The Relation Between Early Abuse and Adult Sexuality.

Despite an expansive literature on the psychological correlates of childhood abuse, the possible relationships between specific categories of early abuse (e.g., sexual, physical, psychological) and adult sexuality have received little research attention. Those studies that have examined a relationship between childhood abuse and adult sexuality understandably have focused almost exclusively on sexual abuse. Decreases in sexual functioning such as sexual aversion, anxiety or avoidance, decreased sexual desire or sexual self-esteem, inhibited sexual arousal or orgasm, vaginismus, dyspareunia, and negative attitudes toward sexuality and intimate relationships in general have all been linked to a history of early sexual abuse (for reviews, see Beitchman et al, 1992; Browne & Finklehor, 1986; Gilmartin, 1994). The prevalence and severity of these findings have been questioned as being dependent on the type of sample studied (Okami, 1991; Rind, 1995). Although Laumann, Gagnon, Michael, and Michaels' (1994) random probability U.S. sample found that men with sexual abuse histories reported three out of nine sexually related problems and women reported four out of eight problems, the differences between sexually abused and non-sexually-abused groups were modest. This was further quantified in meta-analyses by Rind and Tromovich (1997), and Rind, Tromovitch, and Bauserman (1998) who calculated only small effect sizes on all the significant sexual problem variables. Other research on both clinical and nonclinical samples has noted that early sexual abuse is correlated with more sexuality in adulthood as indicated by greater variety of sexual activity and more lifetime sexual partners (deYoung, 1982; Herman, 1981; Laumann et al., 1994; Seidner & Calhoun, 1984).

Most studies on the consequences of childhood physical abuse have tended to focus on adult interpersonal relationships rather than sexuality per se (e.g., Briere & Runtz, 1988), or have used childhood physical abuse as a family risk factor or covariate (Mullen, Martin, Anderson, Romans, & Herbison, 1994). The focus has been primarily on adult aggressive and criminal behavior, substance abuse, self-injurious behaviors, emotional problems, and academic and vocational difficulties (see review by Malinowsky-Rummell & Hansen, 1993). In their national telephone survey of 10- to 16-year-old youths, Boney-McCoy and Finklehor (1995) found a significant correlation between severity of psychological symptoms and sexual and parental assault histories. There is also a large, primarily clinical, literature which links childhood physical abuse, substance abuse, and other self-injurious behaviors (Malinowsky-Rummell & Hansen, 1993). The consequences of childhood physical abuse for adult sexuality are potentially important. Indiscriminate sexuality, which might include multiple partners, unsafe sex practices, or drug/alcohol related sexual behavior, could put a person's health and safety at risk and has been shown to be related to the combination of childhood physical and sexual abuse histories (Cunningham, Stiffman, & Dore, 1994). In addition, partner selection, the capacity to make longer-term relationship commitments, and the ability to engage in meaningful sexual relationships, could be affected by a history of childhood physical abuse.

Like physical abuse, the independent impact of both emotional and neglectful abuse on adult sexual behavior has not been explored. Research indicates that these forms of childhood maltreatment are associated with a number of negative child conditions/characteristics, including depression, withdrawal, suicide, low self-esteem, dependency, prostitution, underachievement, emotional maladjustment, and inability to trust others (see review by Hart, Brassard, & Karlson, 1996). Given the nature of these consequences and the fact that acts of emotional maltreatment and neglect are in direct conflict with basic psychological needs such as belongingness, love, and esteem (Hart, Germain & Brassard, 1987), it is reasonable to expect that these forms of abuse may be contributing marker variables for the development of adult sexuality.

Overall, studies on the relationship of early abuse and adult sexuality have not addressed the potential contribution of multiple forms of maltreatment. This is surprising given that sexual, physical, and psychological abuse frequently occur together (Briere, 1992), and different types of maltreatment are often associated with different longterm sequelae (e.g., Briere & Runtz, 1990; Crittenden, 1988). For example, the combination of physical and sexual abuse may account for some of the association between childhood sexual abuse and adult sexuality that has, thus far, only been attributed to sexual abuse experiences, and may also help to explain some of the inconsistencies reported in past findings.

