Context counts: long-term sequelae of premarital intercourse or abstinence.

For a sex education program to be eligible for federal and state funding as an abstinence-only program, its curriculum must be consistent with the federal government's definition of abstinence education. According to Title V of the Social Security Act (1998), abstinence education must teach that "Sexual activity outside the context of marriage is likely to have harmful psychological and physical effects" (Sec. 510 [b] [2] [E]). Drawing from religious texts, proponents of abstinence education maintain that premarital sex precludes the development of satisfying sexual relationships in marriage, will "hinder intimacy," and results in "weak marriages, guilt, anger, and hatred" (Cornerstone Booklet Ministry, 2004; FriendsFirst, 2004). Many abstinence-only programs use scare tactics to make their case that premarital sex has negative consequences. For example, the Sex Respect program, which qualifies for federal funding under the Social Security Act, teaches "Abstinence also keeps us from developing an association between sex, guilt, fear, and shame. The negative associations can be carried into marriage, robbing us of the pleasure and fulfillment we could have enjoyed" (Mast, 1997, p. 64.).

Yet, despite continual increases in federal funding to abstinence-only programs (Rosenberg, 2002), claims that premarital sex has harmful psychological and physical effects have not been tested. For many people, the first sexual experience does not occur within marriage. Do such individuals have different developmental trajectories because they do not abstain from sex until marriage? Does premarital sex per se have negative consequences for sexual functioning, as the abstinence-only advocates claim? The current study contends that one's sexual debut is a potential turning point in sexual development. However, we propose that the context of that experience--rather than simply age or relationship status at the time--is the crucial factor in later sexual functioning.

Effects of Sexual Experience on Sexual Development

According to script theory, early sexual experiences are important in sexual development because they influence one's sexual script development (Browning & Laumann, 1997; Cate, Long, Angera, & Draper, 1993; Davidson & Moore, 1997; Gagnon & Simon, 1973). Script theory contends that sexual scripts, which are learned schemas about sex, provide meaning for the internal sensations we feel during sexual desire and arousal. Scripts organize the order of sexual acts and help the individual decode novel situations. They also set limits on sexual responses and help a person understand the socially appropriate behaviors and responses.

Although the possibility of positive effects of early sexual experiences has rarely been considered (Planned Parenthood Federation of America, 2003), research on the negative effects is plentiful. Regarding the immediate effects of the first sexual experience, responses vary. People are more likely to report guilt and less likely to report pleasure when it occurs with a casual partner (Moore & Davidson, 1997; Sprecher, Barbee, & Schwartz, 1995). Gender differences in responses to the first experience are also evident; women tend to report more guilt and men tend to report more physical pleasure (DeLamater, 1987; Donald, Lucke, Dunne, & Raphael, 1995; Guggino & Ponzetti, 1997; Holland, Ramazanoglu, & Thomson, 2000; Sprecher et al., 1995; Weis, 1983). Among women, feelings of guilt regarding the first experience are more likely when the debut occurred under the influence of drugs or alcohol or at a younger age (Moore & Davidson, 1997). The gender difference in emotional reactions to first experience may be linked to the gender difference in sexual scripts: women engage in sex with a person-centered script and men engage in sex with a body-centered script (see Dekker & Everaerd, 1989; DeLamater, 1987). That is, women are likely to have sex to strengthen relationships and increase intimacy, and men are likely to have sex to gain physical pleasure. These gender differences are consistent with men's more permissive attitudes toward premarital sex (Oliver & Hyde, 1993).

Regarding the long-term effects of early sexual experiences, research on childhood sexual abuse is informative for the current study. That is, childhood sexual abuse is often an individual's introduction to sexual behavior, and is thus a potential turning point in sexual development. Childhood sexual abuse has been linked to an increased prevalence of depression, anxiety, sexual guilt, sexual maladjustment, sexually inappropriate behavior, sexual aversion, poor self-esteem, and interpersonal relationship problems (Bartoi & Kinder, 1998; Browning & Laumann, 1997; Fleming, Mullen, Sibthorpe, & Bammer, 1999; Kendler et al., 2000; Kinzl, Traweger, & Biebl, 1995; Meston & Heiman, 2000; Meston, Heiman, & Trapnell, 1999; Noll, Trickett, & Putnam, 2003; Walser & Kern, 1996; Wyatt, Newcomb, & Riederle, 1993). Moreover, the increased prevalence of such problems in the extreme case of child sexual abuse suggests that negative sexual experiences may not only have long-term effects, but may specifically have effects that can influence later sexual functioning. The experience of sexual abuse may be so salient that individuals have difficulty altering their sexual scripts to include appropriate and consensual sexual relationships. Browning and Laumann (1997) argued that the effects of adult-child sex are indirect and mediated by the development of sexual scripts: "even if the adult-child sexual relationship is experienced as highly negative, it may be difficult for a child with no alternative model of sexual interaction to 'rewrite' the original sexual script enacted with the adult" (p. 543).

It is tempting to assume a direct causal link between childhood sexual abuse and later sexual dysfunction, but it is important to remain cautious about such associations, given the quasi-experimental designs of these studies (Hyde, 2003). A meta-analysis by Rind, Tromovitch, and Bauserman (1998) found little evidence to support the link between childhood sexual abuse and later adjustment, including sexual dysfunction. Despite serious methodological flaws in Rind et al.'s meta-analysis (Hyde, 2003), it is wise to consider the possibility that survivors of childhood sexual abuse may be more resilient to their experiences than we have theorized them to be. The current study examines the relationship between early sexual experiences, including childhood sexual abuse and other sexual activity, and later psychological and physical health, allowing for evidence of stress or resilience to emerge.

Adult sexual assault also has been linked to poorer sexual functioning (e.g., Becker, Skinner, Abel, & Cichon, 1986; Becker, Skinner, Abel, & Treacy, 1982; Golding, 1996; Wyatt et al., 1993). Wyatt and colleagues (1993) argued that for some women, rape might alter their perceptions of their own sexuality as well as of sexual relationships. For these women, sex is no longer perceived as a medium for expressing affection; it is now a medium for displaying power. In a review of research on post-sexual-assault sexuality problems, van Berlo and Ensink (2000) described the consistent, albeit quasi-experimental, finding that victims report a decrease in sexual satisfaction and frequency of sexual behavior following rape. In addition, they proposed that following rape, "sexual stimuli are perceived as anxiety provoking and are subsequently relabeled as reduced or absent altogether" (van Berlo & Ensink, 2000, p. 252), suggesting changes in sexual scripts. That is, sexual scripts following rape involve a response to sexual stimuli that includes anxiety and reduced sexual arousal.

Although rape differs from childhood sexual abuse in that the assault may not be the individual's first sexual experience, it is similar in that both involve coercion, are socially inappropriate and developmentally problematic, and represent highly salient and personally meaningful sexual encounters. These qualities make such experiences potential turning points in sexual development.

