The first known sex surveys in the United States were the Kinsey Reports, published in 1948 and 1953. Alfred Kinsey was an evolutionary biologist and a professor of zoology from Indiana University who had never studied human behavior. The main focus of his early research was in fact the mating practices of gall wasps [4]. In 1938, he was asked to teach the sexuality section of a course on marriage. As he was looking up literature in preparation for his lectures, he realized that scarcely anything had been published on the topic. He then decided to conduct his own study, using taxonomic methods transposed from entomology with which he was familiar. He started by handing out questionnaires to his class students, soon switching to face-to-face interviews. Along with his three colleagues, he ended up interviewing close to 18,000 subjects. He used samples of convenience starting with his own students, later expanding to other college students, prison inmates, mental hospital patients, a group of homosexuals, and hitchhikers. His findings were published in two books, Sexual Behavior in the Human Male in 1948 [5], followed 5 years later by Sexual Behavior in the Human Female [6]. These two works were later collectively dubbed the Kinsey Reports. They challenged preconceived notions about sex, a topic which was previously deemed taboo. Notably, Kinsey’s data was presented free of any social, cultural, or political taboos [7]. The controversial nature of the topic naturally invited a backlash. Many critics, including religious and political figures, feared that Kinsey’s candid discussion of sexual practices would disrupt the moral order of the nation. Nevertheless, the Reports soon reached the top of bestseller lists, turning their author into a celebrity.

Some of Kinsey’s most famous and controversial findings concerned homosexuality . He reported that 37% of men had at least one sexual encounter with another man at some point in their life and that 46% of men “reacted” sexually to persons of both sexes. Ten percent of men had had sex exclusively with other men for at least 3 years [5]. This figure may be the basis of the oft-repeated assertion that “10% of Americans are homosexual.” Kinsey himself however avoided the categories “homosexual” and “heterosexual,” instead using a seven-point scale—later dubbed “Kinsey scale ”—to describe a person’s sexual orientation, ranging from 0, or “exclusively heterosexual,” to 6 or “exclusively homosexual” [5]. Regarding women, Kinsey reported that 7% of single females aged 20–35, as well as 4% of previously married women from that same age group, had equal heterosexual and homosexual experiences or responses (a rating of 3 on the Kinsley scale). One to three percent of single females of that age group were reported to be exclusively homosexual [6].

Kinsey estimated that the average frequency of marital sex among women in their late teens was almost three times a week, and it decreased to a little over two times a week at age 30 and once a week by age 50 [6]. On the topic of masturbation , he reported that 62% of female responders partook of it, 45% of which indicated that they could achieve orgasm in less than 3 min [6]. The rate of masturbation in males was markedly higher, at 92% [5].

Kinsey reported that around 50% of all married males had had an extramarital affair at some point during their married life, as did 26% of females under 40. He estimated that between 1 in 6 males and 1 in 10 females aged 26–50 had engaged in extramarital sex. Eighty-six percent of men reported that they had engaged in premarital sex, as did half of the women who married after World War I.

Additionally, Kinsey reported that 50% of males and 55% of females endorsed having responded erotically to being bitten. Furthermore, 22% of males and 12% of females reported an erotic response to a sadomasochistic story [5, 6].

Although his reports are still regarded as foundational works in sex surveys and the study of sex in general, Kinsey’s methodology was heavily criticized, especially his use of convenience samples. His rationale for not using a random sample was that he could not persuade a random sample of Americans to answer deeply personal questions about sexual behavior. The result is a sample which, despite being considerable in size, may not be representative of the general population, as is generally the case with convenience samples [8]. Some groups were overrepresented in the sample. For instance, 25% of respondents were or had been prison inmates, and 5% were male prostitutes. One vocal critic even asserted that “a random selection of three people would have been better than a group of 300 chosen by Mr. Kinsey” [9]. Additionally, many of Kinsey’s respondents volunteered to be in the study. Subsequent investigations showed that individuals who volunteer for surveys are often not representative of the entire population [10]. Nevertheless, Kinsey’s work has been described as “monumental ” [11] and has been named as one of the factors that changed the general public’s perception and eventually led to the sexual revolution of the 1960s. It also led the way for the next generation of sex researchers. Indiana University’s Institute for Sex Research (later renamed Kinsey Institute) has strived to expand and update the trove of information left behind by its founder. Subsequent studies were conducted, most notably in 1970 and in 2009. They are presented below. Researchers in other countries also emulated Kinsey. A notable example is the survey conducted in the United Kingdom in 1949 by the social research organization Mass-Observation. It was dubbed “Little Kinsey” although, unlike its model, it was conducted using random sampling [12].

