Book Review

Sex in America: A definitive survey

N Engl J Med 1995; 332:1452-1453May 25, 1995

Article

Sex in America: A definitive survey
By Robert T. Michael, John H. Gagnon, Edward O. Laumann, and Gina Kolata. 300 pp. Boston, Little, Brown, 1994. $22.95. ISBN: 0-316-07524-8

It is a measure of both the prudery and the prurience of our times that this book derives from a survey that evaded the S word and was named the National Health and Social Life Survey. Names are influential. They are shaped by one's assumptions and in turn shape one's formulations. In Sex in America the conceptual assumption is explicit: it is the doctrine of social or sociological determinism — namely, that human sexual behavior is socially scripted. Thus, although the authors do not use the term social constructionism, that is the social-science doctrine to which they adhere. Accordingly, Sex in America omits consideration of biomedical (so-called essentialist) determinants and explanations. Too often, the discussion becomes defensively opinionated in adhering to a social or sociological explanation of the survey's statistics. It would have been better to present the prevalence, frequency, or incidence data dispassionately, letting them stand alone to speak for themselves, as is the case in the statistically dense companion volume, The Social Organization of Sexuality (E.O. Laumann, J.H. Gagnon, R.T. Michael, and S. Michaels. Chicago: University of Chicago Press, 1994). That approach would have allowed for not just one but several alternative attempts at ex post facto causal explanations. Tendentiousness would have been avoided.

There is not as much scholarly prestige, however, in presenting survey percentages, means, and ranges alone as there is in also offering an interpretation or explanation of what they mean. One way of doing this is to compare the data with those of a parallel study conducted in another place (the transcultural method) or in the same place at a different time (the before-and-after method). In the present instance, the data from prior studies have been criticized as irrelevant. Unlike the data from the present survey, the earlier data were not obtained from a nationwide random probability sample of informants. The place of these data has been absurdly usurped by a chimera, salvaged in part from the sexual revolution of the 1960s and 1970s. This chimera ostensibly represents Americans' beliefs about the sexual histories of other people, also Americans, which are supposedly a fantastic menu of orgasmic promiscuity, erotic variety, and dissolute abandon.

Far too many of the first 90 or so pages of Sex in America are devoted to a comparison of the mostly mundane findings of the survey with anecdotal examples of the chimera. Many of the anecdotes are quoted from well-known writers and other celebrities. The scientific alternative would have been to ascertain whether the survey respondents actually believed in this chimera. Then the book would have escaped the criticism of being tilted toward a moralistic return to sexual minimalism.

Subsequent chapters are more successful in escaping this criticism, in large part because, in the place of the chimera, there are statistical comparisons between legitimate subgroups. The statisticians are back on firm ground. The comparisons are based on such factors as sex (male or female), age, educational level, ethnicity, religious affiliation, availability of a sexual partner (spouse or other), number of partners, number of divorces or breakups, and history of forced sex (juvenile, adolescent, or adult). There are, of course, uncounted other factors that, funding and time permitting, might have been included, such as age of parents at informant's conception, mother's pregnancy history, number of siblings, order of sibling births, history of juvenile sexual-rehearsal play, history of degree of physical exhaustion, number of offspring, history of use of hormones, handedness, hirsutism, proneness to allergies, and so on. For any survey, however, the line must be drawn somewhere.

The variables of sexuality that are subjected to comparative analysis are number of partners, partner fidelity, frequency of having sex, and sexual practices and preferences (including oral and anal sex). Only male and female homosexual–heterosexual comparisons are presented, since so few respondents (1.4 percent of the women and 2.8 percent of the men) in the total sample (n = 3159) identified themselves as homosexual, although a higher proportion of respondents (5.5 percent of the women and 6 percent of the men) reported an erotic attraction to persons of the same sex. Similarly, there are no comparisons of people with AIDS and those without AIDS, since only five respondents reported that they had had a positive test for the human immunodeficiency virus (HIV) or had AIDS.

Chapters 8 through 12 make up most of the second half of the book, which ends with chapter 13, “Sex and Society” (a classification of orientations toward sexuality as traditional, relational, and recreational). Chapters 8 through 12 deal with masturbation and erotica, homosexual partners, sexually transmitted diseases, AIDS, and forced sex. Each of these five chapters constitutes an essay that sets the prevalence data obtained from the survey responses in historical and comparative context. The chapter on AIDS is extraordinary, for it is virtually a manifesto telling middle-class heterosexual whites that HIV will not be transmitted to them provided they keep away from intravenous drug users and from homosexual and bisexual men. It is an example of the written word misused as a lethal weapon.

Throughout the book there are so many reminders of the inferiority of survey data collected from so-called convenience samples to data collected with the authors' method that they seem to protest too much. For all its advantages, a sexological survey using a national random probability sample has not yet proved to give the same results as would a survey using a national census sample. A sexological survey is not like a political or consumer-spending survey, in which predictions are validated by how people actually vote or by how much they actually spend. Take the hypothetical case of an alleged connection between some aspect of sexual behavior and race — not race defined as black or white, irrespective of the degree of skin coloration, but race defined quantitatively by the degree of coloration as measured on a colorimeter. It is by no means certain that the criteria for selecting a representative sample for a political or consumer-spending survey would be the same as the criteria for selecting a sample representative of the degree of coloration in the United States. In other words, national probability sampling, for all its virtues, has its limits. Sex in America must be read with this in mind.

Some findings of the survey invite controversy on ideological grounds. The first experience of penovaginal intercourse is prevalent in the middle to late teenage years, before marriage. A plurality of partners is more prevalent before cohabitation or marriage and also after separation or divorce. After the initial bloom of a relationship, the frequency of penovaginal intercourse tapers off, the typical frequency being once or twice a week. Penovaginal intercourse is the most prevalent form of sexual activity, but a large proportion of people sometimes engage in oral sex, and a smaller proportion in anal sex. Watching the partner undress is erotically arousing for a large proportion of people. The frequency of masturbation does not decrease with marriage and may actually increase, and masturbation is not a substitute for but an addition to sex with a partner. Irrespective of the frequency of sexual activity and the degree of explicitly sexual gratification, most people report satisfaction with their present sexual partner. Types of sexual activity are not closely correlated with religious affiliation, either liberal or conservative.

Overall, this is a “pop sexology” book. It is not addressed to professionals in the fields of sexology and reproduction. The dedicated professional and academic reader will need to refer also to the scientific version of the survey, The Social Organization of Sexuality, to find statistical data applicable to the various claims and conclusions in this book.

John Money, Ph.D.
Johns Hopkins University and Hospital, Baltimore, MD 21205