Sex Guilt and the Double Standard

Feelings of guilt regarding sex can have short-term effects; for example, such feelings may be linked to decreased sexual arousal and poorer sexual functioning immediately following the experience (Cado & Leitenberg, 1990; Darling, Davidson, & Passarello, 1992; Davidson & Moore, 1994; Walser & Kern, 1996; Wiegel, Wincze, & Barlow, 2002). Such effects may attenuate over time as the individual works through the experience (e.g., through more positive sexual experiences or effective coping efforts), or they may develop into long-term problems. Over time, feelings of sex guilt as a result of a traumatic or negative sexual experience may develop into long-term effects such as sexual dysfunction.

Some researchers argue that societal values--such as the sexual double standard--provide a link between casual sex and resulting sex guilt, contending that "sexual involvement with an occasional dating partner or a person just met is a potentially guilt-inducing behavior in a society that equates sex and love" (Moore & Davidson, 1997, p. 41). This may be particularly true for women. Reiss (1960) proposed that Americans hold men and women to different standards in their premarital sexual behavior. The traditional double standard allowed and even encouraged men to engage in premarital intercourse, whereas it prohibited women from doing the same, condemning them as immoral for such behavior (Reiss, 1960, pp. 91-92). Given the liberalization of sexual attitudes since Reiss's book was published, we might expect less endorsement of the double standard today. More recent research on the double standard of premarital sexual behavior provides mixed conclusions. For example, Milhausen and Herold (1999) found that although women perceived a societal double standard, they did not personally endorse it. Kelly and Bazzini (2001) and Gentry (1998) also found little evidence that college students endorsed the double standard. On the other hand, both Oliver and Sedikides (1992) and Hynie and Lydon (1995) found evidence for the endorsement of the double standard. Nonetheless, either personal endorsement or perception of societal endorsement of the double standard might influence whether a woman has premarital intercourse. It might also influence the way she thinks and feels about her experience, such that she might feel regret or shame about her first intercourse if it occurred before marriage.

The Current Study

We tested abstinence-only advocates' claims that premarital sex is linked to negative psychological and physical health outcomes. Specifically, we asked whether relationship status at the time of the first intercourse (i.e., marital or premarital) is important, such that participants who have their first intercourse premaritally would have more negative outcomes--for example, poorer physical health, more sexually transmitted diseases, more sexual dysfunction, more sex guilt, and poorer life satisfaction--than those who abstain until marriage. Age at first coitus was also tested as a predictor of these negative psychological and physical health outcomes.

In addition, we investigated an alternative hypothesis that the context, rather than age or relationship status at first intercourse, is an important factor in the development of psychological and physical health outcomes. Using data from the National Health and Social Life Survey (NHSLS; Laumann, Gagnon, Michael, & Michaels, 1994), we hypothesized that experiences that occurred in negative contexts (i.e., those that are likely to lead to negative cognitions or affect on the part of the person involved and affect his or her sexual script) would be associated with poorer psychological and physical health outcomes.

We characterized negative contexts somewhat differently for men and women, based on documented gender differences in sexual scripts (Dekker & Everaerd, 1989; DeLamater, 1987). The gender differences in scripts may prime men to be less affected by the context of their first sexual experience. In contrast to women, men's sexuality tends to be less subject to contextual factors, such as relationship status (Baumeister, 2000), with the result that men and women may experience sexual activity differently. Thus, the qualities that characterize a context as negative are theoretically different for men and women. Although sexual scripts are potential mediators of these relationships, they were not measured in the current study. Instead, we used script theory to inform the characterization of sexual experience.

METHOD

Participants

We analyzed the NHSLS data (Laumann et al., 1994) for this study. In an attempt to acquire a sample representative of the United States population, the NHSLS used probability sampling of households. This method excluded 3% of Americans (i.e., those living in institutional or group quarters), and resulted in a response rate of 78.6%. The NHSLS interviewed 3,432 English-speaking adults between the ages of 18 and 59 (558 were between the ages of 18-24, 1,070 were 25-34, 947 were 35-44, and 850 were 45-59). Of these participants, 56% (n = 1,912) were women and 44% (n = 1,520) were men; 53% (1,809) were married. In addition, 71.4% (n = 2,452) were White (not Hispanic), 16% (n = 550) were Black (not Hispanic), 9.4% (n = 321) were Hispanic, 1.9% (n = 67) were Asian/Pacific Islander, and 1.2% (n = 42) were Native American/Alaskan. Finally, 14.5% of participants had not graduated high school, 63% were high school graduates, 15.6% were college graduates, and 6.9% held an advanced degree.

Procedure

Interviews took place between February and October of 1992. Participants answered numerous questions regarding demographics, lifetime sexual activity, childhood, adolescence, sexual victimization, physical health, and sexual attitudes. Based on those questions, we constructed several variables for the current study.

Measures

Relationship status at first vaginal intercourse. Participants who stated that their first vaginal intercourse occurred within the context of marriage were compared with those who had premarital intercourse. Of the women in the NHSLS sample, 77.4% (n = 1,101) had had premarital intercourse, and 22.6% (n = 322) had not. Of the men, 89.5% (n = 1,064) had had premarital intercourse, and 10.5% (n = 125) had not.

Age at first vaginal intercourse. Participants were asked the age at which they first had vaginal intercourse. The mean age at first coitus was M = 17.67 years (SD = 3.25). This is consistent with other studies of age at debut (e.g., Singh, Wulf, Samara, & Cuca, 2000). Men (M = 17.24, SD = 3.39) experienced their first vaginal intercourse at an earlier age than women did (M = 18.01, SD = 3.10), F (1, 3286) = 45.73, p < .001.

We categorized participants into 3 groups (early, average, and late) based on age at their first intercourse. Because men and women differed in mean age of first intercourse, groupings were based on different ages for men and women. The early group (n = 1,000) was comprised of men and women who had their first sexual experience at a noticeably younger-than-average age for their gender; for men this was at or before age 15 (n = 422), and for women this was at or before age 16 (n = 578). The average group (n = 1,462) was comprised of participants who had their first intercourse at or around the mean age; for men this was between ages 16 and 18 (n = 634), and for women this was between ages 17 and 19 (n = 833). The late group (n = 821) was comprised of participants who had their first intercourse later than average; for men this was at or after age 19 (n = 385), and for women this was at or after age 20 (n = 436).

Context of first sexual experience. We defined the first sexual experience as the first vaginal intercourse or, if applicable, prepubertal sexual contact by an adult. Prepubertal sexual contact by an adult was included because even though it might not include intercourse, such experiences are important in sexual development, as scripting theory predicts and as past research suggests. The NHSLS did not ask a question specifically on the respondent's appraisal of the first sexual experience as we have defined it. We therefore constructed a proxy variable for assessing the context of the first sexual experience, which we scored as either positive or negative, based on its consistency with sexual scripts.