In the late 1950s and early 1960s, Washington University gynecologist William Masters and his research associate and eventual wife Virginia Johnson studied sexual behavior using a medical model. They described the anatomy and physiology of human sexual response by observing 382 women and 312 men engaging in “10,000 complete cycles of sexual response.” Their findings were published in Human Sexual Response (1966) [13] and Human Sexual Inadequacy (1970) [14].

The next major study was funded by the NIMH and was conducted by Indiana University’s Institute for Sex Research, under the direction of Albert D. Klassen, subsequently joined by Eugene E. Levitt and carried out by the National Opinion Research Center (NORC) at the University of Chicago. It was originally conceived as a survey of the public’s perception of homosexuality but was later expanded to cover a wide variety of sexual behaviors and attitudes. A fairly large sample of Americans (3018 respondents) were surveyed from 1970 to 1972. The data gathering process was slower and more expensive than anticipated, and the data analysis and writing processes were mired in controversy and personal disputes. As a result, the manuscript, which was finished as early as 1979, was not published. It was not until 1988 that the text re-emerged [15]. What Science had dubbed “The Long, Lost Survey on Sex” [16] was eventually published in 1989 as Sex and Morality in the U.S. [15], with additional data published in Science [17]. The main interest of the study lies in the fact that it dealt primarily with perceptions of sexual practices rather than the prevalence of the practices themselves. The researchers reported that 48% of respondents disapproved of masturbation, and a vast majority disapproved of extramarital coitus and homosexual relations without affection (87 and 88%, respectively). Even homosexual relations with affection had a disapproval rating of 79%. Similarly, a majority (65–82%) disapproved of premarital sex in teenagers, whether boys or girls, with or without romantic love. Adult age and the presence of love were predictors of a lower disapproval rate. Still, 29% of responder approved of outlawing premarital sex, 59% favored laws against homosexuality, and 14% believed a person convicted of homosexuality should be sentenced to at least a year in prison [15].

In subsequent years, there were other smaller studies, targeting certain specific population groups, such as young women [18–20] or college students [21]. These studies tended to focus more on social issues, such as contraception and teenage pregnancy, rather than actual sexual practices [22].

In the absence of rigorous scientific studies of sexual practices , a number popular nonscientific or pseudoscientific reports attempted to assuage the public’s curiosity about the topic. More often than not, these studies used convenience samples. For instance, a number of these studies were funded by and published in magazines, and they drew their samples from the readership of said magazines. These include Psychology Today [23], Redbook [24], and Playboy [25].

The samples used in the studies may be large—20,000 in the case of Psychology Today , more than five times more for the Redbook study. However, these samples are drawn from the readers of the publications, and it has been argued that these are an already preselected population not necessarily representative of the general population, especially in terms of liberalism and education [15]. An additional issue is that of response rate. In the case if the Redbook survey, a survey was sent out to 4,700,000 Redbook readers , and only 2% of them responded. Such a response rate casts doubts as to the representative abilities of the findings, a shortcoming that will plague many other such surveys, such as that conducted by American-born German sex educator Shere Hite. Hite sent out surveys to women whose names she obtained from women’s organizations and subscriber lists of women’s magazines. In total, she distributed 100,000 questionnaires and received 3000 of them back, a 3% response rate [26].

The Janus Report (1993), by Samuel S. Janus and Cynthia L. Janus, was distributed to 4550 subjects with 2795 returned and were “satisfactorily completed” [27]. These included volunteers who came to the offices of sex therapists. It presents itself as an updated snapshot of American sexual practices in the context of the AIDS epidemic that had started a decade earlier. The report claimed to be going against preconceived notions about the sexual habits of senior citizens. It reported that over 70% of Americans ages 65 and older have sex once a week. The rate reported is almost as high in men over 65 (69%) as in men aged 18–26 (72%). It even went on to claim that a similar proportion of men over 65 have sex every day (14%) as men aged 18–26 (15%). The Januses’ conclusions contradicted previous—and subsequent—studies that found a gradual decline in sexual activity beginning in the fifties. The Janus Report was heavily criticized for its skewed sampling, which resulted in overestimating the rates of sexual behaviors. For instance, Arthur Greeley [28] systematically compared the results obtained by the Januses to results drawn from the General Social Survey (GSS) , which was based on a national household-based probability sample conducted by the National Opinion Research Center over two decades. He found that the Janus Report estimates were often 2–10 times higher than those of the GSS. For instance, as mentioned above, the Janus Report estimates that 69% of men aged 65 and above have sex at least once a week, as do 72% of men aged 18–26. By comparison, GSS rates for these age groups are 17 and 57%, respectively. The fact that the Januses’ findings were never replicated elsewhere seems to give credence to their detractors.

