Human sexuality is a topic of endless fascination and frequent research study. However, that research is of limited value – not no value, but limited value – and should always be taken with a grain of salt. Here are four reasons why.
1. A lot of studies about sex are based on self-reported behavior. There is, however, no way to confirm the results without observing what people really do, as opposed to what they say they do. There are two reasons why the results may thus be inaccurate.
a. The subjects lie directly.
Recently OfftheCuff linked to some NIH/CDC data on my site showing that the average number of sexual partners for females is 3ish and for males is 5ish. However, a study published in 2003 showed that when subjects were hooked up to lie detectors, women admitted to higher numbers of partners and men admitted to lower numbers of partners, with men and women having about the same number overall. Another similar study published in 2013 found that:
Back in 2003, women went from having fewer sexual partners than men (when not hooked up to a lie detector) to being essentially even to men (when hooked up to the lie detector.)
In this new study, women actually reported more sexual partners than men when they were both hooked up to a lie detector and thought they had to be truthful.
b. Subjects lie indirectly.
I was wondering aloud about this discrepancy in data recently, and Deti made an interesting point:
I think the average male lifetime N of 5 or 6 is probably accurate. I think women are able to convince themselves that their Ns are lower than they actually are, and can lie convincingly about it. I think they can lie so convincingly about it that they can even deceive themselves into believing their Ns are lower than they are. Another factor is sex while intoxicated and the distorted memories and perceptions associated with it. I suspect a lot of women don’t remember a good portion of the sex or the sex partners they’ve had. Polygraphs are notoriously unreliable and can be defeated.
2. Sex researchers, like many social science researchers, often have specific social agendas that they want their research results to support. For example, I automatically discount any research I hear about homosexuality now because there is close to zero chance that it is going to be unbiased; any results that show homosexuality to be negative in any way are automatically thrown out and attempts are made by gay activists to discredit the researcher, as the Mark Regnerus case proved recently. Other examples of sex researchers who truly used shoddy research methods and whose findings have been questioned in recent years are Margaret Mead and Alfred Kinsey, both of whom had radical social agendas.
3. It is hard to make generalizations about sexual preferences and relationships because there is so much variability.
In Be The Researcher, blogger Private Man does a nice job of explaining the flaws in social science research and explains that he prefers to base his dating advice on behavior he has directly observed:
Look to people’s actions and not necessarily their words. If a dating and attraction study is based on actually watching human behavior “in the wild” then it is far more valuable when presenting truth.
All of us can become social and psychological researchers if we carefully observe actual human behavior in the context of attraction and dating. I do this constantly when in social situations so I can pass along useful information to my readers. I make a point to be acutely aware of social scenarios, particularly when men and women are together.
When observing, the challenge is to notice the subtle things like tone of voice, eye contact, and body language. It takes practice. The best source for seeing human interactions is go places where first, second, and third dates are happening on a regular basis. Also, singles Meetup group events are a great source for observation.
I am positive that his observations are useful to his client base, so his observational research results serve the purpose that he needs them to (advising primarily secular men and women on dating).
However, are such research results useful for making generalizations about all men and women? I would say no, they aren’t. I recently realized that our chatter in the blogosphere cannot really be widely generalized, either, and here are two examples to support that:
a. Women’s experiences are widely divergent.
I have always assumed that if I could stay with my husband through difficult times, then so could any other woman. And I still believe women have a moral imperative to do this, but I realize now that I am generalizing my own experience of marriage onto women for whom it doesn’t fit at all. I am strongly attracted to my husband, and there is just no way I can understand what it is like to try to honor one’s marriage vows when one is very unattracted to one’s husband, and I believe I have underestimated the difficulty that women in that situation face.
b. Men’s preferences are wildly variable.
I recently participated in a conversation at RT’s place which clearly demonstrated this. The topic of the post was the story of a man whose wife would not perform certain sex acts that he desired while they were dating (presumably they are not Christians). Because he really liked her, he married her anyway, but they had a lackluster sex life. Recently, he had found an amateur porn video that his wife had participated in while in college, in which she enthusiastically performed all the sex acts (and many more) that she would not do with him.
