Category Archives: David M. Buss

Why Young White Unmarried and Non-cohabiting Humans in Psychology Classes Have Sex (in America): Part II

Part two of my critique of the new sex study everybody is talking about! Part one is here.

Yesterday I wrote about my methodological concerns regarding the study by Cindy M. Meston and David M. Buss, “Why Humans Have Sex,” published in the Archives of Sexual Behavior.

Today I’m looking at the reasons themselves and discussing some of the conclusions they drew, and some of the conclusions I’d draw looking at the same data.

First of all, I want to dispense with the notion that there were 237 reasons. Quantifying things is an important part of scientific research, of course, and coding data (fitting responses into categories, etc.) is a process that can never be wholly objective. (Somebody at least has to create the categories!) In this case, my criticism arises because the authors indicate that they whittled 715 initial “reasons” down to 237 by eliminating or merging responses that were “too similar” to other responses. That, they claim, produced a list of 237 “distinct reasons”.

I disagree. How distinct is “I wanted to experience the physical pleasure” from “I wanted the pure pleasure,” or “It feels good”? All three of those made the top 15 for men and for women. For the women in the study these were reasons 2, 13 and 3 respectively and for men they were reasons 3,12, and 2.

I’m also not really sure how “I wanted to keep a partner from straying,” is different from “I was afraid my partner would have an affair if I didn’t have sex with him/her” or how “I wanted to get my partner to stay with me” is different from “I wanted to prevent a breakup.” (All were less common than the top 50 but more common than the bottom 50 for both the men and the women in the study.) There are other examples of very similar “distinct reasons” but you get the idea.

Now sometimes researchers use similar answer options to test whether respondents are consistent or not in their reporting of whatever is being studied (personality traits, motivations, what have you). But that doesn’t mean these are “distinct reasons.” Also usually in that case the similar items are scattered throughout the list of items. In this study similar answers are generally clustered together on the list of items raising both a methodological and analytical problem: were people more likely to give consistent answers because they were faced with similar choices clustered close together?

I have not gone through the list of 237 and figured out how many “distinct reasons” I’d come up with, but is clearly fewer than 237. if you’ve downloaded the study (which you can do in PDF form here) it would be interesting for a bunch of us to try it and compare notes!

Second, it’s important to note that most of the reasons were not reasons most of the time for most of the students in the study. Even in the top 50 for both men and for women, most items have a mean score of less than 3. Remember, individuals were asked to indicate whether each of the 237 reasons was true for “none”, “a few”, “some”, “many”, or “all” of their sexual experiences. (This raises a separate methodological issue in that fatigue sets in for lots of survey takers well before they’d have reached their 237th item on the survey!) Those categories were numerically coded 1-5 with “none”=1 and “always”=5. So an item with a mean score of 3 would be true, on average, for “some” of respondents’ sexual experiences. Of the top 50 reasons for women, only the top 8 had mean scores of 3 or above. For men the top 10 did.

Bracketing the methodological problems for a minute, this is interesting because it indicate that people’s self-reported reasons for having sex are pretty varied, and it would seem that few people always have the same set of reasons.

An aside: In a disheartening interview on the Brian Lehrer show I heard Lehrer, who I usually think asks pretty good questions, as “Did we really need a scientific study to show that?” As if the scientific confirmation our hunches about sex is somehow unimportant! When we have hunches about other things, global warming for example, we certainly expect to use science to confirm whether our hunches are accurate. Why would we not do the same for sex?

(Here’s a link to the Lehrer piece. If you listen to the clip, at 7 minutes 51 seconds you can hear Leonore Tiefer, noted sex therapist, researcher and sex educator, call in to raise the same methodological questions I raised in my blog entry yesterday. Leonore, I wish I’d heard the piece in time to site it in yesterday’s post!)

Anyway, given those variations, and still bracketing the methodological issues for a moment, it’s interesting to look at the top 50 reasons with a purpose slightly different from that of the researchers. Their interest was to categorize peoples’ reasons. Mine is to examine what they mean. Remember that these are self-reported reasons. They are self-reported at two levels: first, the list itself is the result of people’s own reports about why they have sex. Second, the ranking comes from people’s reports about the relationship between their thoughts and their behavior. (Our self-reporting is not always accurate but it is interesting because it does represent the stories we tell ourselves about why we do what we do.)