In 2004, ABC News Primetime Live published its own American Sex Survey . The researchers interviewed 630 American adults by telephone, out of a random national sample of 1501. The survey lavishly describes “eye-popping sexual activities, fantasies and attitudes in this country.” Among the results is 42% of Americans call themselves “sexually adventurous,” 30% of single men aged 30 and older have “paid for sex,” and that half of women acknowledge having “faked an orgasm” [29].

The scarcity of serious up-to-date studies and the urgent necessity of accurate information about sexual behaviors as a matter of public health, especially in the era the HIV epidemic, were the drivers behind the National Health and Social Life Survey (NHSLS) in 1990 [30]. The NHSLS, sometimes dubbed the Chicago Study or Chicago Survey, sought to remedy what its authors saw as methodological flaws in all previous sex surveys, from Kinsey on. The purpose of the NHSLS, according to one of its authors, was to “collect and analyze data on the social organization of sexual behavior, particularly the social structuring of sexual action, and the ways in which that structuring influences behaviors that increase the incidence and prevalence of a variety of health-related problems” [22]. The inception of the NHSLS was not without controversy. The study of people’s private sexual behaviors has long been contentious, as evidenced by the vivid reactions to the Kinsey Reports. The government was reluctant to fund a project which would “provide a mandate of excessive sexual expression” [31], and although the study was originally requested by the National Institute of Child Health and Human Development (NICHD), its funding was soon terminated. The researchers were thus forced to turn to private donors to fund the study, including the Robert Wood Johnson Foundation, the Rockefeller Foundation, and the American Foundation for AIDS Research. The NHSLS was headed by Edward O. Laumann, Research Associate at NORC; John H. Gagnon, Professor of Sociology and Psychology at the State University of New York at Stony Brook; Robert T. Michael, NORC founding director; and James Coleman, NORC Research Associate [8].

The NHSLS used a novel approach in sampling. It involved a multistage area probability sample designed to give each household an equal probability of inclusion. The sampling returned 4369 eligible respondents. These subjects were administered a face-to-face 90-min survey, 3432 of which responded and were thus included in the study, resulting in a response rate of 78.6%. The subjects were aged 18–59, including 75% Whites, 12% African Americans, and 8% Hispanic Americans.

NHSLS researchers reported on a wide array of topics, such as sexual fantasies, masturbation, orgasm, and emotional satisfaction. Some of the findings from this survey are highlighted later in this chapter. The results of the study were published in Sex in America: A Definitive Survey [8] and The Social Organization of Sexuality [22]. Among the NHSLS findings was the fact that Americans tend to engage in sexual intercourse with peers who are similar to themselves in age, education, and ethnicity. It also revealed that, at the time, most sexually transmitted infections were contracted by young adults. 16.9% of respondents reported that they had at some point been diagnosed with at least one sexually transmitted infection. The survey also reported a correlation between the number of both lifetime and simultaneous sexual parwtners and the likelihood of contracting a sexually transmitted infection. Those with more than 10 lifetime sexual partners were 20 times more likely to have contracted such an infection as those with only one. Interestingly, the NHSLS found that the populations at highest risk of contracting sexually transmitted infections were starting to change their sexual practices accordingly, for instance, with increased condom use.

Overall, according to Cooks and Baur, the NHSLS “stands alone as the most representative U.S. sex survey and as one that reliably reflects the practices of the general U.S. adult population in the 1990s” [31]. Nevertheless, the study had some important shortcomings. As the NHSLS was conceived in the wake of the AIDS epidemic, it was in part designed to understand sexual behaviors associated with the transmission of the disease, so the researchers focused mainly on sexual practices that may lead to infection, omitting many significant non-genital sexual acts such as hugging, kissing, and body stroking, as well as the events surrounding the act such as courtship. Another major limitation has been the exclusion of adolescents and older adults, as the NHSLS researchers have chosen to focus solely on adults aged 18–59. It was in order to remedy to these limitations that some more cohort-specific surveys have appeared in recent years. The Youth Risk Behavior Survey (YRBS) , conducted by the Center for Disease Control and Prevention (CDC) has strived since 1991 to gain data in the adolescent population, studying health risk and health protective factors in 9th to 12th graders, including sexual behavior [32–34]. The National Social Life, Health and Aging Project (NSHAP) , on the other hand, researches the elderly population [35].