The present study is intended to begin to examine the associations between reported childhood/adolescent maltreatment and adult sexuality. This study extends previous research in this area by (a) examining the correlates of four relatively distinct measures of early abuse (physical, emotional, sexual, neglect) with later adult sexual function, (b) examining the correlates of early abuse with a wide range of sexuality measures, (c) using a comparatively large sample of 1,037 undergraduates, (d) examining potential association differences between genders, (e) examining potential association differences between persons of Southeast Asian and European ancestry, and (f) testing in a confidential laboratory setting. Given that the sample is young and educationally homogeneous, the data provide an opportunity to examine emerging, rather than necessarily definitive, sexual patterns. Understanding how specific forms of abuse influence adult sexuality has important implications for the assessment and treatment of both sexual and nonsexual abuse victims, and for deriving an etiological theory of the mechanisms by which early abuse influences adult sexuality.

METHOD

Participants

Participants were 1,032 University of British Columbia undergraduate volunteers (376 males and 656 females) who completed this study in exchange for introductory psychology course credits. The study was conducted between November, 1993 and December, 1996. Ethnic composition of the sample was approximately 55% non-Asian and 45% Asian. For the purposes of simplicity and brevity, here and throughout this manuscript, East and Southeast Asians are referred to as Asians. All other individuals, including a small portion of South and West Asians, are referred to as non-Asians. Asian status was coded on the basis of participants' responses to the following question: "What is your predominant ethnic background?" Approximately 85% of the Asian subsample were ethnic Chinese. The remaining 15% of Asian respondents endorsed their ethnicity as: Hong Kong, Japanese, Korean, Philippino, Taiwanese, or Vietnamese. Among non-Asians, approximately 84% listed English as their first language, and 83% listed Canada as their country of birth. Among Asians, approximately 21% listed English as their first language, and 29% listed Canada as their place of birth. Participants ranged in age from 17 to 48, with 91% between the ages of 18 and 25. Given that people in their 40s would have been adolescents prior to the onset of the AIDS epidemic and during the height of the "sexual revolution," variables measuring sexual behavior, particularly unrestricted sexual behaviors, may be expected to have a different meaning for participants in their late 30s and 40s than for participants in their late teens and 20s.(1) Because of this, subjects over the age of 30 were excluded from further analyses. Five female subjects fell into this category. Maximum final sample sizes, grouped by gender and ethnicity, were 275 Asian females, 381 non-Asian females, 191 Asian males, and 185 non-Asian males. Mean age of each of the four sub-samples was 19.8 (SD = 2.5) for non-Asian females, 20.6 (SD = 2.8) for non-Asian males, 19.5 (SD = 1.8) for Asian females, and 19.6 (SD = 1.9) for Asian males. Approximately 29% of Asians were born in Canada, 45% were recent immigrants (i.e., 5 years or less in Canada), and 26% were long-term immigrants (i.e., more than 5 years in Canada). Among females, 0.9% reported being exclusively homosexual, 1.2% mostly homosexual, 3.6% bisexual, 19.1% mostly heterosexual, and 75.2% reported being exclusively heterosexual. Among males, 2.1% reported being exclusively homosexual, 1.3% mostly homosexual, 1.6% bisexual, 8.1% mostly heterosexual, and 87% reported being exclusively heterosexual.

Abuse measures

Physical and emotional abuse and neglect were assessed using 28 of the 32 items from the childhood and adolescent sections of the Emotional and Physical Abuse Questionnaire (Carlin et al., 1994). Items on this scale range from relatively common (e.g., "I was shaken") to relatively severe (e.g., "I have had broken bones following a beating") forms of abuse. Subjects were asked to indicate whether or not, and the degree to which, they experienced the abuse items as a child (before age 18) by a parent or guardian. As per Meston, Heiman, Trapnell, and Carlin (1999), items were grouped into the following three categories, physical abuse (13 items) (e.g., "I have received black eyes from being hit," "I was hit or spanked with a board, stick, or wire"), emotional abuse (9 items) (e.g., "I was mocked or taunted," "At home I was criticized and made to feel worthless"), and neglect (6 items) (e.g., "I was locked out or went out of the house without sufficient or appropriate clothes or shoes in the winter," "I have had food or water withheld from me for more than a day"). Participants were asked to indicate on a 5-point Likert scale (0 = never to 5 = very frequently) the frequency with which they experienced each of these items. Abuse composites were formed by summing responses to the items included in each respective composite. Coefficient alpha reliability estimates based on the current sample were .81, .80, .65 (non-Asian males); .82, .87, .65 (non-Asian females); .76, .77, .46 (Asian males); and .80, .83, .44 (Asian females) for physical abuse, emotional abuse, and neglect, respectively.