I’m pleased that actual desire to have sex and enjoyment of sex is reported as often as it is (the second and third most common reasons for both men and women in the study). I’m pleased that affection, attraction and love are mentioned as often as they are (all rank in the top 20 for students of both genders). I’m pleased that the students in the study were able to acknowledge and own their horniness (the seventh most common reason for women and men in the study).

That said, there are some troubling reasons in the top 50 for both the men and women students who participated in the study.

For male students in the study, the 34th most common reason was “The person was too ‘hot’ (sexy) to resist” (mean score of 2.17), 38th most common reason (with a mean score of 2.15) was “I saw the person naked and could not resist” and the 42nd most common reason (with a mean score of 2.11) was “The person was too physically attractive to resist.” Framing one’s sexual activity in terms of “being unable to resist” is troubling if it is accurate because it implies a lack of ability to control oneself. In addition, pointing to another person’s characteristics (attractiveness, sexiness) as the cause of one’s own inability to control oneself reinforces the victim-blaming that often surrounds rape, and acquaintance rapes particularly.

That for women students in the study the 49th most common reason (with a mean score of 1.89) was “I was drunk,” certainly doesn’t make the situation easier. And the fact that both men and women cited “heat of the moment” (mean score of 2.84 for the men and 2.89 for the women), “it just happened” (mean score of 2.23 for the men and 2.21 for the women) and “my hormones were out of control” (mean score of 2.20 for the men and 2.11 for the women) is only reassuring in that I’m glad the students were aware of those things as factors in their own sexual behavior.

Sexual motivation is complicated and deserves serious study. This study, though disappointing in many ways, at least has us talking about the need for better research on why people have sex. Understanding people’s sexual motivations, both their conscious motivations and their unconscious motivations, could be important to designing effective public health campaigns, and address the still-too-real problems of sexual abuse in our society.

So lets move beyond this study. Let’s move beyond the trivializing focus on how many people said they had sex to get rid of a headache, or the focus on the number 237 and what the 238th reason might be, and lets start talking about the different ways to design really good sexual motivation research.

(Note: This post is also published on

Technorati Tags: , , , , , , ,

1 Comment

Filed under Archives of Sexual Behavior, Cindy M. Meston, David M. Buss, psychology, public discourse, research, sex, sexuality

Why Young White Unmarried and non-cohabiting Humans in Psychology Classes Have Sex (In America)

That should probably be the title of the new study by Cindy M. Meston and David M. Buss of University of Texas at Austin (PDF).

The study is an important one because it does begin to explore people’s conscious, expressed motivations for having sex, a subject that has been largely ignored or taken for granted in the past. We know much more about what kinds of sex people have than we do about why they have it (or why they think they have it).

And when I read the New York Times article about the study and saw that there was such a wide range of reasons people gave, I was excited: it seemed that the researchers were breaking open some interesting ground and finding lots of diversity.

The news coverage only mentioned a few of the reasons, and I wanted to see the whole list of 237, so I downloaded the study, which you can do here (PDF). I skipped straight to the Table 1 (p. 481) labelled “Top 50 reasons why men and women have sex.” And while I was not at all surprised to find pleasure-oriented reasons among the top reasons for both men and women, I was rather surprised that nowhere in the top 50 for either gender was “conceiving.” Then I read the methods section.

Always read the methods section!

The study occurred in two phases. In the first phase, where the actual list of reasons was generated, the sample was slightly more diverse. It included undergraduate and graduate students in psychology and “community volunteers who were participating in several other ongoing studies in the Sexual Psychophysiology Laboratory at the University of Texas.” (The demographic characteristics of these respondents are not broken down in the methods section of the article so we can’t say much about them.)

These participants were given an opportunity to respond to the following prompt: “Please list all of the reasons you can think of why you, or someone you have known, has engaged in sexual intercourse in the past.” Collectively they came up with 715 reasons, and after the repetitious ones were weeded out the researchers were left with the 237 “distinct” reasons they took into phase two of the study.