Several men mentioned that the problem wasn’t that she had done these things but rather that she would not do them with HIM. They said they wouldn’t mind what their wives had done in the past; what mattered was if their wives were still willing to do these things now with them.
I asked the men there about that. Would they really be willing to marry/partner with a woman who had participated on film in oral and anal sex with numerous partners? The men said yes, so long as she was still willing to do so with him. I was startled by this because many men here say that they are absolutely firm in their resolve not to marry a non-virgin, but some of the men at RT’s say they would be willing to marry/partner with a hardcore slut so long as she was still willing to perform a variety of sex acts. This really demonstrates that there is a great deal of variance among men concerning what they desire in a sex partner and/or wife.
4. The group of subjects in any kind of objective study of sexuality are inherently going to exhibit selection bias (H/T to Zippy Catholic for being the first person I ever heard mention this idea).
Let’s look at one example. The way that female sexual arousal is studied is usually by showing the female subjects erotic or pornographic video clips while they are hooked up to a vaginal photoplethysmograph. Here is a description of that device:
A vaginal photoplethysmograph is a clear acrylic, menstrual tampon-shaped device that contains a light source, and a light detector. The light source illuminates the capillary bed of the vaginal wall and the blood circulating within it. As the amount of blood in the vaginal tissue increases, more light is reflected into the photosensitive cell of the device.
In plainspeak: they stick that plastic thing right in her woo-woo. What kind of woman is willing to participate in that? Well, women who are being paid to participate in the study. However, I personally (like many other women) do not accept cash money in exchange for access to my coochie. Women who accept money in exchange for access to their sexuality are more commonly called whores. Would my arousal cues be different than those of the kind of women (i.e. whores) who are willing to participate in such a study? There is literally no way to separate out the selection bias here because you cannot measure the vaginal vasocongestion of a woman who will not allow you to stick a photoplethysmograph in her vagina.
However, here’s a little bit from Wikipedia about this device which I found fascinating:
Past research suggests that studies using VPG result in small and unrepresentative samples. However, more recent assessments of women’s willingness to participate in sexual psychophysiological research using VPG found no evidence of sampling bias influencing results.
No evidence of sampling bias? I was intrigued. No link to the study cited for this was provided, but I googled it and found the abstract. Here it is:
In this study, the authors investigated the hypothesis that women’s sexual orientation and sexual responses in the laboratory correlate less highly than do men’s because women respond primarily to the sexual activities performed by actors, whereas men respond primarily to the gender of the actors. The participants were 20 homosexual women, 27 heterosexual women, 17 homosexual men, and 27 heterosexual men. The videotaped stimuli included men and women engaging in same-sex intercourse, solitary masturbation, or nude exercise (no sexual activity); human male-female copulation; and animal (bonobo chimpanzee or Pan paniscus) copulation. Genital and subjective sexual arousal were continuously recorded. The genital responses of both sexes were weakest to nude exercise and strongest to intercourse. As predicted, however, actor gender was more important for men than for women, and the level of sexual activity was more important for women than for men. Consistent with this result, women responded genitally to bonobo copulation, whereas men did not. An unexpected result was that homosexual women responded more to nude female targets exercising and masturbating than to nude male targets, whereas heterosexual women responded about the same to both sexes at each activity level. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
There is no mention of the study of selection bias; why is this being used to disprove the hypothesis that selection bias is inherent in vaginal photoplethysmograph use? This proves to me that you should never trust anything that is said about problems with sex research procedures without carefully investigating the claims.
Due to the inherent selection bias in sex research, the bias exhibited by many of the researchers, and the lack of generalizability of the results due to wide variation in human experiences and preferences, I do not take any sex research at face value. I advise my readers to question the results from sex research studies and take them with a grain of salt.