently do you experience an orgasm during intercourse’ and ‘Overall, how frequently do you experience an orgasm during masturbation’. Respondents could answer on a 7 point Likert scale where 1 was ‘never’ and 7 was ‘always’. Those who were not in a sexual relationship were asked to think about their past relationships when answering the question.
What’s wrong with those questions, I hear you ask? Let’s go back to 2005 when the research team behind this current study published research where they argued the female orgasm had a genetic basis and orgasmic difficulties were hereditary.
The same measure of orgasm was used then, and I’ll refer you to the criticism I raised at the time:
“Let’s look at those questions in more detail. First off, what does ‘overall’ mean? Is it ‘overall’ in your life at the moment? ‘overall’ across your lifespan? ‘overall’ in your current relationship, or perhaps past relationships? If you don’t define parameters, you can’t be certain what people are responding to. That’s why reputable studies of sexual functioning define a specific period (e.g. the past month, the past year, your current relationship) to direct participants’ answers.
And what about ‘frequently’? Even novice social scientists know to avoid the term ‘frequently’. In this study ‘frequently’ could mean daily, weekly, monthly, or yearly. Participants are left to decide what frequently means to them, and you can be sure it’ll vary across respondents. Again, in kosher sex research you give participants amounts to respond to, rather than using abstract terms like ‘frequently’.
The question on masturbation is confusing because it’s two questions in one. Existing sex research shows women have different orgasmic experiences when masturbating alone or with a partner, and usually are more likely to orgasm through masturbation on their own, than when their partner’s masturbating them. Again, reliable sex research always breaks down questions to avoid confusing participants.
But the main problem with the questions asked, was that frequency of orgasm doesn’t equate to orgasmic dysfunction. Because you don’t have an orgasm ‘frequently’ doesn’t make you dysfunctional. Only this research assumed it did. Those were the questions asked of women to identify sexual problems”.
In any reputable study on sexual functioning you should include a measure of distress. Indeed the authors of this current paper do acknowledge this “we did not use a standardised definition of FOD [female orgasmic disorder] nor did we measure the component of distress”.
This is problematic on two levels. If you don’t ask people if their orgasmic frequency bothers them, you can’t conclude the absence of orgasm is a disorder. One woman who has rare but mindblowing orgasms might be happy. You can’t conclude because women don’t orgasm through intercourse or masturbation they are dysfunctional unless they tell you it is a problem for them.
The second reason this is a problem is the authors had been critiqued four years ago for using this non-standarized measure without an additional distress component, and yet have not apparently rectified this problem in their current study. If I were a reviewer I’d want to know why that hadn’t happened – particularly if it had been so clearly stated as an omission in the current paper.
Most telling is in the close of the paper the authors list all the things that link to female sexual problems they did not study “relationship imbalances, relationship length, intimacy, lack of trust, partner’s sexual performance, partner’s sexual dysfunction, stress and negative body image”. Without including any measure of those within the analysis you cannot completely conclude the low EI scores were not affected by other life events.
As mentioned, the researchers involved in this study had previously published research that claimed a heritability of orgasmic problems. The study itself and subsequent press coverage did cause a lot of concern among sex researchers and therapists.
In the previous study using the twin database around 4000 women were approached. The current paper suggests the database is now around 8000 women. That’s not a problem, but what isn’t clear is whether the women who completed the first survey (which compared orgasmic frequency with heritability) were the same or different women who completed this current study (with the additional EI questionnaire).
If so, there would presumably be a potential bias where some participants would be more familiar with questions than others. It may have no effect on the research but it would have been helpful for the authors to make clear exactly how the sampling worked on this particular survey as compared to previous ones they have run on their pool of participants.
The current research was funded by numerous sources, but the lead author’s PhD (which this paper seems to be drawn from) was funded by Pfizer, who are one of the main players in making drugs for sexual dysfunctions. As you may expect, no journalist noted there might be a potential conflict of interest around research sponsorship as they covered this study.
Should we dismiss the study completely?
While I think the current paper is problematic in methodological terms, and while I would have rejected it for those reasons if I were a reviewer, I don’t think we should be too hasty and reject it completely. Undoubtedly the measurement of orgasm is unsatisfactory in this paper, and that does impact on the reliability of the research. However, the inclusion of the questions around EI I think are interesting and certainly exploring EI in future research on psychosexual behaviour is a worthwhile activity.
My only caveat would be that simply comparing EI with sexual functioning won’t really tell you that much. If you add a measurement of the multiple factors influencing people’s sex lives to an EI questionnaire you certainly would get some interesting and worthwhile findings. The only downside would be you’d have a very lengthy questionnaire!
Why is bad media coverage of this sex study a problem?
The problem with this study is twofold. The research itself could be methodologically stronger, and the media coverage could have been accurate. The muddle over the study measuring IQ as opposed to emotional intelligence is certainly going to mislead the public who would be forgiven for believing smart girls get more sex. Or the media myth that more sex = ‘better’ sex.
The other problem is studies like this and subsequent coverage always presents female sexual functioning through a deficit model. Women’s sexual problems are consistently constructed in the media as caused by a lack of something – hormones or the g-spot to name a couple of popular examples. That means there’s room to suggest to women if they just sorted out their hormone levels or enhanced their g-spot they’d be fine. Unsurprisingly there’s a lot of commercial interests at stake in creating problems and offering solutions. It’s a very negative and disempowering way to talk about women’s sex lives.
I don’t think it’s too much to ask a journalist to read a paper if they’re going to report on it. Not least because they then might identify flaws within the methodology, or potential conflicts of interest from research sponsors. If a journalist doesn’t have the time to read a paper they could always ask a researcher to explain it to them. And that might even give them more story angles and an edge on their competitors.
So what’s the take home message?
It’s simple. The majority of press coverage of this story was wrong. The measure of orgasm in the paper wasn’t reliable, and there are questions over sample size and sponsorship. However, the idea of exploring emotional intelligence within sex research shows promise.
This study did not prove that more intelligent women get more sex. Which is just as well. Because, as all smart women know, it’s never about quantity. It’s all about quality.Tweet