We categorized the first sexual experience as occurring in a negative context if any of the following criteria were met: (a) the first vaginal intercourse was forced by the partner (women, n = 75, 3.9%; men, n = 7, 0.5%); (b) the first vaginal intercourse was with a blood relative (women, n = 15, 0.8%; men, n = 13, 0.9%); (c) the first vaginal intercourse was with someone who paid the participant to have sex (women, n = 4, 0.2%; men, n = 2, 0.1%); (d) the main reason the participant chose to have or decided to go along with having the first vaginal intercourse was peer pressure (women, n = 141, 7.4%; men, n = 78, 5.1%) or the influence of drugs or alcohol (women, n = 33, 1.7%; men, n = 11, 0.7%); or (e) participants said that they had been touched sexually by an adult prior to puberty (women, n = 317, 16.6%; men, n = 242, 15.9%). Because men and women have different sexual scripts and motivations for sex (Dekker & Everaerd, 1989; DeLamater, 1987; Oliver & Hyde, 1993), the definition of a negative context for women (but not men) also included situations in which the first intercourse was with someone they knew but not well (n = 89, 4.7%), someone they had just met (n = 18, 0.9%), or a stranger (n = 12, 0.6%). These are situations presumed not to be consistent with a person-centered approach.

According to these criteria, 68.6% (n = 1,312) of women were categorized as having their first intercourse in a positive context, 28.5% (n = 545) of women had it in a negative context, 78.2% (n = 1,189) of men had it in a positive context, and 21.1% (n = 321) of men had it in a negative context. Because some participants refused to answer these questions or had not had sexual intercourse, 2.9% (n = 55) of women and 0.7% (n = 10) of men had missing data and were excluded from further analyses.

Sexual dysfunction. Participants responded "yes" or "no" to seven questions regarding sexual functioning during the previous 12 months. Questions assessed lack of interest in sex, inability to come to climax (experience an orgasm), coming to climax (experiencing an orgasm) too quickly, experiencing physical pain during intercourse, not finding sex pleasurable (even if it was not painful), feeling anxious just before having sex about own ability to perform sexually, and having trouble maintaining an erection (for men only), or having trouble lubricating (for women only). For each question to which the participant responded "yes," he or she received one point. These points were summed so that a score of 0 reflected no dysfunctions and a score of 7 reflected seven dysfunctions. Because the distribution was positively skewed, a square root transformation was used to normalize it. Of the full sample, 1.7% (n = 57) of participants chose not to answer questions regarding sexual dysfunction; they were omitted from further analyses.

Sexual guilt. As a measure of sex guilt, we used responses to the question, "How often does thinking about sex make you feel guilty?" Participants responded using a Likert scale ranging from 1 (never) to 5 (nearly always). Of the full sample, 79.4% (n = 2,725) of participants responded "never" to this question. Because the distribution was positively skewed, we used a log transformation to normalize it.

Overall health. Participants responded to a single-item rating of overall physical health, scaled from 1 (excellent) to 4 (poor). We reversed health ratings such that higher scores reflected better health.

Lifetime STDs. Participants responded "yes" or "no" to questions regarding the lifetime incidence of sexually transmitted diseases (STD). Questions assessed gonorrhea, syphilis, herpes, chlamydia, genital warts, hepatitis, and AIDS or HIV. For each STD the participant reported ever having, he or she received one point. We summed these points such that a score of 0 reflected no STDs and a score of 7 reflected seven STDs. Because the distribution was positively skewed, a log transformation was used to normalize it.

Life satisfaction. Participants responded to a single-item rating life satisfaction in the past year, scaled from 1 (extremely happy) to 5 (unhappy most of the time). The item read, "Generally, how happy have you been with your personal life during the past 12 months?" We reversed life satisfaction ratings such that higher scores reflected greater satisfaction.

Statistical Analyses

We used the multivariate analysis of covariance (MANCOVA) model to examine mean differences between groups categorized according to relationship status at first intercourse, age (early, average, or late) at first intercourse, and context of first experience. The relationship status at first intercourse (premarital or marital), age at first intercourse, or context of first sexual experience was used as the independent variable in 3 individual MANCOVAs; gender was used as a second independent variable in each of the MANCOVAs. The five dependent variables included sum of sexual dysfunctions, sex guilt, overall health, sum of STDs, and life satisfaction.

We computed correlations among education, age, and alcohol consumption and all dependent variables to determine covariate use. Education was negatively correlated with sum of sexual dysfunctions (r [3,354] = -.11, p < .001) but positively correlated with sex guilt (r [3,287] = .04, p < .05), health (r [3,405] = .24, p < .001), STDs (r [3,405] = .04, p < .05), and life satisfaction (r [3,401] = .04, p < .05). In addition, age was negatively correlated with sex guilt (r [3,306] = -.09, p < .001), health (r [3,424] = -.14, p < .001), STDs (r [3,405] = .04, p < .05), and life satisfaction (r [3,401] = .04, p < .05). Report of alcohol consumption in the past year, which was scored from 1 (not at all) to 5 (daily), was positively correlated with health (r [3,416] = .09, p < .001) and STDs (r [3,413] = .10, p < .001), but negatively correlated with life satisfaction (r [3,401] = -.05, p < .05). Thus, although correlations are modest, education, age, and alcohol consumption served as covariates in all MANCOVAs.

RESULTS

Descriptive statistics regarding sexual dysfunction in the NHSLS sample are provided in Table 1. Descriptive statistics regarding sex guilt, overall health, STDs, and happinesslife satisfaction are shown in Tables 2, 3, 4, and 5, respectively.

Long-Term Sequelae of Premarital Intercourse or Abstinence

Is premarital sex linked to negative psychological and physical health outcomes? We ran a 2 (gender) x 2 (relationship status at first intercourse: marital vs. premarital) between-subjects MANCOVA (education, age, and alcohol consumption as covariates) with five dependent variables (sexual dysfunction, sexual guilt, STDs, overall health, and life satisfaction). Participants whose first intercourse was coerced (i.e., forced or prepubertal with an adult; n = 618, 18.0%) were excluded from these analyses because their experiences were likely distinct from those of people who chose to have premarital intercourse. That is, the abstinence-only agenda refers to choosing premarital abstinence; insofar as rape or childhood sexual abuse victims did not choose to have coercive first experiences, their experiences should not be categorized the same way as those of individuals who chose to have premarital intercourse. Cell means and standard deviations for dependent variables according to relationship status at the time of the first intercourse are shown in Table 6. The multivariate test of the interaction between gender and relationship status at time of first intercourse was not significant, F (5, 2521) = .71, p > .05. The multivariate test of gender was significant, F (5, 2521) = 51.30, p < .001. Women had significantly more sex dysfunctions (F [1, 2525] = 16.60, p < .001), less sex guilt (F [1, 2525] = 5.16, p < .05), and less life satisfaction than men (F [1, 2525] = 4.16, p < .05). Men and women did not differ significantly on overall health (F [1, 2525] = 0.26, p >. 05) or STDs (F [1, 2525] = 1.52, p > .05).

The multivariate test of relationship status at time of first intercourse was significant, F (5, 2521) = 8.31, p < .001. Univariate F-tests of relationship status and the associated effect sizes are shown in Table 6. Individuals who had premarital sex did not differ significantly from those who abstained until marriage on sex dysfunction, sex guilt, or overall health. However, those who had premarital sex had significantly more STDs and less life satisfaction. These effects are considered small to moderate by Cohen's (1988) standards.

