Level of social desirability – “the need to respond in a culturally appropriate manner” – is correlated with the degree of misreporting of various behaviours. A paper published in Archives of Sexual Behaviour reviews socially desirable responding in sex research.
Psychology professor Bruce M. King argues that extreme under- and over- reporting of desired and undesired sexual behaviours is comparable to the misreporting observed in other fields. Given that sexual activity is highly private, sex researchers largely rely on self-report measures; meaning that, it is not possible to verify many sexual behaviours (e.g., number of sexual partners, frequency of sex, sexual fantasies, exposure to pornography, etc.).
The social desirability bias has been established in other areas of research. For example, in health research, under-reporting of body weight and energy intake and over-reporting of height has been observed across various studies and has been correlated with measures of social desirability.
Numerous studies have examined the presence of social desirability bias in sexual research. Studies of men who have sex with men have revealed that social desirability bias influences responses about HIV serostatus and needle injection. Surveys of the general population have revealed that many men over-report condom use, with one study finding a correlation of only 0.51 between men’s self-reported condom use and their partner’s reports. Social desirability has also been found to influence responses about extramarital affairs and exposure to pornography.
A good example of extreme misreporting (comparable to nutritional self-reports) is men’s self-reports of erect penis size. Several studies which relied on self-reports of erect penis length found a mean of 6.0-6.4 inches, while the “combined mean length in 10 studies in which researchers took measurements (of pharmacologically induced full erections) was 5.36 inches, and 5.11 inches in 21 studies of fully stretched penises.” Importantly, the impression management component of social desirability is distinguishable from “intentional mischievous responding.”
King writes, “The results of many studies now indicate that social desirability responding in studies of self-reported sexual behaviors is as pervasive and often as extreme as is found in other research areas.”
The author makes a few recommendations for navigating socially desirable responding in sex research. He notes that extreme misreporting can occur even with anonymous and highly private studies (e.g., online surveys). One technique to reduce the social desirability bias is indirect questioning. For example, instead of asking “To what extent have you used an illegal drug or consumed alcohol on the job?” researchers could ask “To what extent has the average worker in your workplace used an illegal drug or consumed alcohol on the job?” Indirect questioning has proven effective in reducing social desirability bias.
It is also important to include measures of social desirability for studies that rely on self-reported data (e.g., the Marlowe-Crowne scale). Example statements include “I’m always willing to admit it when I make a mistake” and “I am always courteous, even to people who are disagreeable”. In this way, raw scores can be adjusted “by an amount commensurate with the degree of socially desirable responding.”
When presenting data, researchers can also consider using the median (vs. the mean) as the measure of central tendency, given that the median is unaffected by extreme scores and mischievous responding. It is also important to be aware of potential differences among demographic groups (e.g., between men and women) when adjusting data to account for social desirability bias. For example, under-reporting of energy intake is more common among women (vs. men), and older (vs. younger); in other words, different subgroups can have different perceptions of socially desirable behaviours.
King concludes, “When asking individuals about their private sexual behaviors, attitudes, and desires, sex researchers should minimize social desirability responding (beyond anonymous responding) and attempt to ascertain the magnitude of any social desirability bias.”
The review, “The Influence of Social Desirability on Sexual Behavior Surveys: A Review”, was authored by Bruce M. King.
