William Marsiglio, John H Scanzoni, Kendal l Broad. Encyclopedia of Sociology. Volume 4. 2nd Edition. New York: Macmillan Reference USA, 2001.
In the face of significant political and methodological obstacles, social science researchers have continued to advance the understanding of human sexuality. Clearly, the political climate surrounding sex research has improved since Kinsey and his colleagues conducted their pioneering studies on male and female sexuality in the late 1940s and early 1950s (Kinsey et al. 1948, 1953), but the politics of sex research continue to impede progress in this area. One prominent research team, for example, was forced to abandon its efforts to secure federal funding and turn to private foundations to support a landmark study on human sexuality in the general population (Laumann et al. 1994b).
Despite the efforts of some conservative politicians, policymakers have become more willing to fund research on sexuality and related issues. This funding pattern is documented by the numerous large-scale national surveys that were supported by federal monies during the late 1980s and 1990s. These studies dealt extensively with sex and related issues (e.g., National AIDS Behavioral Survey, National Surveys of Adolescent Males, National Surveys of Men) or included a small battery of sex-related questions in surveys dealing primarily with other topics (e.g., ADD Health Surveys, National Household Survey of Drug Use, National Surveys of Families and Households, National Survey of Labor Market Experience-Youth, Youth Risk Behavior Survey). Funding agencies such as the Social Science Research Council Sexuality Research Fellowship Program are expanding studies of sexuality by offering crucial support for both quantitative and qualitative research projects about sexuality (Di Mauro 1997). Efforts also are being made to expand international studies of sexual behavior (Parker 1997). Most observers agree that the human immunodeficiency virus/acquired immune deficiency syndrome (HIV/AIDS) epidemic has provided the major impetus for this turnaround.
While a great deal has been learned about sexual behavior in recent years, especially among adolescent and young adult men, much of this research has focused on sexuality within a “social problems” context. This research typically has dealt with issues associated with teenage sexuality and pregnancy, HIV/AIDS, coercive sexuality, or another form of sexuality that has attracted a “deviant” label. As a result, relatively little attention has been devoted to studying the expression of sexuality in noncontroversial, everyday life circumstances.
Sex research took an impressive step forward with the publication of The Social Organization of Sexuality: Sexual Practices in the United States (Laumann et al. 1994a) and Sex in America (Michael et al. 1994), the less technical version of this study prepared for the general public. These publications were based on data drawn from the National Health and Social Life Survey (NHSLS) that included face-to-face interviews and supplemental self-administered questionnaires given to 3,432 respondents aged 18-59 who were selected randomly from the noninstitutionalized population. The NHSLS represents the most comprehensive sex survey to date that uses a probability sample of the general noninstitutionalized U.S. population.
Conducting sex surveys or other types of research in this area continues to be difficult because sex researchers, compared with researchers in most other areas, deal with particularly sensitive and personal topics and therefore face serious difficulties with issues related to response bias, sample representativeness, measurement, and ethics. With these methodological issues in mind (Bancroft 1997; Bentler and Abramson 1980; Jayne 1986; Kelley 1986) and given this article’s space limitations, this review assesses the available research on sexual behavior in the United States. The brevity of this article precludes a review of all the literature on individuals’ subjective perceptions of sexuality. Notably, much of the research on gay, lesbian, and bisexual experiences has revolved around identity, etiology, community, and AIDS (Risman and Schwartz 1988). While this entry discusses same-sex sexual behavior patterns, the reader should consult “Sexual Orientation” in this encyclopedia for a summary of literature about gay, lesbian, bisexual, and transsexual/transvestite identities and experiences. Since much of the literature on sexuality subsumes behavior under identity, this discussion uses the terms “heterosexual,” “homosexual,” and “bisexual” as adjectives rather than nouns and uses the terms “gay,” “lesbian,” and “bisexual” to refer to self-identity as reported in research studies (Risman and Schwartz 1988).
This discussion takes into account five basic and interrelated features of sexuality: (1) patterns of behavior, emphasizing gender, race, and sexual orientation, (2) the varied meaning of sexuality at distinctive periods throughout the life course, (3) social control aspects, particularly as they relate to prostitution, (4) the consensual and coercive contexts within which sex occurs, and (5) the relationship between the HIV epidemic and sexual behavior. While this entry reviews primarily social science literature, a number of studies have been published in the popular press, some of which have received considerable attention from the lay population, that are based on self-selected samples of persons who returned magazine surveys (Kelley 1986 provides a review of these works) or volunteered for qualitative interview studies (Rubin 1990).
Sexuality in the Life Course
Childhood Sexual Behavior. Given the Western view of children as asexual beings, little social science research has been conducted on childhood sexuality in the United States; most of the limited research has occurred within northern Europe. Data that are available typically rely on adults’ recollections of their childhood experiences or the reports of parents about their children. While this research is methodologically limited, it does offer a glimpse of children’s sexual behavior and play.
Anthropological research has shown clearly that while individuals in Western cultures tend to view children as asexual, children are seen as being capable of sexual activity in many nonindustrialized, non-Western countries (Ford and Beach 1951). Parents in non-Western countries sometimes tolerate and even encourage their children to pursue heterosexual behaviors (including intercourse), homosexual behaviors (e.g., fellatio), or both. In some societies mothers masturbate their children to soothe them.