As a result of the sexual double standard, women who had premarital intercourse prior to the sexual revolution may have had different experiences from those women who had premarital sex after the sexual revolution. To investigate further the potential role of the sexual double standard in the development of negative outcomes for women who had premarital intercourse, we conducted a cohort analysis. Participants ranged in age from 18 to 59 at the time of interview. Women were separated into two groups: those who came of age before the sexual revolution (i.e., they were at least 18 years old in 1970) and those who came of age after the sexual revolution. We ran a 2 (cohort: came of age before v. after the sexual revolution) x 2 (relationship status at first intercourse) MANCOVA (education, age, and alcohol consumption used as covariates) with women, using sex dysfunction, sex guilt, STDs, overall health, and life satisfaction as dependent variables. Women whose first experience was coercive were excluded. If the timing of the debut (i.e., relationship status at first intercourse) in relation to the sexual revolution is important in the development of sexual dysfunction, the interaction of cohort and relationship status at first intercourse should he significant, with the worst outcomes among women who had premarital sex before the sexual revolution.

The multivariate interaction of cohort and relationship status at first intercourse was significant, F (5, 1351) = 3.05, p < .01. We found significant results for univariate tests with sex dysfunction (F [1, 1355] = 3.76, p < .05), overall health (F [1, 1355] = 9.64, p < .01), and life satisfaction (F [1, 1355] = 4.82, p < .05), but not with sex guilt (F [1, 1355] = .91, p > .05) or STDs (F [1, 1355] = .73, p > .05). See Figure 1 for graphs of the significant interactions. The pattern of interactions indicated that for women who came of age after the sexual revolution, premarital sex was associated with worse outcomes. This pattern is inconsistent with the hypothesis that women who had premarital sex prior to the sexual revolution would have poorer outcomes than women who had premarital sex after the revolution. Although age at time of interview was statistically controlled, cohort differences are inextricably linked to age in that older participants are grouped in Cohort 1 and younger participants are grouped in Cohort 2. To the extent that age and cohort are linked, these results should be interpreted cautiously.

[FIGURE 1 OMITTED]

Long-Term Sequelae of Age at First Intercourse

To test the association between age at first intercourse and later psychological and physical functioning, we ran a 2 (gender) x 3 (age at first intercourse: early, average, or late) MANCOVA (education, age, and alcohol consumption as covariates) with 5 dependent variables (sexual dysfunction, sexual guilt, STDs, overall health, and life satisfaction). As with the analyses of relationship status at time of debut, participants whose first intercourse was coerced (n = 618, 18.0%) were excluded from these analyses. Cell means and standard deviations for all 5 dependent variables are shown in Table 7.

The multivariate test of the interaction between gender and age at time of first intercourse was not significant, F (10, 5092) = 1.28, p > .05. As in the previous analysis, the multivariate test of gender was significant, F (5, 2545) = 10.76, p < .001. Women had significantly more sex dysfunction (F [1, 2549] = 43.33, p < .001), less sex guilt (F [1, 2549] = 6.31, p < .05), and less life satisfaction than men did (F [1, 2549] = 3.29, p < .05). Men and women did not differ significantly on overall health (F [1, 2549] = .69, p > .05) or STDs (F [1, 2549] = 1.02, p > .05).

The multivariate test of age at time of first intercourse was significant, F (10, 5092) = 18.95, p < .001. Univariate F-tests of age at debut and the associated effect sizes and Bonferroni post-hoc tests of significance are shown in Table 7. Individuals who had their sexual debuts at early ages had significantly less sex guilt, poorer overall health, and more STDs than individuals who had their debuts at average or late ages. There were no significant group differences in sex dysfunction or life satisfaction. In sum, age at first intercourse was consistently linked to 3 of 5 outcomes, but the direction and pattern of the effects is unreliable and inconsistent with the abstinence-only agenda. This issue will be addressed further in the Discussion.

Long-Term Sequelae of Context of First Intercourse

To investigate the contribution of the context of the first sexual experience to later psychological and physical functioning, we ran a 2 (gender) x 2 (context: positive vs. negative first experience) MANCOVA (education, age, and alcohol consumption as covariates) with five dependent variables (sexual dysfunction, sexual guilt, STDs, overall health, and life satisfaction). Cell means and standard deviations for these dependent variables are shown in Table 8. The multivariate test of the interaction between gender and context was not significant, F (5, 3173) = .92, p > .05. The multivariate test of gender was significant, F (5, 3173) = 12.19, p < .001. Women had significantly more sex dysfunctions (F (1, 3177) = 42.22, p < .001), less sex guilt (F [1, 3177] = 11.28, p < .001), and fewer STDs (F [1, 3177] = 3.87, p < .05) than did men. Men and women did not differ significantly in overall health (F [1, 3177] = 3.17, p > .05) or life satisfaction (F [1, 3177] = .84, p > .05).

The multivariate test of context of first intercourse was significant, F (5, 3173) = 23.90, p < .001. Univariate F-tests of context and the associated effect sizes are shown in Table 8. Consistent with our hypotheses, the results indicate that negative contexts were significantly linked to more sex dysfunctions, more sex guilt, poorer health, more STDs, and less life satisfaction--that is, to all five outcomes.

DISCUSSION

Is Abstinence the Answer?

While federal funding for abstinence-only education increases to $102 million annually (Dailard, 2002), a debate rages regarding the effectiveness of abstinence-only education versus comprehensive sexuality education (US Surgeon General, 2001). Despite the surplus of evidence indicating the inefficacy of abstinence-only programs (e.g., Kirby, 2002; Silva, 2002), there is no source of federal funding for comprehensive sexuality education programs (Sexuality Information and Education Council of the United States, 2001). The research reported here provides an empirical test of one of the major claims made by advocates of abstinence-only education. To investigate such claims, we assessed the sequelae of premarital sexual intercourse for the outcomes of sexual dysfunction, sex guilt, overall health, incidence of STDs, and life satisfaction. Two outcomes--STDs and life satisfaction--were linked to premarital sex to some extent. However, variables not accounted for in the current study may explain these findings. For example, greater number of lifetime sexual partners is expected among those who have premarital intercourse; by increasing the likelihood of contracting an STD, premarital sex is likely indirectly linked to STD incidence. In addition, religious beliefs may both guide premarital sex decisions and affect life satisfaction, providing a potentially spurious link between premarital sex and life satisfaction. The results suggest that premarital sex per se does not result in later sexual dysfunction, feelings of sex guilt, or poor health, though it appears to be linked in some way to STDs and life satisfaction. Because of the correlational nature of the data, causal inferences between premarital sex and negative outcomes are unwarranted.

The cohort analysis/did not suggest that women who had premarital sex before the sexual revolution had worse outcomes than women who had premarital sex after the sexual revolution did. Indeed, the pattern of interactions indicated just the opposite: women who had premarital sex after the sexual revolution had worse outcomes than women who had premarital sex prior to the sexual revolution. The confound between age and cohort is unavoidable with the current design, and conclusions from the cohort analysis should be viewed cautiously.