It was in phase two, the one with the especially skewed population, that respondents were asked to look at each of the 237 “distinct reasons” and, using each one to complete the sentence “I have had sex in the past because…” to indicate whether that statement was true of “none,” “a few,” some,” “many” or “all” of their sexual experiences. For those who had not had sex in the past (27% of women and 32% of men for whom sexual experience data were available reported not having had sexual intercourse, for example) the instruction was to use that same scale to rate the “likelihood that each of the following reasons would lead you to have sex.” In their published study the authors do not distinguish these responses from those of people who were reporting on actual experience, and while I can’t tell whether that had any major influence on the data, it seems to represent a seriously flawed assumption that guesses about what might motivate one to have sex are the same as reports on what actually has motivated one to have sex.

The respondents in that part of the study, the part where the “reasons” were analyzed for frequency and relatedness, the participants were 1,549 undergraduate students enrolled in Introductory Psychology courses. This kind of sampling is fairly common in academic studies, especially psychological ones, but in this case it makes, I think, a very significant difference in the results. And it means that the title of the study, “Why humans have sex,” and the overall interpretation of the data are rather overstretched. I don’t think that the motivations college students might have for having sex are the same as the motivations that married non-students might have, just for example.

The sample is interesting in its homogeneity in other ways too. Ninety-six percent were between 18-22. And most were not married or living with a sex partner. (Only 4% of the women and 2% of the men were married. Only 6% of the women and 5% of the men were living with a sexual partner.)

So at the time these people were filling out their surveys, they represented a group that is generally young, single or dating students who are focused on their educations and perhaps the beginnings of their careers.

This doesn’t sound like “humans” to me. And it doesn’t sound like a good way to make conclusions about the reasons that people have sex.

While the range of 237 reasons that people have sex might be broad enough to encompass most people’s experience, I don’t think that the priorities or motivations of 18-22 year old college students is representative of the priorities or motivations of, say 30-35 year old people in long-term relationships.

And none of this begins to address the hubris of claiming that any study performed on an American sample represents “humans” in general.

Now, the authors of the study do devote two paragraphs in the discussion section (always read the discussion section) to the limitations of their study. Specifically they mention the fact that their study is based on people’s “expressed reasons” thus can’t account for subconscious or unconscious motivations, and that social approval of some reasons and stigma around others might have affected what people were willing to claim about their own motivations. They also mention the limitations of their sample. They write:

“A third limitation pertains to the relative youth of most of the sample. Reasons for engaging in sexual intercourse undoubtedly differ by age cohort … and would be expected to change over the life span. For example, compared with the student sample assessed in this study, we would expect having sex for reproductive purposes to be endorsed much more frequently among 30 and 40 year olds and having sex simply to gain social status to decline with age.”

They also acknowledge the limitation of conducting their study “within a single culture” and simply say that researchers should explore these same issues in a range of other cultures.

If the authors acknowledge these limitations at the end of their study, why am I harping on them? Mostly because the vast majority of folks who hear about this study won’t have read the study itself. They’ll have read news coverage or commentary that quotes from the body of the results, or directly from the tables, and the authors of the study are not terribly cautious with their language in those sections. They talk about “men” and “women” and “people” but not about “male college students” and “female college students” for example. It will be very easy for these findings to be widely misinterpreted.

Still, the authors of the study raise excellent questions for future research to explore, and those questions acknowledge the limitations of their own work. For example, the authors suggest that future research examine whether or not the 13 major clusters of reasons found in this study are found as primary sexual motivations in other cultures as well, and “to what extent do the reasons for having sex change across the life span.”

It would be just as interesting to ask the question “Are we really motivated by the things we think are motivating us. Ironically, on the same page of the Times was an article suggesting that is not likely to be the case!
Note: In another post I’ll try to address some of the actual findings. It is interesting, for example, how few of the reasons are “common” in the sense of representing most people’s experience much of the time! Even among the top 50 reasons, for example, most had mean scores that indicated that people said they were true only for “a few” or “some” of their sexual experiences!
This is also published on SexInThePublicSquare.Org

Technorati Tags: , , , , , , , ,


Filed under Archives of Sexual Behavior, Cindy M. Meston, David M. Buss, New York Times, psychology, public discourse, research, sex, sexuality