The available research indicates that prepubertal boys are much more likely to experience intense sexual interest and masturbate and to do so at younger ages than their female counterparts. Boys also appear to participate more frequently throughout childhood in both heterosexual and homosexual play activities that have sexual overtones (e.g, “doctor and nurse”). These activities typically include an element of exhibition, exploration, and experimentation. In a unique longitudinal study of children in California, researchers found that about 48 percent of mothers reported that their children had engaged in interactive sex play (77 percent when masturbation was included) before age 6 and that exposure to such sex play was not related to children’s long-term adjustment at ages 17 to 18 (Okami et al. 1997).
Other researchers focusing on heterosexual and lesbian women in Brazil, Peru, the Philippines, and the United States found similarities across cultures in regard to memories of childhood sexual behavior. Self-identified lesbians were found to be more sexually active as children and reported earlier contact than did heterosexual women. Lesbians also were more interested in girls than heterosexual women were in boys, even though lesbians reported more early attractions to men than did heterosexual women (Whitam et al. 1998). Another study of gay and bisexual male youth ranging in age from 17 to 23 also depends on memories of childhood to understand early childhood same-sex attractions. The results of this study show that in most cases, same-sex attractions began in childhood and were given sexual meaning at the onset of puberty (Savin-Williams et al. 1996). Both studies suggest that same-sex attractions are felt in childhood and are acted on in adolescence.
Adolescent and Young Adult Sexual Behavior. In contrast to childhood sexuality, an expansive body of literature on adolescent heterosexual behavior has emerged during the past two decades, and much more extensive data are now available on young males (Moore et al. 1995, 1998; Sonenstein et al. 1997). Much of this research has used one of several national data sets to document and examine rates and trends for age at first intercourse and sexual activity patterns, with particular attention given to racial patterns.
The bulk of the evidence suggests that there was a sizable increase in the rate of sexual activity among teenage females in the 1970s and 1980s, although that increase appears to have leveled off in the 1990s. In fact, among 15- to 19-year-old females, while the proportion that had ever had intercourse increased from 29 percent in 1970 to 55 percent in 1990, it declined slightly to 50 percent in 1995. About 77 percent of 19-year-old females have had sex at least once. The main reason teenage females remain virgins is that having sex would violate their religious or moral values. While rates among African-American and white females have converged over the past two decades, black females are still more likely to report being sexually active than are their white counterparts. Whereas 90 percent of black females have had intercourse by age 19, about 75 percent of whites and Hispanics have had coitus.
Sexual activity rates among comparably aged males have increased since the 1970s and have always been higher than those of females. A 1998 study compared three separate national samples of males aged 17-19 living in metropolitan areas in 1979, 1988, and 1995 and found that the percentage of respondents who had ever had heterosexual intercourse shifted from 66 percent, to 76 percent, to 68 percent (Ku et al. 1998). However, these changes were restricted to nonblack males because the rates for blacks increased from 1979 to 1988 and then remained basically constant between 1988 and 1995. A similar curvilinear pattern for the overall sample (but not for blacks) was observed when researchers examined the proportion of teenage males who had had intercourse in the four-week period before the interview. Meanwhile, the average number of times young males engaged in sexual intercourse during the previous twelve-month period increased from fourteen in 1979 to seventeen in 1988 to twenty-one in 1995. While black males did not experience a change in frequency between 1979 and 1988, they reported a significant increase from 1988 to 1995 (thirteen versus twenty-four acts per year).
Attempts to explain why youth initiate sex have focused primarily on the direct and indirect influence of sociodemographic, social psychological, and biological factors. Most of the research designs, especially those used in studying females, are not ideal for concluding a causal relationship between beliefs, attitudes, or values and sexual behavior because these variables tend to be measured simultaneously. Despite this methodological shortcoming, a number of factors appear to be related positively (when controlling for numerous variables) to the probability of individuals engaging in heterosexual intercourse at a young age: being black, living in a poverty area, having weak religious beliefs, attending a segregated school (for blacks), attending an integrated school (for whites), lower parental education, having a mother who was sexually active at a young age, living in a single-parent household, having more siblings, and having a low level of academic achievement.
In addition to these social variables, a small number of researchers have used cross-sectional and longitudinal designs to study the relationship between adolescent hormones and heterosexual behavior (Udry 1988). While this research has produced mixed results, hormones and biological markers associated with puberty appear to be related to adolescents’ sexuality (both attitudes and behavior). Consequently, a growing number of social scientists are advocating the development of biosocial models that take into account the complex interrelationship among the pubertal process, sexual identity development, sexual behavior, and societal norms. Collecting saliva and blood samples has become an acceptable, and increasingly expected, component of national data collection efforts that target adolescent sexuality and fertility.
The discussion of social and biological factors leading toward the initiation of sexual behavior is also present in studies of same-sex attractions, desires, and behaviors. Similar to heterosexual onset of active sexuality, gay, lesbian, and bisexual youth report the development of sexual interest during puberty. Gagnon (1977) explains that young men typically self-identify as gay and have homosexual experiences at an earlier age than do young women. Jay and Young (1979) report awareness of same-sex attraction for young men developing at the median age of 13 or 14 and for young women at age 18. Cohen and Savin-Williams (1996, pp. 120-121) summarize the results of ten different studies of the initiation of sexual behavior (mostly among young men) and find that the average age of reported first experience of homosexual sex is 15 for young men, with young women reporting an average age of 16. Importantly, same-sex sexual behavior is not always correlated with the development of a gay, lesbian, or bisexual identity. Studies show that youth will participate in homosexual activities without later “coming out” as gay, lesbian, or bisexual. In particular, as a result of cultural understandings of gender and sexuality, Chicano men who participate in same-sex anal intercourse do not consider it “gay sex” unless one is in the subordinant (receptor) role (Almaguer 1993; Alonso and Koreck 1993; Carrier 1989). Research in this area supports the supposition that social and cultural factors are important in the initiation of same-sex sexual behavior and its attendant meaning.
