Most sex surveys you read about are really just marketing exercises masked as science. Company X hires a private firm to survey 1,000 people to find out how many of them use Product Y, and throws in a few questions that are designed to be catchy for journalists (for example, a car insurance company asks men how many of them buy Valentine's Day gifts for their cars;  you know, just for one example).

The National Survey of Sexual Health and Behavior (NSSHB), conducted by a team of researchers at The Center for Sexual Health Promotion based at Indiana University, with funding from Church & Dwight (the company that brings us, among other household products, Trojan condoms), should not be confused with a marketing survey. While the company that paid for is obviously doing so not out of a love of science, but a love of money, the fact is that the research and the data are as good as this sort of data gets, and the survey itself, the largest nationally representative survey of sexual behaviors conducted in the U.S. since 1994, in many ways is a significant leap in sophistication and specificity in terms of what we mean when we talk about sex.

In total 5,865 people ages 14 to 94, completed surveys which included questions about who they have sex with, what kind of sex they are having, where and how often they have it, and how much pleasure and/or pain they get from it. They also collected data about age, gender, race and ethnicity, self-identified sexual orientation, education, income, and more.

The survey was conducted through a company that uses technology not unlike ratings services like Nielsen to produce and deliver participant pools that represent the general population across a number of demographic factors based on U.S. census data. Of course they don't represent all of us, nor do they represent any of us in particular. But with such a large and carefully generated sample, the snap shots the data provide are detailed and, as far as numbers go, rich and robust. Participants filled out the survey online. Adults received no compensation for their participation, adolescents (who were only able to participate with parental consent) were given $5. Some of the findings include:

Not All Teenagers are the Same
Much of the data highlights an important factor that public health educators know, but the general public and certainly the media tend to forget. Not all teenagers are the same. By collecting data from adolescents age 14 and up, the survey foregrounds the transition moment between the ages of 16-17 and 18-19, where teenagers show a huge leap in both kind and frequency of sexual activities. To offer one example, twice as many women between 18-19 report having had oral sex, vaginal intercourse, and anal sex as women who are 16-17. The data also let us see that condom rates also drop during this transition, in some groups by as much as 50% between the ages of 14-17 and 18-24.

Sexual Orientation and Sexual Activities are Not the Same
This is another one that educators know well, but bares repeating. Sexual activities are not the same as sexual orientation, and in this study, which didn't attempt to over sample from sexual minority communities and as such has predominantly heterosexual participants, we see that far more people reported same sex sexual activities than identified as gay or lesbian. For example:

• While 6% of male participants identified as gay, 12% received oral sex, 11% performed oral sex, and 8% had anal sex with a male partner.
• Only 1% of women who participated in the survey identified as lesbian, but 15% received and 16% performed oral sex with a female partner.

 

The Sex Are We Having
While the survey was unique in offering people the chance to describe multiple sexual activities within one sexual encounter, it still focused on four kinds of sexual activities: masturbation (solo or partnered), oral sex (giving, receiving, both), vaginal intercourse, anal intercourse. In terms of lifetime experience, here are a few of the numbers:

• 90% of men and 74% of women report solo masturbation.
• 77% of men report receiving and 75% report performing oral sex with a female partner.
• 76% of women report receiving and 73% report performing oral sex with a male partner.
• 31% of men and 34% of women report engaging in anal sex with an opposite sex partner.

 

Who We're Having It With
We're mostly having sex with people we are in relationships with, but not only. Describing sex in the past year:

• 53% of participants were with a romantic partner
• 24% had a casual sex partner
• 9% had sex with a "new acquaintance"

 

Most of Us Lack Friends with Benefits (But Age Doesn't Matter)
Overall, 13% of men and 11% of women reported having sex with a friend in the past year. While it was most common for women between 18-24 to report having sex with a friend, for men there wasn't a direct correlation between age and likelihood of having sex with friends. So while "friends with benefits" seems to be far from the epidemic we hear about on talk TV and radio, when it does happen, it's not jus the kids who are doing it.

Sex for Pay or Exchange
Particularly exciting was the choice to identify the exchange of sex for money or other things as something worthy of studying in a general survey on sexual behavior. The researchers chose to use the term "transactional sex" to describe sex where one or the other partner paid or exchanged something for sex. Overall 3% of participants described having sex that they either got paid for or paid for, or were given something for, or gave something for. More on this here,

Where We're Having Sex
The most common place to have sex was one's own home (76%), followed by a partner's home (14%). Only 1% of participants reported having sex in public within the past year.

How Much We Enjoy Sex, and How Much We Think Our Partners Enjoy It
Overall 91% of men and 64% of women report having an orgasm at last sexual encounter. At the same time, 85% of men and 92% of women believed their partner had an orgasm the last time they had sex. These data can't be directly compared given the small group of people who have same sex partners, but the discrepancy, particular among men, is notable. Clearly more men think their partners are having orgasms than they are. Of course with only numbers to give you, the most interesting stuff is left for all of us to talk, moan, and scream dramatically about.

Sex Hurts
Across age groups 6% of men and 30% of women reported some pain at their last sexual encounter. These percentages are striking both in terms of the sheer number of people they represent who are experiencing pain during sex, and the significant (but not surprising) gender discrepancy.

Not Just Girls Going Wild
Researchers asked participants about their use of alcohol and marijuana during sex. Interestingly, the image of out of control young people being the ones to rely on drugs or alcohol during sex wasn't borne out completely. If you sum the data provided, men and women 50-59 were most likely to report drinking alcohol or smoking marijuana at their last sexual encounter. Which isn't to say young people didn't also partake, but it's worth noting that the vast majority of men and women reported not drinking or smoking marijuana at their last sexual encounter.