Results regarding the relationship between age at first coitus and later outcomes, although statistically significant, were inconsistent with the abstinence-only agenda. Although individuals who had their first intercourse at early ages reported slightly poorer health, more STDs, and slightly poorer life satisfaction, they actually reported healthier feelings regarding sex guilt than those who had later debuts. In addition, sex dysfunction was not linked to age at first intercourse. The results suggest no clear or consistent pattern of effects of having sex at younger ages. In light of the effects of context of early sexual experiences, it is apparent that having sex at a developmentally immature stage is potentially harmful. These results should be interpreted cautiously, especially given the correlational nature of the research design. Moreover, the finding that individuals who have their first intercourse at older ages are likely to report more sex guilt than those who had their first intercourse at younger ages is inconsistent with the abstinence-only agenda. It is plausible that high levels of sex guilt play a role in postponing the first intercourse. That is, feelings of guilt about sexual activity may create anxiety in sexual contexts or discourage an individual from engaging in sexual activity. This causal relationship cannot be tested in the current study, but has been supported by other research (Long, Care, Fehsenfeld, & Williams, 1996; Sack, Keller, & Hinkle, 1984). The link between age at first intercourse and STD prevalence may be explained by poor use of condoms by adolescents (Morrison, 1985). Generally, abstinence-only programs cannot provide education regarding condoms and their correct usage.

Does Context Count?

Analyses of the potential effects of relationship status at first intercourse and age at first intercourse indicate that if early sexual experiences are antecedent to later psychological and physical health outcomes, the relationship is more complex than the abstinence-only advocates claim. With this in mind, we examined the context of those experiences as a potential predictor. The contexts of the first sexual experiences were categorized as negative if the experience was prepubertal, forced, with a blood relative, or the result of drugs or alcohol, if the individual was paid for sex, or, for women, if the sex partner was a stranger or not well-known to the individual. That is, the contexts were categorized as negative if they were inconsistent with sexual scripts. First sexual experiences occurring in negative contexts were associated consistently with greater sexual dysfunction, greater sex guilt, poorer health, more STDs, and less life satisfaction. Sex that is nonconsensual, occurs before puberty, is with a blood relative or stranger, or occurs as a result of drugs, alcohol, or peer pressure is reliably associated with development of these negative outcomes, even after controlling for SES. Although these results are inconsistent with Rind et al. (1998), who claimed that childhood sexual contact was only weakly associated with negative outcomes, they are consistent with a great deal of past research in identifying an association between coercive and inappropriate sexual contact with later psychological problems (e.g., Browning & Laumann, 1997; Kendler et al., 2000; Wyatt et al., 1993). These results lend support to the hypothesis that sexual experiences occurring in contexts inconsistent with gender-typical sexual scripts are linked to negative outcomes. The use of scripting theory in the characterization of contexts was validated by the significant effects that we found and suggests future research should be aimed at linking early experiences to sexual dysfunction, guilt, STDs, health, and life satisfaction via script development.

Limitations

Like all others, the current study has several limitations. First, the data on the sexual debut were collected retrospectively, in some cases more than 40 years after the first sexual experience. Ideally, research on this question should be conducted prospectively to reduce measurement error in reporting of such meaningful experiences. Although the events are highly salient and the data were carefully collected using probability sampling and in-depth interviews, participants may have remembered events or aspects of those events incorrectly or may have engaged in some restructuring of the events. Interviewing participants earlier (i.e., soon after first coitus) should reduce measurement error due to cognitive restructuring over time. Some research indicates that individuals' recollections of their sexual debuts are unreliable (Turner et al., 1998; Upchurch, Lillard, Aneshensel, & Li, 2002; Wu, Martin, & Long, 2001). However, inconsistencies in teenagers' reports of sexual behavior appear to be random and not problematic for conclusions about age at first intercourse (Upchurch et al., 2002). In addition, the magnitude of the inconsistencies is generally very small (Wu et al., 2001). Interviewers rated participants' apparent frankness in responding to questions; over 95% of participants were rated as "mostly" or "entirely" frank in their responses. To the extent that interviewers' ratings and participants' memories are accurate, the results of the current study are valid.

A second limitation is that the current study did not specifically address the role of sexual orientation. By examining the links between first vaginal intercourse and development, we respond to abstinence-only programs but also define sexual experiences with a heterosexual lens. Although specific sexual experiences may differ, the developmental path we have described with scripting theory should hold regardless of sexual orientation. Nonetheless, future research might examine this developmental path within lesbian, gay, and bisexual individuals as well.

The variables assessed by Laumann et al. (1994) were mostly sociological in nature. A more psychological study of these issues might highlight the personal factors involved, such as history of depression and anxiety as well as medication use. In particular, no direct measure of participants' sexual scripts was collected. The results of the current study suggest that context has a clear link to adult functioning; however, this link may be better understood with the use of validated measurement tools. For example, a validated depression scale is preferable over the single-item measure of life satisfaction used here.

Given the correlational nature of the data, there are potential alternative explanations for the current findings. One plausible explanation is that a third factor, such as personality, may cause individuals both to put themselves in negative contexts and to develop sexual dysfunction, sex guilt, poor health, STDs, and poor life satisfaction (Caspi et al., 1997). This explanation was not testable in the current study, but does suggest a future line of research.

Future Directions

Future research should focus on scripts as hypothesized mediators of the relationship between the contexts of first sexual experiences and later psychological and physical health outcomes. Sexual experiences occurring in contexts that create a negative sexual script may be linked to both immediate and long-term negative outcomes. Negative or positive cognitions or affect as a result of the qualities of the first sexual experience may mediate theis relationship; for example, anxiety or shame associated with previous experiences may inhibit sexual arousal in potentially positive contexts.

If data were collected longitudinally, preexisting cognitions and associated emotions regarding sexuality could be assessed. For example, the roles of depression, anxiety, and drug or alcohol use could be explored further. Identification of sexual scripts and their potential for change or development could also be assessed in a longitudinal design. For example, do sexual scripts change in response to experiences in positive or negative contexts?

It is important to test empirically all assertions of the abstinence-only education programs. Are the other components of Title V scientifically based and medically accurate? Title V of the Social Security Act (1998) also requires that abstinence programs teach "abstinence from sexual activity outside marriage as the expected standard for all school age children" (See. 510 [b] [2] [B]) and that "bearing children out-of-wedlock is likely to have harmful consequences for the child, the child's parents, and society" (See. 510 [b] [2] [F]). Such educational goals are derived from religious texts and moral or political beliefs rather than from adequate public health research. These educational goals dictate lifestyle choices that are inconsistent with those made by average Americans, as we can see from the NHSLS data. When investigating the effects of abstinence-only sex education programs, it is important to look for both positive and negative effects. Curricula using fear and shame may do more to foster negative attitudes about sex, such as sex guilt, than to prevent inappropriate or premature sexual experiences. In addition, curricula that fail to teach safer sex practices in favor of teaching abstinence may actually increase risk for STDs and unwanted pregnancy by neglecting to prepare youth for sexual behavior.