Research on adolescent homosexual and bisexual behavior is complicated by the stigma associated with a gay, lesbian, or bisexual identity, and so many youths do not self-report same-sex activities (Savin-Williams et al. 1996). While many self-identified gay, lesbian, and bisexual adolescents report desiring steady, loving same-sex relationships, many maintain heterosexual relationships and fear revealing close same-sex friendship because of the stigma associated with being “gay” (Hetrick and Martin 1987). Thus, studies of sexual behavior among gay, lesbian, and bisexual adolescents have documented heterosexual behavior, although with a lesser frequency than is the case with heterosexual counterparts. Interestingly, lesbian and bisexual female adolescents tend to report more heterosexual experiences than do gay and bisexual male adolescents (Herdt and Boxer 1993). Research on the frequency of sexual activity among gay, lesbian, and bisexual young adults indicates that for some populations, entry into college allows an increase in the frequency of sexual experiences (Evans and D’Augelli 1996).
While most research has focused on heterosexual intercourse, a number of studies of adolescent and young adult populations have examined issues related to the number of sexual partners, frequency of sexual intercourse, other types of sex acts, and the sequencing of petting behaviors (Laumann et al. 1994a; Moore et al. 1995 ; Warren et al. 1998). The majority of teenagers and those in their twenties tend to have one sexual partner at a time, a form of serial monogamy. One study found that 72 percent of women aged 18-29 reported having only one partner in the previous year, and another found that only about 10 percent of females have two or more partners in a three-month period. Among the sexually experienced, about 77 percent of 15- to 19-year-old females and 85 percent of those aged 20-24 report having sex more than once a month. Among sexually experienced teenage males aged 15-19, 54 percent report having had no more than one partner in the previous year, 80 percent report having had two partners, and 6 percent report having had five or more.
Researchers also have shown that the prevalence of oral sex has grown tremendously in this century. Kinsey’s data revealed that very few college women born between 1910 and 1935 performed fellatio (11 percent) or received cunnilingus (12 percent). More recent studies in California and North Carolina suggest that between one-third and one-half of adolescents aged 15-18 have engaged in oral sex (Hass 1979; Newcomer and Udry 1985), while nonrepresentative studies of college students in the United States and Canada indicate that between 32 and 86 percent of females have administered oral sex and between 44 and 68 percent say they have received it (Herold and Way 1983; Young 1980). Furthermore, Kinsey’s data suggested that oral sex was primarily experienced only among those who also had experienced coitus (only 5 percent of male and female virgins reported performing it), but more recent research indicates that a sizable minority of youth are experiencing oral sex while they are technically virgins.
Single-Adult Sexual Behavior. Because researchers have not clearly and consistently distinguished between young adult (18- to 24-year-olds) and adult sexual behavior in presenting their findings, it is difficult to present a clear-cut review of “adult” sexual behavior among single persons. Research has shown that rising divorce rates and postponement of marriage for heterosexual men and women have increased the population of single adults (Blumstein and Schwartz 1983). The inability of gay and lesbian partners to marry legally also contributes to the growing population of “single adults” because some persons in committed relationships are categorized as single. This growing population of single adults and the typical transition from experimentation and dating to long-term, committed relationships bring into focus a significant type of adult sexual behavior. While these demographic patterns are noteworthy, a large proportion of individuals still experience a significant proportion of their adult sexual histories within committed relationships or marriage.
For single adults, it is known that individuals today move through the sequence from first kiss to first intercourse much more quickly than did older cohorts of a similar age. Many of these first sexual experiences with a new partner occur within an arrangement that is perceived in some ways to be a relationship. Although men are more likely to find recreational sex outside a relationship acceptable, both men and women prefer to have sex within an ongoing romantic relationship. Thus, contrary to stereotypical images, most adults have sex infrequently when they are not in a relationship. Indeed, national data based on a sample of 18- to 59-year-olds who report being sexually active in the past year indicate that without controlling for age, 48 percent of men and 54 percent of women who have never been married and are not currently cohabiting have had sex only a few times or not at all in the past year. These figures varied only slightly for men (46 percent) and women (58 percent) who were divorced, separated, or widowed but were not married or cohabiting. The mean monthly frequency of sex (vaginal, oral, or anal) was 5.6 and 5.3 for men and women who never married and were not currently cohabiting and 5.4 and 5.1 for men and women who were divorced, separated, or widowed. Only 8 to 9 percent of men and women who had never married but were currently cohabiting—persons who could be perceived technically as being single—reported similarly low levels of sexual activity: a few times a year or not at all. Meanwhile, among never married, noncohabiting persons, 29 percent of men and 21 percent of women reported giving oral sex to their partners during the most recent sexual experience. Among the entire sample, regardless of whether the respondents had sex in the past year, 67 percent of men and 70 percent of women reported that they had engaged in active oral sex at some point during their lives.