A sexual abuse score was derived using the following items from the sexual abuse questionnaire (Carlin & Ward, 1992):

1. "I have been casually touched, pinched, or fondled in a sexual manner when I did not want that to happen."

2. "I have been touched or fondled in a sexual manner for a long period of time when I did not want that to happen."

3. "I have been undressed and have had my genitals and/or breasts touched when I did not want that to happen."

4. "I have been forced to touch or fondle another person's genitals when I did not want to."

5. "I have been forced to engage in oral-genital sexual relations when I did not want to."

6. "I have been forced to have sexual intercourse when I did not want to."

Participants responded to these questions on a yes/no basis. A sexual abuse composite score was calculated as follows. A positive endorsement (yes) to question 1 was scored as one, positive endorsements to questions 2 through 4 were each scored as two, and positive endorsements to questions 5 and 6 were each scored as four (Meston et al., 1999). These scores were summed to provide an overall index of sexual abuse. Coefficient alpha reliability estimates based on the current sample were .79 (non-Asian males), .79 (non-Asian females), .61 (Asian males), and .69 (Asian females). Only endorsements of sexual abuse which occurred in either childhood or adolescence before age 18 (not adulthood) were included in the composite scoring.

Sexuality Measures

Interpersonal sexual behavior. Experience with intercourse was scored as a dichotomous variable. Persons were considered to have experienced intercourse if they wrote an age for first intercourse or if they had experienced any of the sexual intercourse items included in the Sexual Experience Subscale of the Derogatis Sexual Functioning Inventory (DSFI; Derogatis, 1978). Frequency of intercourse was assessed by asking respondents to circle a number between 0 (not at all) and 8 (4 or more times a day) corresponding to how often they typically engage in intercourse. The Sexual Experience Subscale was used to evaluate variety of sexual experience. This scale consists of 24 specific sexual acts representing a broad spectrum of sexual experiences, ranging from behaviors such as "Kissing on the lips" to "Mutual oral stimulation of genitals" and various positions of coitus (e.g., "Intercourse-sitting position"). These 24 items are summed to provide an overall index of the range of sexual experience (coefficient alphas = .96, .97 for males and females, respectively). Unrestricted sexual behavior was scored by summing two behavior items from the Sociosexual Orientation Inventory (Simpson and Gangestad, 1991) ("With how many different partners have you had sex in the past year," and "total number of one-night stands") plus four additional items. The four additional items were (a) "With how many different partners have you engaged in any sexual foreplay (e.g., deep kissing, petting, genital caressing) in the past year?," (b) "With how many partners have you had sexual intercourse in your lifetime (include all long-term relationships, brief relationships, and one-time-only relationships and encounters)?," (c) "Have you ever been involved in more than one ongoing sexual relationship at the same time (yes/no)?," and (d) "If you were in a happy and committed relationship, can you imagine a situation in which you might have a brief sexual encounter with someone other than your partner (assume there was absolutely no risk of the affair continuing, or of pregnancy, disease, or discovery by your partner) (yes/no)?" This latter item was adapted from Symons (1987). Prior to the computation of this composite, individual items were first standardized within each gender by ethnicity (i.e., Asian vs. non-Asian) subsample. Coefficient alphas for the unrestricted sexual behavior composite were .77 and .78 for males and females, respectively.