An all-encompassing survey to cover all age groups remained absent until two decades later, with the National Survey of Sexual Health and Behavior (NSSHB) conducted by the Kinsey Institute and Indiana University. The NSSHB not only provided an expanded age group but also allowed insight into changing trends and developments in sexuality over the past 20 years [36]. The NSSHB studied nearly 6000 subjects aged 14–94 and provided the scientific community with the first batch of thorough data on human sexuality and sexual behaviors in the past two decades. The NSSHB also took into consideration new developments in society which could influence sexuality. Such changes include the use of medications to treat erectile dysfunction (sildenafil, marketed as Viagra^®, was released in 1998) which had extended the sexual lives of many and increased sexual behaviors in older ages. Also, in contrast to the 1970 Kinsey survey, same-sex relationships were now viewed differently, as same-sex marriage had become legal in some states and there had been increased recognition of same-sex partnerships and lifestyles. Another major development was the possible influence of the Internet on sexual perceptions and practices. The ability of the NSSHB to consider this influence on sexuality provides data, which researchers hoped would be more representative of the national population [37].

The NSSHB set out with the expressed aim to be the most nationally representative sexual survey conducted to date. It enrolled a total of 5865 participants (2936 males and 2929 females) through a population-based cross-sectional study conducted in the spring of 2009, randomized using random digit dialing and address-based sampling. This allowed the sampling frame to cover approximately 98% of all US households. Ages included ranged from 14 to 94, with adolescents requiring consent from their parent or guardian. Sample adjustments made included gender, age, race (Black, Hispanic, White, or other), geographic region (Midwest, North, South, West), sexual orientation (heterosexual, homosexual, bisexual, asexual, or other), household income, level of education, and relationship or marital status. Participants were asked to report whether or not they had engaged in certain solo or partnered sexual behaviors and, if so, how recently (never, within the past month, within the past year, or more than a year ago). These reports were obtained via the Internet, as opposed to face-to-face interviews in past surveys. Findings were presented as 95% confidence intervals sorted by age cohorts. Sexual behaviors included masturbation (solo and partnered), vaginal intercourse, anal intercourse, and same-sex behaviors. Condom use during vaginal intercourse was also assessed [37].

The NSSHB claimed to be a more modern sexual survey in comparison to previous studies, arguing that its design and items of investigation allow for a better insight into the sexual health of the US population. Investigation of earlier ages (younger than 18) provided insights into young teens, who have been considered a higher-risk demographic. Investigation of older ages (over 60) has allowed a better understanding of sexuality in an age group which has seen an extended sexual life due to the pharmacological advancements, an expanded range of sexual enhancement products (vibrators, lubricants), and consumer marketing messages that shape expectations for sex and relationships at an advanced age. Another particularity of the NSSHB lies in its methodology for data acquisition; the researchers claimed that, since answers were sent via the Internet, they were likely to be more honest than face-to-face answers such as those of the NHSLS, the latter being thought as more anxiety provoking and not allowing for the same degree of honesty. Nevertheless, the study did have its own major limitations. In particular, in a recurrent problem common to many sex surveys, the sample may have been subject to self-selection, as those who chose to participate may represent different sexual personalities than those who chose not to participate. Also, subjects sampled were only those accessible in the community and not those in group homes, hospitals, or long-term care facilities. Such factors are important and should be taken into consideration when designing future sexual surveys.

Although this chapter focuses primarily on surveys of sexual behaviors conducted in the United States, there were many more carried out in various parts of the world. Listing them all would be far outside the purview of this volume, but one notable study is the Sexual Well Being Global Survey, or SWGS , conducted by Kevan Wylie, M.D., of the United Kingdom in 2006 and published in 2009. The researchers surveyed 26,032 participants across 26 countries, including the United States, the United Kingdom, China, Brazil, and Nigeria. The study was conducted online except for Nigeria where the respondents were interviewed face-to-face. The listed objective of the study was to “identify the variety of sexual behaviors undertaken by adults across the world.” The researchers found that three out of five respondents globally said that sex was important to them, with comparable findings in men and women. Sixty-nine percent of those surveyed enjoyed sex. Forty-four percent of participants were “very or extremely satisfied” with their sexual life [38].

The following data regarding the prevalence of sexual behaviors are mainly derived from the most recent nationally representative sexual surveys, the NSSHB [8, 22] and NHSLS [36–41] (Figures 3-1 and 3-2).