One of the revelations of Kinsey’s early research was the extent to which people had same-sex sexual experiences without identifying themselves as gay, lesbian or bisexual. Since Kinsey’s research, this finding has been reproduced in a variety of different contexts. Laumann et al.’s comprehensive sex study (1994a) illustrates how the interaction of behavior, attraction, and sexual identity complicates the measurement of the number of people who could be understood to be gay, lesbian, or bisexual. Ultimately, Laumann et al. (1994a) conclude that there is a core group of people who define themselves as “homosexual or bisexual,” have same-gender partners, and express homosexual desires. Yet there are also a number of men and women who have adult same-sex sexual experiences or desires but do not identify themselves as “homosexual or bisexual.” Additionally, variables (e.g., place of residence and level of education) are found to influence the number of homosexual experiences. Because of the complexity of measurement, Laumann et al. (1994a) conclude that there is no single answer to the prevalence of homosexuality. Studies of adult sexual experiences among men of color in the United States and men of color in other cultures also have illustrated the extent to which heterosexual adults engage in same-sex sexual behavior: Research has identified a large population of men who have sex with men (MSM) yet do not self-identify as gay or bisexual (Manalansan 1996). Research since Kinsey’s original attempt to develop a comprehensive (although nonrepresentative) understanding of sexual behavior in the United States has shown that some adults participate in homosexual behavior without self-identifying as such.
Despite the complexity of sexual identity and behavior, a great deal of research on same-sex adult behavior patterns has been comparative, creating a binary heterosexual-homosexual comparison. This dualistic approach is so prevalent that many sexual activities are understood as being in the domain of only heterosexual behavior or only homosexual behavior. For example, tranvestitism (cross-dressing) has been defined in the psychological literature as primarily the domain of heterosexual men, yet more recent research suggests that men with cross-dressing habits behave as homosexual, bisexual, and asexual as well as heterosexual, suggesting a complexity to adult sexual behavior patterns that is not fully understood (Bullough and Bullough 1997). In a significant study comparing same-sex sexual behavior patterns with opposite-sex behavior patterns, Masters and Johnson (1979) found some general trends of similarity for adult “heterosexuals” and “homosexuals” (both men and women). Overall, their research shows that “heterosexuals” and “homosexuals” have similar fantasy patterns and physiological responses to sexual stimuli. In other words, response to sexual stimuli is not conditioned by a particular sexual identity. Some differences between heterosexuals and nonheterosexuals emerge when women’s experiences are highlighted. A study of 70 self-categorized heterosexual, bisexual, or homosexual women showed that bisexual and lesbian women were significantly more likely than were heterosexual women to describe their orgasms as “strong.” In addition, bisexual and lesbian women put more emphasis on oral and manual sexuality, while heterosexual women put more emphasis on intercourse as a source of sexual response (Bressler and Lavender 1986). Overall, comparative studies suggest a similarity of sexual response and fantasy for both heterosexual and homosexual behaviors but some distinctions in terms of actual sexual activities participated in by self-identified gay, lesbian, or bisexual individuals.
Another arena of research that compares heterosexual behavior to gay and lesbian behavior is the study of committed relationships. A great deal of research illustrates the prevalence of stereotypes about gay, lesbian, and bisexual people being promiscuous and not involved in long-term, committed relationships. However, in one of the few large-scale studies of relationships (including heterosexual, gay, and lesbian couples), Blumstein and Schwartz (1983) found that many lesbians and gay men establish lifelong partnerships. Research also shows that in established couples, gay men and lesbians report as much satisfaction (measured in numerous ways) as heterosexuals do (Peplau 1982). Specific research about sexual behavior among gay and lesbian couples has shown some interesting trends. A notable finding in Blumstein and Schwartz’s (1983) research is that the frequency of sex in lesbian couples is low compared to that in other couples. Among lesbian couples who had been together two to ten years, more than 25 percent reported having sex once a month or less. This finding has been explained in terms of women’s sexual socialization, different definitions of sexuality, and age of couples (Blumstein and Schwartz 1983; Johnson 1990). By contrast, Blumstein and Schwarz (1983) report that gay male couples report higher sexual frequency than do married couples (of one to ten years). This often is explained in terms of men’s socialization to value sexuality. Research on sexual frequency in committed couples illustrates the significance of gender and thus the diversity of relationships and attendant sexual behavior among self-identified gay men and lesbians.
Research about the diversity of sexual behaviors among gay, lesbian, and bisexual adults expanded in the 1990s. In general, gender is understood to account for variation of sexual behavior in the gay, lesbian, and bisexual population: Men having sex with men are understood to do so differently (in terms of frequency and activities) than women having sex with women. Variation also has been found within sexual identity categories. For example, it has been found that individuals involved in bisexual behavior have a variety of relationship types (Blumstein and Schwartz 1976). Meanwhile, distinctions in regard to adult same-sex sexual behavior in terms of race are not well understood. In a national-level study of more than 700 coupled, homosexually active African-American men and women, most of the respondents reported satisfying sex lives with their current partners and variation of sexual frequency among couples but no systematic difference between women and men (Peplau et al. 1997). The varieties of sexual behavior within gay, lesbian, and bisexual categories and couples are just being discovered.
The politics of sexuality research in the era of HIV/AIDS has expanded the understanding of sexual behavior among adults, but only in particular arenas. For example, national-level research on sexual behaviors among bisexual men and women and lesbians remains limited (Doll 1997). In addition, the way in which studies of male homosexuality have been conducted differentially operationalizes “homosexual” as an identity, so that populations of men who have sex with men (MSM) sometimes are included and at other times are excluded (Carballo-Dieguez 1997; Sandfort 1997). Clearly, studies of sexuality expanded dramatically in the 1990s, yet this growth pattern has been shaped by concerns associated with an era of HIV/AIDS.