Finally, because negative contexts are associated with poorer sexual functioning, more sex guilt, more STDs, poorer health, and less life satisfaction, researchers and clinicians should investigate potential interventions both to prevent the negative contexts and to promote the development of healthy sexual functioning and behavior. Instead of increasing federal funding to medically inaccurate educational programs, legislators should increase funding to services designed to aid sexually victimized youth and to improve Americans' knowledge of sexual health.

Conclusion

The current findings indicate that characterizing premarital sex as necessarily harmful is misleading and inaccurate. Although abstinence-only advocates claim that premarital sex is detrimental to well being, there is little evidence of this in the current study. Insofar as first sexual experiences are related to later functioning, the context of the experience is the crucial element.

Table 1. Number of Sexual Dysfunctions During Past 12 Months
in the NHSLS Sample

                           0             1            2

Women   (n = 1,871)    842 (45.0)    392 (21.0)   236 (12.6)
Men     (n = 1,504)    831 (55.3)    335 (22.3)   157 (10.4)

Total   (n = 3,375)   1673 (49.57)   727 (21.5)   393 (11.6)

            3           4          5          6          7

Women   170 (9.1)   122 (6.5)   75 (4.0)   27 (1.4)   7 (0.4)
Men      88 (5.9)    60 (4.0)   17 (1.1)   15 (1.0)   1 (0.1)

Total   258 (7.6)   182 (5.4)   92 (2.7)   42 (1.2)   8 (0.2)

Note. Sample sizes may vary across analyses as a result of
missing data. Percentages are in parentheses.

Table 2. Sex Guilt in the NHSLS Sample

                         Never        Rarely

Women   (n = 1,815)   1538 (84.7)   150 (8.3)
Men     (n = 1,494)   1188 (79.5)   183 (12.2)

Total   (n = 3,309)   2726 (82.4)   333 (10.1)

        Occasionally    Often     Almost always

Women     97 (5.3)     18 (1.0)     12 (0.7)
Men       99 (6.6)     10 (0.7)     14 (0.9)

Total    196 (5.9)     28 (0.8)     26 (0.8)

Note. Responses to the question, "How often does
thinking about sex make you feel guilty?" Sample
sizes may vary across analyses as a result of
missing data. Percentages are in parentheses.

Table 3. Self-reported Overall Health in the NHSLS Sample

                        Poor        Fair         Good        Excellent

Women   (n = 1,535)   30 (2.0)   154 (10.0)    701 (45.7)    650 (42.3)
Men     (n = 1,271)   19 (1.5)   101 (7.9)     574 (45.2)    577 (45.4)

Total   (n = 2,806)   49 (1.7)   255 (9.1)    1275 (45.4)   1227 (43.7)

Note. Sample sizes may vary across analyses as a result of missing
data. Percentages are in parentheses.

Table 4. Number of STDs Ever Diagnosed in the NHSLS Sample

                           0            1            2

Women   (n = 1,908)   1542 (80.8)   274 (14.4)    77 (4.0)
Men     (n = 1,519)   1247 (82.1)   188 (12.4)    74 (4.9)

Total   (n = 3,427)   2789 (81.4)   462 (13.5)   151 (4.4)

           3          4         5        6        7

Women   11 (0.6)   2 (0.1)   0 (0)     0 (0)   2 (0.1)
Men      6 (0.4)   3 (0.2)   1 (0.1)   0 (0)   0 (0)

Total   17 (0.5)   5 (0.1)   1 (0.1)   0 (0)   2 (0.1)

Note. Sample sizes may vary across analyses as a result of
missing data. Percentages are in parentheses.

Table 5. Life Satisfaction in the NHSLS Sample

                      Unhappy most     Sometimes
                      of the time    fairly unhappy

Women   (n = 1,908)     66 (3.5)       209 (11.0)
Men     (n = 1,517)     40 (2.6)       139 (9.2)
Total   (n = 3,425)    106 (3.1)       348 (10.2)

        Generally
        satisfied,      Very happy      Extremely
         pleased     most of the time     happy

Women   542 (28.4)      782 (41.0)      309 (16.2)
Men     451 (29.7)      625 (41.2)      262 (17.3)
Total   993 (29.0)     1407 (41.1)      571 (16.7)

Note. Responses to the question, "Generally, how happy
have you been with your personal life during the past
12 months?" Sample sizes may vary across analyses as a
result of missing data. Percentages are in parentheses.

Table 6. Sex Dysfunction, Sex Guilt, Health, and General Life
Satisfaction According to Relationship Status at Time of Debut

                                       Sex dysfunction   Sex guilt
                                          Mean (SD)      Mean (SD)

Female
  Premarital             (n = 1,067)     0.43 (.49)      1.05 (.14)
  Marital                  (n = 296)     0.36 (.43)      1.05 (.15)
Male
  Premarital             (n = 1,053)     0.28 (.38)      1.07 (.15)
  Marital                  (n = 120)     0.29 (.42)      1.08 (.16)
Total
  Premarital             (n = 2,120)     0.36 (.45)      1.06 (.15)
  Marital                  (n = 416)     0.34 (.43)      1.06 (.15)
Univariate test (F) of
  relationship status
  at debut                                .41             .48
Effect size (d) of
  relationship
  status at debut                         .05             .00

                                                        Life
                           Health        STDs       satisfaction
                         Mean (SD)     Mean (SD)     Mean (SD)

Female
  Premarital             3.32 (.68)    0.08 (.15)    3.59 (.96)
  Marital                3.23 (.79)    0.02 (.08)    3.78 (.95)
Male
  Premarital             3.37 (.66)    0.07 (.14)    3.67 (.92)
  Marital                3.29 (.79)    0.01 (.07)    3.89 (.90)
Total
  Premarital             3.34 (.67)    0.07 (.15)    3.63 (.94)
  Marital                3.25 (.79)    0.02 (.07)    3.81 (.94)
Univariate test (F) of
  relationship status
  at debut                .07         27.10 ***     14.09 ***
Effect size (d) of
  relationship
  status at debut         .01           .45           .19

Note. Degrees of freedom = 1, 2525 for univariate F tests.

*** p < .001.

Table 7. Sex Dysfunction, Sex Guilt, Health, and General Life
Satisfaction According to Age at Debut

                                      Sex dysfunction   Sex guilt
                                         Mean (SD)      Mean (SD)

Female
  Early                   (n = 394)     0.46 (.52)      1.05 (.14)
  Average                 (n = 649)     0.44 (.48)      1.06 (.14)
  Late                    (n = 341)     0.37 (.46)      1.07 (.17)
Male
  Early                   (n = 308)     0.34 (.43)      1.06 (.14)
  Average                 (n = 541)     0.29 (.38)      1.08 (.16)
  Late                    (n = 325)     0.33 (.44)      1.09 (.17)
Total
  Early                   (n = 702)     0.43 (.49)      1.05 (.14)
  Average               (n = 1,190)     0.37 (.44)      1.07 (.15)
  Late                    (n = 666)     0.35 (.45)      1.08 (.17)
Univariate test (F)
  of age at debut              0.25    13.41 ***        3.77 *
Effect size (d)
  of age at debut (a)           .13      .14 ***         .13 *
Effect size (d)
  of age at debut (b)           .17      .19 ***         .21 ***
Effect size (d)
  of age at debut (c)           .04      .06             .09