We All Want Connection AND Hot Sex
Data from the survey challenge the traditional gendered characterization of men wanting hot fast sex and women wanting a connection. Men who were having sex with a romantic partner were more likely to report orgasm than those who were having sex with a friend or casual partner. The same was actually true for women too, although the relationship was not as strong.

On the flip side, both women and men were more likely to report having an orgasm if the sex involved multiple activities. This relationship was stronger for women.

Significant Advances
Before getting to the strings part, I feel it's worth pointing out a few great things the researchers did in this survey. I'll be going into more depth in later posts, but just a bulleted list here:

• collecting data in a way that allows us to distinguish sexual activities within a sexual event, and sexual behaviors from the partners and relationships we have them in
• considering race and ethnicity outside of the context of sexual risk, attempting to situate it in a discussion of sexual health and pleasure
• asking about transactional sex in a general survey of the U.S. population
• attending in both the production of data and the write up to the ways that sexual health services are delivered, and how information might best be presented to have a positive impact on sexual health service provision broadly defined

About Those Strings

There are always strings attached to funding. There are strings attached to funding from the government, from non-profits, and from for profit companies. This is the reality of doing large scale research like this (and the reason why some choose not to do research like this at all). Instead of discounting all data that comes from research funded by organizations or companies with obvious or hidden self-interests, it makes sense to think about what sorts of influence funders, as well as the researchers, have on the data they produce, and then to try and understand the data in context. Here are some concerns with considering in relation to the NSSHB.

There are theoretical underpinnings which we should watch as, despite what most quantitative researchers will tell you, why a person goes looking for answers will influence the answers they come back with. Much of the data is presented within a context of normalization. The theory goes that people want to know if they are normal. And if researchers can give them answers they'll feel better. The problem with this is two-fold. First of all, the idea that curiosity about others sex lives is inherent, something born into us, is absurd. Most people want to know what Oprah's going to give away on her next show, or what Lady Gaga is going to be for or against next. Were we born wanting to know these things? Of course not. We may be born with inquiring minds (although that's debatable) but the focus of our inquiry can hardly be understood outside of the context of socialization and the intense mediated lenses through which we experience so much of our lives. Secondly, the theory supposes that the data they have collected represents normal, or a norm, that is the one that people should judge themselves against. This is a value judgment, not a scientific one. I would argue that, no matter how well intentioned, the idea of telling people they are normal is always counter productive in the end. It doesn't make people feel better (perhaps momentarily it does) and believing your own data to be normal is much like believing your own press. It makes you lazy, it makes you stop thinking. Good data should make you think more. This is damn good data, and deserving of better justification.

In more than one of the introductory pieces in the journal, the research of Alfred Kinsey is invoked as the beginning of this whole systematic enterprise of learning about what people do sexually. It's worth taking a moment to note how fundamentally different Kinsey's enterprise was from the kind of data collection that the NSSHB represents. The bulk of Kinsey's data (perhaps all of it, I'm not completely sure) came from face to face interviews. The interviewers were painstakingly trained and were able to conduct interviews without a script. They were real people in a room. Data collected by a computer should not be discounted and the information need not be diminished. But neither should we forget that how we share stories and who we share them to matters. Data that Kinsey and colleagues spent years collecting were here collected in three months in the spring of 2009. It seems silly to pretend that the way you ask a question, and what you ask, doesn't impact the kinds of answers you get.

Finally, there are those financial strings, being pulled by Church & Dwight, the makers of Trojan condoms, who also happen to be doing a major push on their new line of vibrators. Last Friday I listened in on a media teleconference where three of the primary researchers described the survey and took questions from the media. Most of those asking questions (at least for the first part which I was able to stay for) were from large mainstream news companies; the New York Times, Time Magazine, Reuters, AP, etc. And the most common theme was about conflict of interest. It was encouraging to hear so many journalists start with questions about the funders role in the development of questions and the final reports. The researchers responses were less encouraging. They were emphatic about two things. First, that their funders didn't push them in any directions in particular, and second, that the corporate-academic collaboration they have enjoyed with Trojan over the past several years should serve as a model for more collaboration with corporate partners.

The fact is that the fit for a company like Trojan with this kind of work is such a good one, it becomes hard not to hear every report of new data as a template for a marketing campaign or press release. A company that makes products designed for sexual pleasure and health funding research on how people already experience these things and how they can experience them more makes sense. But the existence of this data helps all companies that make sex toys and condoms. The difference is that the Trojan name is branded to every piece of writing and (they hope) media coverage from now until the next major survey comes along (perhaps another 20 years). Of course who funds a study has influence on the results. So does who conducts the study. With the ever growing body of clinical and theoretical writing about the influence of funders on data collected using their money, the response of the Indiana University research team can only be characterized as naive or disingenuous. Either way, no data is perfectly "clean". It's all biased. What seems important is to talk about the bias, and try to understand the data in it's proper context. Instead of sounding defensive, as at least one member of the Trojan advisory board, who didn't appear to be involved in the research but was on the call anyway, sounded.

All research is imposition, and arguably there is no such thing as purely objective data. But much should be said for intention and execution when it comes to researching sexuality, and on those counts the team who have brought us the first major survey of American sexual behavior in almost 20 years deserve much congratulations and gratitude.