Sexual Behavior among Elderly Persons. Although the proportion of the U.S. population over age 50 continues to grow, research addressing the relationship between aging and sexuality, in particular the sexual behavior of the elderly population (65 and older) is quite limited because of its frequent use of small nonrepresentative samples, its cross-sectional research designs, and its narrow, youth-oriented definition of sexuality as coitus. Consequently, generalizations are difficult to make, and most research on elderly persons’ sexual behavior deals with the physiological and psychological aspects of this phenomenon. The Viagra revolution undoubtedly will prompt researchers to devote more attention to older persons’ sexuality (Butler 1998).
Two of the more frequently cited, though perhaps dated, studies of aging and sexuality issues are the Starr-Weiner Report (1981) and the second Duke Longitudinal Study (George and Weiler 1981), neither of which was based on random sampling techniques. The former study included 800 sociodemographically diverse participants in senior centers, while the latter included a panel design of men and women health insurance program participants who were 46 to 71 years of age at the first observation period in 1969 and were followed for six years (n = 348 for those enrolled in all four data collection points). Seventy-five percent of the respondents in the first study reported that sex felt as good as or better than it did when they were younger. The results from an analysis restricted to the 278 married respondents who had been retained throughout the Duke study revealed that pattens of sexual interest and activity remained fairly stable over time, men reported higher levels of sexual interest and activity than did their female age peers, and younger cohorts of respondents reported higher levels of sexual interest and activity.
Compared to earlier studies (e.g., the Kinsey reports), the Starr-Weider and Duke studies as well as more recent ones have found higher levels of sexual activity among older persons. Accordingly, Riporetelia-Muller (1989, p. 214) concluded that “for those elderly who remain sexually active and have a regular partner, the rate of decline is not as great as formerly believed.” However, many older persons do not remain sexually active. Using data from the second Duke study, George and Weiler (1981) found that among those who were at least 56 years of age at the first observation date, 21 percent of men and 39 percent of women reported six years later that they had abstained from sexual relations throughout the study or were currently inactive.
In a study using data from a nationally representative household sample (Marsiglio and Donnelly 1991), about 53 percent of all married persons 60 years of age and older reported having sex in the past month, with 65 percent of those 60 to 65 years old being sexually active compared to 44 percent of those 66 or older. Among those who had been sexually active during the past month, the overall mean frequency for sexual relations was 4.3 times. In a multivariate context, persons were most likely to have had sex during the past month if they were younger, had a higher sense of self-worth and competency, and were married to a spouse who self-reported his or her health status as favorable. Surprisingly, when an interaction term was used to compare husbands and wives, no significant differences were observed in the way the partner’s health status was related to sexual behavior, although other research has found that both husbands and wives report that males’ attitude or physical condition tends to be the principal reason why they have curtailed or ceased to have sexual relations. Meanwhile, other research indicates that being widowed is the most frequently cited reason for not being sexually active among older women overall. While data on persons who are institutionalized are scarce, it appears that their sexual activity levels are low.
Being without a spouse does not necessarily mean that older persons are sexually inactive. Brecher (1984) reported, for example, that among unmarried persons 60 years of age and older, about 75 percent of men and 50 percent of women were sexually active. This is reinforced by Starr and Weiner’s (1981) finding that 70 percent of their respondents over 60 were sexually active, although only 47 percent were married. Finally, masturbation is an option used by some elderly persons to express their sexuality, presumably in a nonsocial setting. In a few studies, about one-third of women and slightly less than one-half of men over 70 report masturbating.
Just as stereotypes about elderly people and sexuality abound, so do stereotypes about the sexuality of gay, lesbian, and bisexual elders. However, research suggests that gays and lesbians offer a model of successful aging (Berger 1996). In terms of actual sexual behaviors, many of the gay men interviewed for this study reported that their frequency of sex had changed as they aged, supporting the idea that age is associated with a lower frequency of sexual relations (Berger 1996). Similarly, another study based on self-report data obtained from men between 40 and 77 years old found that the majority of these self-identified gay men were currently sexually active, although sexual interest declined somewhat with age. The majority reported no change in their enjoyment of sex from their younger years to the present (Pope and Schulz 1990). The myth of sexual and emotional isolation for lesbians is also challenged by a series of in-depth interviews with 20 women over age 50 who self-identified as lesbian. This research found that lesbians continue to be sexually active and tend to seek out other older women as partners (Raphael and Robinson 1992). Like other research on aging and sexuality, research on self-identified gay and lesbian elders shows that they have continued interest in expressing themselves sexually.
Consent and Coercion
In recent decades, researchers have increasingly focused on “rape and other forms of sexual coercion” (Muehlenhard 1994, p. 143). A few observers have reacted by asserting that the data assessing sexual coerciveness among adults who are known to one another are seriously flawed and thus exaggerate their substantive and policy significance (Roiphe 1993; Muehlenhard et al. 1994). However, most social scientists and clinicians consider sexual coerciveness to be both serious and significant and have begun to scrutinize the conceptual and measurement problems that underlie its use.
Perhaps the most significant historical distinction in this realm was between stranger versus nonstranger, or unknown versus known. Rape occurred most convincingly—to the minds of jurors, for example—when a man who was totally unknown to a woman executed penile penetration. If the man had been known to the woman in any way, her charge of forcible sex could be undermined, depending on how well she knew the man. If, for example, the man was her husband, she could not claim rape at all because consenting to be married carried the obligation to give her body to him unreservedly.