Intrapersonal sexual behavior. Frequency of masturbation was assessed by asking respondents to circle a number between 0 (not at all) and 8 (4 or more times a day) corresponding to how often they typically engage in masturbation. Range of sexual fantasies were assessed using the 20 fantasy items from the DSFI fantasy subscale plus 20 additional items which addressed fantasy areas not assessed by the DSFI (e.g., romance, exhibitionism, voyeurism). The fantasy items were summed to provide an overall index of variety of sexual fantasy (coefficient alphas = .94 for males and females). Follow-up analyses were conducted on six independent fantasy subscales: masochism (e.g., "being sexually degraded"), promiscuity (e.g., "having more than one sexual partner at the same time"), voyeurism (e.g., "secretly watch others engage in sex"), sadism (e.g., "whipping or beating your sexual partner"), exhibitionism (e.g., "engaging in sex while some person or persons are watching you"), and romance (e.g., "making love outdoors in a romantic setting"). The DSFI Attitudes Subscale was used to assess liberal sexual attitudes. This subscale is composed of 15 liberal and 15 conservative statements relative to sexual behaviors. Respondents are asked to answer the 30 questions on a 5-point Likert scale from strongly disagree (1) to strongly agree (5). Endorsements of conservative items are reverse scored and summed together with the liberalism items to provide an overall index of liberal attitudes towards sexuality (coefficient alphas = .82, .88 for males and females, respectively). A composite variable representing unrestricted sexual attitudes and fantasies was constructed by summing z-scores for the following two composites: (a) a four-item cluster of Sociosexual Orientation Inventory attitudes regarding casual sexual involvements (e.g., "I can imagine myself being comfortable and enjoying "casual" sex with different partners": 1 = strongly disagree to 5 = strongly agree), and (b) a three-item composite of DSFI fantasies about unrestricted sex (e.g., "Having intercourse with an anonymous, attractive stranger": 0 = never to 4 = quite often). A composite variable representing subjective sexual drive was constructed by summing z-scores to two variables, (a) "Overall, how would you rate your level of sex drive?" (1 = below average to 5 = above average), and (b) a three-item composite of self-rated sexual arousability ("I am a really sensual and sexual kind of person, someone who feels very aroused, very often," "Once in a while, I feel so sexually aroused, I can't think of anything but sex," "I've never really felt very interested in sex" [latter item reverse scored]: 1 = strongly disagree to 5 = strongly agree).

Sexual satisfaction. Sexual satisfaction was evaluated using the Sexual Satisfaction Scale of the Sexual Experiences Inventory (Trapnell & Meston, 1995). This scale consists of 23 items which evaluate global sexual satisfaction and a range of specific sources of sexual dissatisfaction, including sexual contentment (e.g., "I often feel something is missing from my present sex life"), sexual competence (e.g., "I'm concerned that my occasional problems becoming aroused could occur again or become worse"), sexual communication (e.g., "I usually feel comfortable discussing sex when my partner wants to"), and sexual compatibility (e.g., "I often feel that my partner's beliefs and attitudes about sex are too different from mine"). Respondents were asked to answer the 23 questions, with reference to their most recent close relationship, on a 5-point Likert scale from strongly disagree (1) to strongly agree (5). Coefficient alphas for the sexual satisfaction composite were .91, .92 for males and females, respectively. Body image was assessed using the Body Image subscale of the DSFI. This scale consists of self-ratings on 5 gender specific physical attributes (e.g., "Women/men would find my body attractive") and 10 general body attributes (e.g., "My face is attractive"), for a total of 15 ratings. Respondents are asked to answer the 15 questions on a 5-point Likert scale from not at all (1) to extremely (5). These items are summed in a positive direction to provide a single numerical index of satisfaction with one's physical appearance. Coefficient alphas based on the current sample were .83, .82 for males and females, respectively.

Sexual orientation. Sexual identification was assessed using a 5-point Likert scale similar to the Kinsey Heterosexual-Homosexual Rating Scale (Kinsey, Pomeroy, & Martin, 1948). Participants indicated their sexual orientation on the scale from 1 (exclusively homosexual) to 5 (exclusively heterosexual).

Self-Esteem Measures

Self-esteem was measured using the Rosenberg Self-Esteem Inventory (Rosenberg, 1965). This scale consists of 10 items which assess positive (e.g., "Oil the whole, I am satisfied with myself") and negative (e.g., "I certainly feel useless at times") feelings about oneself. Participants indicate the degree to which they believe each of the statements accurately describes themselves on a scale from 1 (strongly agree) to 5 (strongly disagree). Negative items are reverse scored and summed with the positive items to provide an overall measure of self-esteem.

Data Analyses

Meston, Trapnell, and Gorzalka (1996, 1998) recently reported significant differences in a wide range of sexuality measures between undergraduate students of Southeast Asian and European ancestry. In addition, Meston et al. (1999) reported significant differences in the incidence and forms of abuse experienced by students of Southeast Asian and European ancestry. As a result of these findings, a series of moderated multiple regressions were performed separately on the females' and males' data to test for ethnic differences in associations between the abuse and sexuality measures. In each analysis, measures of either physical abuse, emotional abuse, sexual abuse, or neglect were first regressed on the dichotomous ethnicity variable and the sexuality variable of interest (e.g., "variety of sexual experience"), and then the cross-product of ethnicity and the sexuality variable was statistically evaluated for any residual association with the abuse variable. None of the 96 cross-products (12 sexuality variables x 4 abuse variables x gender) fell below a nominal pairwise alpha of .05 (for all Fs, p [is greater than] .05) indicating there were no significant ethnic differences in associations. Subsequent analyses were therefore performed on data collapsed across ethnicity.

