In this paper we use data from the 2002 National Survey of Family Growth (NSFG) to ascertain and analyze patterns of asexuality in the United States. We endeavor to extend the earlier work of Bogaert (2004) on this topic, which focused on patterns of asexuality in Great Britain. Using a social constructionist perspective to study asexuality, we conceptualize and measure the phenomenon in several ways, according to behavior, desire, and self-identification. We use the NSFG respondent sampling weights to produce several sets of unbiased estimates of the percentages of persons in the U.S. population, aged 15-44, who are asexual; each set is based on one or more of the various definitions of asexuality. Finally, we describe some of the characteristics of the asexual population using logistic regression.
In recent years there has been an increase in demographic studies of sexuality. The resulting research has provided insights into broader population patterns of sexual behaviors, desires, and identities, and has emphasized the complexities inherent in the analysis of sexual outcomes. But with very few exceptions, none of the sexuality analyses have focused on asexuality, hence overlooking a sexuality dimension that may well characterize a not insignificant percentage of the population.
In this article, we examine data from the 2002 National Survey of Family Growth (NSFG) dealing with the prevalence and characteristics of asexual persons. Our research highlights the measurement and interpretation challenges that accompany demographic analyses of sexuality in general, and asexuality in particular. Recent research about sexual behaviors, desires, and identity has found that a small, but notable, number of individuals do not appear to fall clearly into the heterosexual, gay, or bisexual categories in terms of their sexuality. Rather, they report that they are not engaging in sexual activity, that they experience no sexual desire, and/or that they self-identify as asexual (Bogaert 2004; Laumann et al. 1994). Asexuality has gained modest attention in the popular media as a sexual orientation (Sohn 2005; Jay 2005), but little quantitative and generalizable information is available about the prevalence of asexual identity or behavior in human populations or the characteristics associated with asexual persons. In the one quantitative study of which we are aware, Bogaert (2004) conducted a study of asexuality using nationally representative data, concluding that around 1.1 percent of the sampled British population provided an asexual response to a survey question on sexual desire.
Drawing on 2002 NSFG data, we seek to extend Bogaert's analysis in two ways. First, we explore whether similar patterns of asexuality are exhibited in the U.S. population as in the British population. In addition, and more importantly in terms of broader demographic studies of sexuality, we incorporate multiple dimensions of asexuality, in contrast to the single dimension permitted by Bogaert's data.
The limited literature on asexuality presents three kinds of definitions dealing with the phenomenon, namely, definitions based on one's behavior, one's desires, and one's self-identification. Our analysis of asexuality across these three dimensions highlights the manner in which both the prevalence and implications of asexuality vary depending on the dimension employed. Further, our analysis demonstrates some of the challenges faced by demographers using survey data on sexuality, as well as in crafting questions to explore population sexuality.
There is a limited social science literature on asexuality. This is due in part to the presumed low levels of asexuality in human and nonhuman populations. Asexuality is thought by some to be low because "one would expect strong selection pressures against such nonreproductive tendencies" (Bogaert 2004: 279). But the fact that a behavior has "nonreproductive tendencies" does not necessarily mean that it will have a low prevalence. …