Koss and her colleagues were among the first to expand the discourse on sexual coercion beyond its conventional historical understanding. Using a national survey of college students, they reported that 38 percent of female college students reported sexual victimization that met the legal criterion for rape or attempted rape and almost 8 percent of males admitted to behaviors that could be classified as raped or attempted rape with at least one woman since the fourteenth birthday (Koss et al. 1985, 1988). Their conclusions were supported by other studies (Mosher and Anderson 1986; Rapaport and Burkhart 1984). Using data from the 1987 wave of the National Survey of Children, Moore et al. (1989) found that about 7 percent of U.S. adults aged 18-22 (females being more likely than males) confirmed that they had sex against their will or had been raped on at least one occasion. Other researchers have reported that between 10 and 12 percent of women report having been raped by dates and between 8 and 14 percent of wives report that their husbands have sexually assaulted or raped them (Finkelhor and Yilo 1985; Russell and Howell 1982).
Estimating the prevalence of the various forms of coercive sexuality is exceedingly difficult because the Uniform Crime Reports are widely believed to underreport the true rate of sex crimes and anonymous, self-report surveys vary widely in sampling techniques and findings. Nevertheless, Grauerholz and Solomon’s (1989) review of research in this area suggests that a large proportion of the U.S. population has experienced or will experience coercive sexual relations as a victim, a perpetrator, or both. A comparison of sexual coercion among university students in the United States and Sweden found higher rates in the United States (Lottes and Weinberg 1996). Researchers’ estimates of the pervasiveness of incest in the United States may vary the most. Whereas some researchers have observed that about 1 percent of U.S. females have been incest victims (Kempe and Kempe 1984), others have reported much higher figures. For example, Russell (1984) found that 16 percent of her large household sample of women 18 years of age and older in San Francisco had experienced incest before age 18, and 20 percent of Finkelhor’s (1979) predominantly white, middle-class New England college student sample who were raised primarily in nonmetropolitan areas reported having been an incest victim. The 1998 National Violence Against Women survey of 8,000 women estimated that some 17.7 million women in the United States, nearly 18 percent, have been raped or have been the victim of attempted rape. Nearly half the victims were assaulted before their seventeenth birthday. Some three-quarters of those saying they had been raped or assaulted as adults reported that the perpetrator was a current or former husband, a cohabiting partner, or a date.
Not only is sexual coercion an issue for many women, there is evidence to suggest that it is pertinent for some boys and men as well (Finkelhor 1979; Moore et al. 1989). A study of 115 sexually assaulted men found that the majority were assaulted at age 15 or younger, about half were assaulted more than once, and a majority knew their assailants. Among the 115 surveyed, 100 were assaulted by at least one man, 7 by a man and a woman, and 8 by women (King and Wollett 1997). Further, much of the research about sexual assault on men assumes that it is primarily heterosexual men who are the perpetrators of sexual assault on other men, doing so to express power and control. However, in a study of 930 “homosexually-active males” in England and Wales, about 28 percent reported having been sexually assaulted or having had sex against their will at some point in their lives. Moreover, 33 percent of the respondents reported being forced into sexual activity by men with whom they had previously had or were currently having consensual sexual activity. This research suggests that sexual assault in the gay, lesbian, and bisexual community needs to be researched further (Hickson et al. 1994). Some studies (Struckman-Johnson and Struckman-Johnson 1994) indicate that some men report that they were pressured or coerced by women into a sexual experience. Other studies focus on men both in and outside of prison who coerce and/or physically force other men into having sex (King and Wollett 1997; Struckman-Johnson et al. 1996).
In an effort to clarify matters of coercion and consent in the case of both sexes, O’Sullivan and Allgeier (1998) proposed a series of points along a continuum. First, there is sexual activity among persons known to each other that is the result of actual or threatened restraint, aggression, or force. It can be done by men against women and by men against men. Logically, it could involve women against women, but empirically that appears to be quite rare even in a prison situation (Struckman-Johnson et al. 1996), although Renzetti’s (1992) research on partner abuse among lesbian couples suggests that lesbian couples experience a cycle of violence that can include sexual assault. Nevertheless, the general lack of empirical evidence about coercive sex between women could indicate either lower rates of occurrence or gaps in the research. Coercive sex also could be initiated by women against men, but again, this seems to be rare and is associated with the male having become intoxicated (Struckman-Johnson and Struckman-Johnson 1994). O’Sullivan and Allgeier (1998, p. 234) summarize the first category as “coercive interactions in which a person [submits] to … sexual activity but does so under duress.”
Their second category of sexual activity occurs as the result of “seduction interactions in which one partner is clearly resistant,” [and a third is] “‘token resistance’ interactions in which a person expresses nonconsent but is willing and intends to engage in the sexual activity.” Their final category is “willing participation in an unwanted sexual activity.” This refers to “situations in which a person freely consents to sexual activity … without experiencing a concomitant desire for the initiated sexual activity.” They review literature suggesting the validity of these four categories and present their own research in support of the last one. Future research, they imply, should utilize these and/or other constructs to sort out differences along the consent-coercion continuum.