Past research has found age to be positively correlated with sexual experience in college samples (Meston et al., 1996). Because age might also be correlated with abuse history, we conducted a series of Pearson correlations between age and sexual experience, and between age and each of the abuse measures, separately for males and females. Age was significantly correlated with sexual experience in both males, r(348) = .30, p [is less than] .001, and females, r(645) = .23, p [is less than] .001, and with emotional abuse in females, r(614) = .08, p [is less than] .05. Age was not significantly related to any of the other abuse variables in females or to any of the abuse measures in males. Because age was significantly related to both sexual experience and emotional abuse in females, all further analyses of sexuality variables with emotional abuse in females were conducted with age controlled for.

To evaluate the relation between abuse and sexuality, partial correlations, controlling for ethnicity, were conducted separately, by gender, between sexuality measures and each of the physical abuse, emotional abuse, neglect, and sexual abuse composite scores. Consistent with past research (e.g., Ney, Fung, & Wickett, 1994), there was a high intercorrelation among abuse measures noted in the present study (see Table 1). We therefore conducted follow-up analyses on all significant correlations using hierarchical regression analyses to evaluate whether the relation between the abuse and the sexuality variable of interest remained significant after controlling for the effects of ethnicity and the remaining three abuse variables. To be cautious about accumulating Type I error on comparisons across the 12 sexuality variables, for all analyses only differences of p [is less than] .004 (p [is less than] .05/12) were considered statistically reliable.

Table 1. Correlations Between Self-Reported Measures of Early Abuse

                   Physical   Emotional             Sexual
                    abuse       abuse     Neglect   abuse

Males
 Physical abuse                .67(*)     .52(*)    .06
 Emotional abuse                          .54(*)    .04
 Neglect                                            .11

Females
 Physical abuse                .67(*)     .49(*)    .23(*)
 Emotional abuse                          .49(*)    .20(*)
 Neglect                                            .16(*)

Note: Maximum n = 359 males; 618 females.

(*) p < .001.

PROCEDURE

Participants completed the questionnaires in same-sex groups of approximately 5-10 individuals in a large testing room arranged to provide maximum privacy for responding (e.g., visual barriers between participants). All participants completed the sexuality measures (DSFI, Sexual Experiences Inventory, Sociosexual Orientation Inventory) first, followed by the Emotional and Physical Abuse Questionnaire and the Sexual Abuse Questionnaire. A same-sex researcher was present during all sessions to provide instructions and to answer any questions. Testing conditions were made confidential by requesting that no names or other personally identifying information be placed on any of the forms, and by using a randomly selected number for participant identification across the two sessions. In addition, confidentiality was stressed verbally, and in written instructions by requesting that each participant fold their completed answer sheets, seal them in a blank envelope, and deposit the envelope in a large drop box upon leaving the testing room. Before beginning the questionnaires, all participants were asked to read a brief statement informing them of the sexual, personal nature of the questionnaires; the voluntary and confidential nature of all responses; and their right to withdraw from the session at any time, for any reason, without any loss of credit. No participants withdrew from the study.

Approximately one to two weeks prior to the study, a subset of participants (160 males; 359 females) completed the Rosenberg Self-Esteem Inventory as part of a separate, unrelated study. The results from the Rosenberg Self-Esteem Inventory were used to conduct follow-up analyses to the present results. This subsample did not differ significantly (p [is greater than] .05) from the overall sample in ethnic composition or sexual experience.

RESULTS

Physical Abuse

At least one physical abuse item was endorsed by 51% of non-Asian females, 69% of Asian females, 71% of non-Asian males, and 82% of Asian males. Severe physical abuse (defined here as the endorsement of one or more of the severe physical items, e.g., "I have received black eyes from being hit," "I was injured seriously enough by a parent or guardian to require medial care," "I have had broken bones following a beating" was reported by 10%, 19%, 14%, and 34% of non-Asian females, Asian females, non-Asian males, and Asian males, respectively.