                                                       Life
                          Health         STDs      satisfaction
                         Mean (SD)    Mean (SD)     Mean (SD)

Female
  Early                  3.18 (.72)   0.12 (.18)     3.50 (1.06)
  Average                3.27 (.73)   0.06 (.13)     3.60 (.98)
  Late                   3.39 (.67)   0.04 (.11)     3.63 (.92)
Male
  Early                  3.28 (.71)   0.12 (.19)     3.50 (.96)
  Average                3.37 (.65)   0.05 (.13)     3.71 (.94)
  Late                   3.37 (.70)   0.03 (.11)     3.66 (.91)
Total
  Early                  3.23 (.72)   0.12 (.18)     3.50 (1.02)
  Average                3.32 (.70)   0.06 (.13)     3.65 (.96)
  Late                   3.38 (.69)   0.04 (.11)     3.65 (.91)
Univariate test (F)
  of age at debut       79.24 ***     3.29 *
Effect size (d)
  of age at debut (a)     .39 ***      .15
Effect size (d)
  of age at debut (b)     .55 ***      .16
Effect size (d)
  of age at debut (c)     .17 **       .00

Note. Degrees of freedom = 2, 2549 for univariate F tests.

(a) Difference between early and average.

(b) Difference between early and late.

(c) Difference between average and late.

* p < .05 ** p < .01 *** p < .001

Table 8. Sex Dysfunction, Sex Guilt, Health, and General Life
Satisfaction According to Context of First Sexual Experience

                                    Sex dysfunction    Sex guilt
                                       Mean (SD)       Mean (SD)

Female
  Positive            (n = 1,220)     0.40 (.47)       1.05 (.14)
  Negative              (n = 508)     0.54 (.53)       1.08 (.16)
Male
  Positive            (n = 1,145)     0.28 (.39)       1.07 (.16)
  Negative              (n = 311)     0.43 (.47)       1.09 (.17)
Total
  Positive            (n = 2,365)     0.34 (.44)       1.06 (.15)
  Negative              (n = 819)     0.50 (.51)       1.08 (.17)
Univariate test (F)
  of context                         56.95 ***        14.15 ***
Effect size (d) of
  context                              .34              .13

                                                       Life
                        Health         STDs        satisfaction
                       Mean (SD)     Mean (SD)      Mean (SD)

Female
  Positive             3.31 (.71)    0.06 (.13)    3.64 (.97)
  Negative             3.18 (.75)    0.11 (.18)    3.40 (1.04)
Male
  Positive             3.35 (.68)    0.06 (.14)    3.68 (.93)
  Negative             3.29 (.72)    0.09 (.16)    3.39 (1.00)
Total
  Positive             3.33 (.69)    0.06 (.14)    3.66 (.95)
  Negative             3.22 (.74)    0.10 (.17)    3.39 (1.02)
Univariate test (F)
  of context          17.66 ***     35.01 ***     43.40 ***
Effect size (d) of
  context               .15           .26           .27

Note. Degrees of freedom = 1, 3177 for univariate F tests.

*** p < .001

Note: Editorial processing of this manuscript was carried out by Associate Editor Lucia O'Sullivan.

We are grateful to Edward Laumann, John Gagnon, Robert Michael, Stuart Michaels, and their team for collecting the NHSLS data set and making it publicly available.

REFERENCES

Bartoi, M. G., & Kinder, B. N. (1998). Effects of child and adult sexual abuse on adult sexuality. Journal of Sex and Marital Therapy, 24, 75-90.

Baumeister, R. F. (2000). Gender differences in erotic plasticity: The female sex drive as socially flexible and responsive. Psychological Bulletin, 126, 347-374.

Becker, J. V., Skinner, L. J., Abel, G. G., & Cichon, J. (1986). Level of post-assault sexual functioning in rape and incest victims. Archives of Sexual Behavior, 15, 37-49.

Becker, J. V., Skinner, L. J., Abel, G. G., & Treacy, E. C. (1982). Incidence and types of sexual dysfunctions in rape and incest victims. Journal of Sex and Marital Therapy, 8, 65-74.

Browning, C.R., & Laumann, E. O. (1997). Sexual contact between children and adults: A life course perspective. American Sociological Review, 62, 540-560.

Cado, S., & Leitenberg, H. (1990). Guilt reactions to sexual fantasies during intercourse. Archives of Sexual Behavior, 19, 49-63.

Caspi, A., Begg, D., Dickson, N., Harrington, H., Langley, J., Moffitt, T. E., et al. (1997). Personality differences predict health-risk behaviors in young adulthood: Evidence from a longitudinal study. Journal of Personality and Social Psychology, 73, 1,052-1,063.

Cate, R., Long, E., Angera, J., & Draper, K. K. (1993). Sexual intercourse and relationship development. Family Relations, 42, 158-164.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed). Hillsdale, NJ: Erlbaum.

Cornerstone Booklet Ministry. (2004). Premarital sex. Retrieved March 29, 2004, from http://www.greenwichbaptistchurch.org/premsex.html.

Dailard, C. (2002). Abstinence promotion and teen family planning: The misguided drive for equal funding. The Guttmacher Report on Public Policy, 5, 1-3.

Darling, C. A., Davidson, J. K., & Passarello, L. C. (1992). The mystique of first intercourse among college youth: The role of partners, contraceptive practices, and psychological reactions. Journal of Youth & Adolescence, 21, 97-117.

Davidson, J. K., & Moore, N. B. (1994). Guilt and lack of orgasm during sexual intercourse: Myth versus reality among college women. Journal of Sex Education & Therapy, 20, 153-174.

Dekker, J., & Everaerd, W. (1989). Psychological determinants of sexual arousal: A review. Behaviour Research and Therapy, 27, 353-364.

DeLamater, J. (1987). Gender differences in sexual scenarios. In K. Kelley (Ed.), Females, males, and sexuality (pp. 127-140). Albany, NJ: SUNY Press.

Donald, M., Lucke, J., Dunne, M., & Raphael, B. (1995). Gender differences associated with young people's emotional reactions to sexual intercourse. Journal of Youth and Adolescence, 24, 453-464.

Fleming, J., Mullen, P. E., Sibthorpe, B., & Bammer, G. (1999). The long-term impact of childhood sexual abuse in Australian women. Child Abuse & Neglect, 23, 145-149.

Friends First (2004). Benefits of marriage. Retrieved March 29, 2004, from http://www.friendsfirst.org/factmarriage.html.

Gagnon, J. H., & Simon, W. (1973). Sexual conduct: The social sources of human sexuality. Chicago: Adline.

Gentry, M. (1998). The sexual double standard: The influence of number of relationships and level of sexual activity on judgments of women and men. Psychology of Women Quarterly, 22, 505-511.

Golding, J. M. (1996). Sexual assault history and women's reproductive and sexual health. Psychology of Women Quarterly, 20, 101-121.