With evidence that coercive sexuality is prevalent, the question remains: Why do many men and some women coerce others to have sex with them? Some theorists argue that males are more likely to engage in various forms of coercive sexuality if they have strong ties to a peer group that supports sexually aggressive behavior. Compared to their less stereotypically masculine counterparts, men studied in the research noted above who possess traditionally masculine personality characteristics and hold rigid views of gender stereotypes are more likely to report that they have used physical force and threats to have sex and probably would use physical force to obtain sex if they could be assured that they would not be prosecuted. Other theorists have argued that a small percentage of women may facilitate different forms of coercive sexuality by playing sexually receptive or seductive roles. Not surprisingly, it is common for sexual assailants to believe, or at least report, that their victims were willing participants who enjoyed themselves while being sexually assaulted, even though these perceptions are clearly inconsistent with the victims’ accounts (Scully and Marolla 1984).
While the several forms of coercive sexuality share a number of themes, such as the objectification of women, individual factors and circumstances may be significant in accounting for why particular types of coercive sexuality occur. One of the important factors that seem to distinguish the typical stranger or acquaintance rapist from the “average” date rapist is the former’s greater tendency to have been sexually or physically abused by his parents or others. Date rapes tend to involve partners who knew one another and had established at least a modicum of interpersonal trust by making a commitment to spend time together. The dynamic nature of the interaction episodes that typify date rapes and the fact that at least one of the persons often has been influenced by drugs, alcohol, or both can obscure the participants’ intentions and behavior. Furthermore, the more individuals’ sexual scripting is influenced by traditional gender socialization, the more likely it is that coercive sexuality will occur because of sexual miscommunication and males’ reliance on coping strategies that emphasize dominance and aggression. Many men assume that women will offer token resistance to their sexual advances to create an impression that they are not sexually “promiscuous” (Check and Malamuth 1985). One study of 610 female undergraduates revealed that almost 40 percent had engaged in this type of token resistance at least once (Muehlenhard and Hollabaugh 1988). While these patterns should not be used to justify date rape, it is not surprising that some men distort the consensual petting that generally precedes date rape as a woman’s way of acknowledging her willingness to engage in more intimate forms of sexual interaction, even if this means that in some cases men will be required to pursue it forcefully. Finally, while there are many factors related to fathers’ incestuous behavior, one of the more frequently noted arguments underscores the common pattern in which a father pursues sexual and emotional intimacy with his female children (usually a series of episodes over time with the oldest female child) to compensate for his unfulfilling relationship with the adult female, who generally has withdrawn from her roles as mother and wife (partner).
Sexual Risk-Taking Behavior
Despite the well-documented overall increase in condom use in the 1990s, most research on risk-taking behavior indicates that a high percentage of people still regularly place themselves at risk of getting pregnant and/or contracting a sexually transmitted disease (STD) and/or HIV/AIDS. These patterns persist despite increased efforts at safe-sex education programs and services. Some trends are becoming increasingly evident. Data from face-to-face interviews with adult men and women aged 18-59 in the 1996 National Household Survey on Drug Abuse showed that 19 percent of the respondents used a condom during their last sexual experience (vaginal, oral, or anal sex, with no distinction being made in terms of the gender of the partner) when the event occurred within a relationship, while 62 percent did so when sex occurred outside an ongoing relationship (Anderson et al. 1999). Ninety-five percent of the respondents indicated that their last sexual experience occurred within a relationship. These data also showed that once relationship status was statistically controlled, individuals with high-risk sex and drug use profiles did not use condoms at a higher rate than did their counterparts who were not classified as being at an increased risk for HIV infection. Only about 22 percent of high-risk individuals used a condom at last intercourse within an ongoing relationship. Thus, both the casual and steady partners of high-risk individuals continue to place themselves at risk.
Meanwhile, a study of college students highlights gender differences in which men are found to be engaged in more risk-taking behaviors relevant to partner choice and sexual practices (Poppen 1995). Another study indicates that college women tend to put themselves at risk for sexually transmitted diseases by rarely using condoms and only minimally discussing sexual history with a partner (Sheahan et al. 1994). Further, differences in behavior based on sexual identity are empirically supported. Interview data from gay and bisexual men suggest different behavior patterns in regard to “safe sex.” These data suggest that gay men were more likely to have had a steady male partner and to have engaged in unprotected anal sex than were bisexual men (Stokes et al. 1997). Other studies indicate that young gay men’s reports of having unprotected anal intercourse were more common when they knew their partners. Also, increased involvement in the gay community was related to higher levels of risk-taking behavior (Meyer and Dean 1995). Bisexual men use condoms inconsistently with male and female partners, seldom disclose their bisexuality to female partners, and are more likely than men who participate exclusively in same-sex sexual relations to report risk behaviors associated with HIV. The four factors found to raise HIV risk for bisexual men were (1) male prostitution, (2) injection drug use, (3) sexual identity exploration, and (4) culturally specific gender roles and norms such as those characterizing some African-American and Hispanic communities (Doll and Beeker 1996).
Sociological research on prostitution emerged from studies of crime and delinquency, with an emphasis on theories of innate criminal drives. Contemporary sociological research on prostitution more often focuses on actual behaviors. In fact, the commercial sex industry has become a renewed area of research interest and funding as it has been identified as a site of potentially high transmission of HIV. Most research on prostitution still typically relies on self-report data with small-scale, situationally specific convenience samples. Thus, Kinsey’s nonrepresentative data, which are more than forty-five years old, remain the best available comprehensive data on commercial sexual behavior. The available data on commercial sexual behavior identifies a variety of types: heterosexual prostitution of women (street prostitutes, brothels, etc.), heterosexual prostitution of men (e.g., escort services), homosexual prostitution of men (hustlers), homosexual prostitution of women, transvestite/transsexual prostitution, and the global/tourist sex industry (Brock and Thistlethwaite 1996; Perkins and Bennett 1985). Among these types of prostitution, some trends are increasingly visible.