Among females, there was a significant positive correlation between the frequency of childhood physical abuse and unrestricted sexual behavior and variety of sexual fantasy (see Table 2). These relations did not remain significant after controlling for the effects of ethnicity, childhood sexual abuse, emotional abuse, and neglect. Frequency of childhood physical abuse in males was significantly positively correlated with range of sexual fantasies and reports of unrestricted sexual attitudes and fantasies. These correlations did not remain significant after the effects of ethnicity, childhood sexual abuse, emotional abuse, and neglect were controlled for. There were no significant correlations at p [is less than] .004 between physical abuse and measures of sexual adjustment or orientation in either males or females.

Table 2. Partial Correlations of Sexuality Reports with Self-Reported Early Abuse Controlling for Ethnicity

                                          Females

                                    Physical   Emotional
                                     abuse       abuse

Interpersonal sexual behavior

Experience with intercourse (1)      .02        -.04
Frequency of intercourse (1)        -.02        -.04
Variety of sexual experience (24)    .05        -.06
Unrestricted sexual behavior (6)     .18(*)      .08

Intrapersonal sexual behavior

Frequency of masturbation (1)        .07         .06
Variety of sexual fantasy (40)       .14(*)      .12(*)
Liberal sexual attitudes (30)        .03         .03
Unrestricted sexual attitudes
 & fantasies (2)                     .10         .09
Subjective sexual drive (2)         -.11        -.07

Sexual adjustment

Total sexual satisfaction (25)      -.04        -.10
Positive body image (15)            -.06        -.14(*)

Sexual orientation

Sexual identification (1)           -.04        -.04

                                        Females

                                              Sexual
                                    Neglect   abuse

Interpersonal sexual behavior

Experience with intercourse (1)     -.05       .21(*)
Frequency of intercourse (1)        -.04       .17(*)
Variety of sexual experience (24)   -.01       .26(*)
Unrestricted sexual behavior (6)     .07       .25(*)

Intrapersonal sexual behavior

Frequency of masturbation (1)        .06       .14(*)
Variety of sexual fantasy (40)       .10       .23(*)
Liberal sexual attitudes (30)        .00       .16(*)
Unrestricted sexual attitudes
 & fantasies (2)                     .05       .19(*)
Subjective sexual drive (2)         -.03      -.21(*)

Sexual adjustment

Total sexual satisfaction (25)      -.13(*)    .04
Positive body image (15)            -.10       .05

Sexual orientation

Sexual identification (1)           -.07      -.10

                                           Males

                                    Physical   Emotional
                                     abuse       abuse

Interpersonal sexual behavior

Experience with intercourse (1)      .04        .01
Frequency of intercourse (1)         .03        .01
Variety of sexual experience (24)    .04        .01
Unrestricted sexual behavior (6)     .05        .01

Intrapersonal sexual behavior

Frequency of masturbation (1)        .07        .12
Variety of sexual fantasy (40)       .21(*)     .22(*)
Liberal sexual attitudes (30)        .07        .05
Unrestricted sexual attitudes
 & fantasies (2)                     .18(*)     .18(*)
Subjective sexual drive (2)         -.09       -.10

Sexual adjustment

Total sexual satisfaction (25)      -.16       -.29(*)
Positive body image (15)            -.01       -.15(*)

Sexual orientation

Sexual identification (1)            .02       -.10

                                         Males

                                              Sexual
                                    Neglect   abuse

Interpersonal sexual behavior

Experience with intercourse (1)      .08      -.00
Frequency of intercourse (1)         .10      -.07
Variety of sexual experience (24)    .12       .02
Unrestricted sexual behavior (6)     .13       .01

Intrapersonal sexual behavior

Frequency of masturbation (1)        .02       .03
Variety of sexual fantasy (40)       .17(*)    .15
Liberal sexual attitudes (30)        .04       .05
Unrestricted sexual attitudes
 & fantasies (2)                     .15       .09
Subjective sexual drive (2)         -.09      -.08

Sexual adjustment

Total sexual satisfaction (25)      -.17(*)   -.10
Positive body image (15)            -.02      -.01

Sexual orientation

Sexual identification (1)           -.05      -.07

Note: Maximum n = 354 males; 617 females. Number of items per scale is indicated in parentheses. Semipartial correlations and beta weights are available from the first author upon request.

(a) Correlations between emotional abuse and sexuality variables in females have been conducted with ethnicity and age controlled for.

(*) p < .004.

Emotional Abuse

At least one emotional abuse item was endorsed by 70% of non-Asian females, 88% of Asian females, 81% of non-Asian males, and 93% of Asian males. Severe emotional abuse (defined here as an emotional abuse composite score of greater than 10) was reported by 25%, 39%, 25%, and 53% of non-Asian females, Asian females, non-Asian males, and Asian males, respectively.