Guggino, J. M., & Ponzetti, J. J. (1997). Gender differences in affective reactions to first coitus. Journal of Adolescence, 20, 189-200.

Holland, J., Ramazanoglu, C., Sharpe, S., & Thomson, R. (2000). Deconstructing virginity: Young people's accounts of first sex. Sexual and Relationship Therapy, 15, 221-232.

Hyde, J. S. (2003). The use of meta-analysis in understanding the effects of child sexual abuse. In J. Bancroft (Ed.), Sexual development. Bloomington, IN: Indiana University Press.

Hynie, M., & Lydon, J. E. (1995). Women's perceptions of female contraceptive behavior. Psychology of Women Quarterly, 19, 563-581.

Kelly, J., & Bazzini, D. G. (2001). Gender, sexual experience, and the sexual double standard: Evaluations of female contraceptive behavior. Sex Roles, 45, 785-799.

Kendler, K. S., Bulik, C. M., Silberg, J., Hettema, J. M., Myers, J., & Prescott, C. A. (2000). Childhood sexual abuse and adult psychiatric and substance use disorders in women: An epidemiological and co-twin control analysis. Archives of General Psychiatry, 57, 953-959.

Kinzl, J. F., Traweger, C., & Biebl, W. (1995). Sexual dysfunctions: Relationship to childhood sexual abuse and early family experiences in a nonclinical sample. Child Abuse and Neglect, 19, 785-792.

Kirby, D. (2002). Effective approaches to reducing adolescent unprotected sex, pregnancy, and childbearing. The Journal of Sex Research, 39, 51-57.

Laumann, E. O., Gagnon, J. H., Michael, R. T., & Michaels, S. (1994). The social organization of sexuality: Sexual practices in the United States. Chicago: University of Chicago Press.

Long, E. C., Cate, R. M., Fehsenfeld, D. A., & Williams, K. M. (1996). A longitudinal assessment of a measure of premarital sexual conflict. Family Relations, 45, 302-308.

Mast, C. K. (1997). Sex respect: The option of true sexual freedom. Bradley, IL: Respect Incorporated.

Meston, C. M., & Heiman, J. R. (2000). Sexual abuse and sexual function: An examination of sexually relevant cognitive processes. Journal of Consulting and Clinical Psychology, 68, 399-406.

Meston, C. M., Heiman, J. R., & Trapnell, P. D. (1999). The relation between early abuse and adult sexuality. The Journal of Sex Research, 36, 385-395.

Milhausen, R. R., & Herold, E. S. (1999). Does the sexual double standard still exist? Perceptions of university women. The Journal of Sex Research, 36, 361-368.

Moore, N. B., & Davidson, J. K. (1997). Guilt about first intercourse: An antecedent of sexual dissatisfaction among college women. Journal of Sex & Marital Therapy, 23, 29-46.

Morrison, D. M. (1985). Adolescent contraceptive behavior: A review. Psychological Bulletin 98, 538-568.

Noll, J. G., Trickett, P. K., & Putnam, F. W. (2003). A prospective investigation of the impact of childhood sexual abuse on the development of sexuality. Journal of Consulting and Clinical Psychology, 71,575-586.

Oliver, M. B., & Hyde, J. S. (1993). Gender differences in sexuality: A meta-analysis. Psychological Bulletin. 114, 29-51.

Oliver, M. B., & Sedikides, C. (1992). Effects of sexual permissiveness on desirability of partner as a function of low and high commitment to relationship. Social Psychology Quarterly, 55, 321-333.

Planned Parenthood Federation of America. (2003). White paper: The health benefits of sexual expression. New York: Katherine Dexter McCormick Library.

Reiss, I. L. (1960). Premarital sexual standards in America. Glencoe, IL: Free Press.

Rind, B., Tromovitch, P., & Bauserman, R. (1998). A meta-analytic examination of assumed properties of child sexual abuse using college samples. Psychological Bulletin, 124, 22-53.

Rosenberg, D. (2002, December 9). The battle over abstinence. Newsweek, pp. 67-71.

Sack, A. R., Keller, J. F., & Hinkle, D. E. (1984). Premarital sexual intercourse: A test of the effects of peer group, religiosity, and sexual guilt. The Journal of Sex Research, 20, 168-185.

Sexuality Information and Education Council of the United States. (2001). Issues and answers: Fact sheet on sexuality education. SIECUS Report Supplement, 29. Retrieved November 25, 2002. from http://www.siecus. org/pubs/fact/FS_issues_answers.pdf.

Silva, M. (2002). The effectiveness of school-based sex education programs in the promotion of abstinent behavior: A meta-analysis. Health Education Research, 17, 471-481.

Singh, S., Wulf, D., Samara, R., & Cuca, Y. P. (2000). Gender differences in the timing of first intercourse: Data from 14 countries. International Family Planning Perspectives, 26, 21-28.

Social Security Act, 42 U.S.C. [section] 710 (1998).

Sprecher, S., Barbee, A., & Schwartz, P. (1995). "Was it good for you, too?" Gender differences in first sexual intercourse experiences. The Journal of Sex Research, 32, 3-15.

Turner, C. F., Ku, L., Rogers, S. M., Lindberg, L. D., Pleck, J. H., & Sonenstein, F. L. (1998). Adolescent sexual behavior, drug use, and violence: Increased reporting with computer survey technology. Science, 280, 867-873.

Upchurch, D. M., Lillard, L. A., Aneshensel, C. S., & Li, N. F. (2002). Inconsistencies in reporting the occurrence and timing of first intercourse among adolescents. The Journal of Sex Research, 39, 197-206.

US Surgeon General. (2001). The Surgeon General's call to action to promote sexual health and responsible sexual behavior. Rockville, MD: Office of the Surgeon General.

Van Berlo, W., & Ensink, B. (2000). Problems with sexuality after sexual assault. Annual Review of Sex Research, 11, 235-257.

Walser, R. D., & Kern, J. M. (1996). Relationships among childhood sexual abuse, sex guilt, and sexual behavior in adult clinical samples. The Journal of Sex Research, 33, 321-326.

Weis, D. L. (1983). Affective reactions of women to their initial experience of coitus. The Journal of Sex Research, 19, 209-237.

Wiegel, M., Wincze, J. P., & Barlow, D. H. (2002). Sexual dysfunction. In M. M. Antony & D. H. Barlow (Eds.), Handbook of assessment and treatment planning for psychological disorders (pp. 481-522). New York: Guilford.

Wu, L. L., Martin, S. P., & Long, D. A. (2001). Comparing data quality of fertility and first sexual intercourse histories. Journal of Human Resources, 36, 520-555.

Wyatt, G. E., Newcomb, M.D., & Riederle, M. H. (1993). Sexual abuse and consensual sex: Women's developmental patterns and outcomes. Newbury Park, CA: Sage.

Manuscript accepted July 9, 2004

Nicole M. Else-Quest, Janet Shibley Hyde, and John D. DeLamater

University of Wisconsin-Madison

Address correspondence to Nicole Else-Quest, University of Wisconsin-Madison Department of Psychology, 1202 W. Johnson St., Madison, WI, 53706; e-mail: [e-mail].