The most historically well known form of prostitution is the heterosexual prostitution of women. Feminists offer a variety of interpretations for why women enter into prostitution, ranging from exploitation to legitimate work ( Jenness 1990). Gagnon (1977) suggests that the six most common reasons men visit prostitutes are (1) sex without negotiation, (2) sexual involvement without commitment, (3) sex for eroticism and variety, (4) a form of socializing, (5) sex away from home, and (6) sex without rejection. Some sociologists believe that prostitution serves a useful societal function by providing men with a convenient sexual outlet, which in turn minimizes the numbers of sexual transgressions against “respectable” women. Other sociologists view female prostitution as an extreme form of sexism.
Rather than addressing this debate, let us note some empirical trends. Kinsey reported that before World War II, between 60 and 70 percent of adult men had visited a prostitute and about 15 to 20 percent used them regularly. However, many commentators have indicated that rates of prostitution, a form of commercialized sex that in the past provided men with sexual opportunities in a less sexually open society, have decreased drastically since World War II. Notably, current research on sexual behavior in the context of prostitution shows the preponderance of incidents of sexual violence and the increased use of condoms. For example, Miller’s (1993) study of 16 women incarcerated for prostitution found that almost all had experienced some form of sexual assault and other violent crimes. Benson and Mathews (1995) surveyed vice squads, women working as prostitutes on the streets and in brothels, and clients and resident groups in England, finding that most street prostitutes began working in their teens and have been victims of repeated sexual and physical attacks, while the majority of clients have regular partners or are married and tend to be middle-class and middle-aged. A study in the Netherlands of 559 male clients of female prostitutes found that 14 percent of the clients do not use condoms (De Graaf et al. 1997). Typically, men who are less educated, have more commercial sex contacts, and have more contacts with prostitutes are the least likely to use condoms. It is not known if similar trends are characteristic of the heterosexual prostitution of men or the homosexual prostitution of women because there is little empirical research. Thus, it is difficult to determine whether similar patterns of sexual violence and safe sex are present across gender and sexuality lines.
The concern about the transmission of HIV/AIDS has, however, stimulated new and more comprehensive research about the homosexual prostitution of men and transvestite/transsexual prostitution. For example, Browne and Minichiello (1996) illustrate that many early studies of homosexual prostitution of men (from the 1960s to the 1980s) focused on biopsychological concerns, creating the need for future studies to focus on the behaviors and attendant risk for HIV infection of both the prostitute and the client. A study of 211 male street prostitutes and 15 male customers further makes this point (Morse et al. 1992). Morse et al. (1992) found that despite knowledge of HIV infection and its transmission, customers engage in high-risk sexual and drug use behaviors with prostitutes. However, another study contradicts these findings. Waldorf and Lauderback (1992) interviewed 552 men who solicit clients in public places (hustlers) and men who solicit by telephone and advertisements (call men) and found that 90 percent of the respondents used condoms in the past year and 75 percent used them in the past week. Similar studies of risk behavior are emerging for the transvestite/transsexual prostitution population. For example, Boles and Elifson (1994) interviewed 52 transvestite prostitutes and found that those who are socially isolated tend to be at more risk of HIV infection and those networked with nontransvestite male prostitutes tend to have a lower risk of HIV infection. Sociological research on men involved in homosexual prostitution and transvestites involved in prostitution suggests that sexual behaviors have been altered in an era of HIV/AIDS. However, little research about sexual behavior patterns outside the context of risk behaviors has been reported.
Another area of research in prostitution concerns the growing global sex industry. Sex tourism has long existed, but increasing globalization of travel and Internet advertising have sparked research in this area. Sociologists often explain the global sex industry as an outgrowth of “third world” poverty and the inequity of a global economy. Sex tourism does not involve only heterosexual prostitution of women; homosexual prostitution of men is also prevalent. Empirical studies of sexual behavior in the global sex industry remain minimal as it is illegal in many countries and most research is still done on European or American prostitution (Brock and Thistlewaite 1996).
Sociological studies of the varieties of prostitution in the United States and around the world point to a more general trend of contemporary empirical research focusing on sexual risk behaviors. Many of these studies indicate that the study of sexuality is complicated by both politics and the complexities of research methodology. However, the emergence of HIV/AIDS also reminds the sociological community of the importance of studies of sexual behavior and patterns.
This review has documented the extensive efforts of social scientists in recent decades to enhance the understanding of human sexual behavior. Indeed knowledge of sexual behavior has increased dramatically since the days of Kinsey’s early studies. This review also serves as a reminder, however, that knowledge about human sexual behavior, in some areas more than in others, remains limited because sex research is a function of both political and ethical decisions. For example, knowledge about homosexual experiences will continue to be less reliable and complete than it is for heterosexual behavior because the political commitment and considerable resources needed to secure nationally representative samples of gays, lesbians, and bisexuals are missing. Thus, just as social scientists can be held accountable for any shortcomings associated with the prevailing theoretical approaches to sexuality issues, the larger society and its institutionalized, politicized mechanisms for providing research support are responsible for impeding the research community’s efforts to understand sexual behavior. Because sexual activity tends to be a highly private social experience, social scientists’ incremental advances in documenting and explaining it are linked to the lay population’s commitment to this type of research endeavor.