Among females, after the effects of age and ethnicity were controlled for, frequency of emotional abuse was significantly positively related to range of sexual fantasies and significantly negatively related to (positive) body image (see Table 2). These correlations did not remain significant after the effects of ethnicity, age, sexual abuse, physical abuse, and neglect were controlled for. Among males, frequency of emotional abuse was significantly positively related to variety of sexual fantasy and unrestricted sexual attitudes and fantasies, and significantly negatively related to sexual satisfaction and (positive) body image. After controlling for the effects of ethnicity, childhood sexual abuse, physical abuse, and neglect, negative relations between emotional abuse and sexual satisfaction, (F(1,275) = 16.00, p [is less than] .001; sr = -.16), and between emotional abuse and (positive) body image, (F(1,274) = 8.41, p [is less than] .004; sr = -.12), remained significant, whereas relations between emotional abuse and the former two variables did not.

One notable pattern of association which was similar between genders was the relation between physical abuse and emotional abuse and variety of sexual fantasies. Because of the large number (i.e., 40) of fantasy items included in this composite measure, follow-up analyses were conducted to examine whether the specific types of fantasies endorsed might differ between genders. Six fantasy subgroups were examined: exhibitionism, voyeurism, promiscuity, romance, sadism, and masochism. Among males, after controlling for the effects of ethnicity, physical abuse and emotional abuse were significantly (p [is less than] .008 [p .05/6]) positively related to promiscuous, sadistic, and voyeuristic fantasies, and emotional abuse was also significantly related to romantic and exhibitionistic fantasies. Among females, after controlling for the effects of ethnicity, childhood physical abuse was significantly positively related to promiscuous sexual fantasies, and after controlling for the effects of ethnicity and age, emotional abuse was significantly positively related to both promiscuous and masochistic fantasies.

Neglect

At least one neglect item was endorsed by 32% of non-Asian females, 46% of Asian females, 46% of non-Asian males, and 64% of Asian males. Severe neglect (defined here as the endorsement of one or more of the severe neglect items, e.g. "I was locked out of the house without sufficient or appropriate clothes or shoes in the winter," "I have had food or water withheld from me for more than a day,") was reported by 9%, 18%, 18%, and 32% of non-Asian females, Asian females, non-Asian males, and Asian males, respectively.

Frequency of neglect was significantly negatively related to sexual satisfaction in both males and females, and significantly positively related to variety of sexual fantasy among males (see Table 2). After controlling for the effects of ethnicity, physical abuse, emotional abuse, and sexual abuse, these relations did not remain significant.

Sexual Abuse

One or more of the six sexual abuse items were endorsed by 40%, 25%, 11%, and 11% of non-Asian females, Asian females, non-Asian males, and Asian males, respectively. Being forced to have sexual intercourse was endorsed by 9%, 4%, 1%, and 1% of non-Asian females, Asian females, non-Asian males, and Asian males, respectively.

Sexual abuse was significantly related (p [is less than] .004) to all measures of inter- and intrapersonal sexuality in women. Frequency of sexual abuse was negatively related to sexual drive, and positively related to experience with intercourse, variety of sexual experience, range of sexual fantasies, liberal sexual attitudes, frequency of intercourse and masturbation, and likelihood of engaging in unrestricted sexual behaviors and fantasies. There were no significant correlations (p [is less than] .004) between sexual abuse and measures of sexual adjustment or sexual orientation among females, and no significant correlation between sexual abuse and any of the sexuality variables among males (see Table 2).

After controlling for the effects of ethnicity, physical abuse, emotional abuse, and neglect, correlations remained statistically significant between sexual abuse and drive, (F (1, 523) = 20.45, p [is less than] .001; sr = -.19); unrestricted sexual attitudes and fantasies, (F (1, 523) = 14.51, p [is less than] .001; sr = .16); liberal sexual attitudes, (F (1, 523) = 11.82, p [is less than] .001; sr = .14); variety of sexual fantasy, (F (1, 523) = 20.64, p [is less than] .001; sr =. 19); frequency of masturbation, (F (1, 513) = 7.93, p = .005; sr = .12); unrestricted sexual behavior, (F (1, 516) = 23.13, p [is less than] .001; sr = .20); variety of sexual experience, (F (1, 523) = 36.43, p [is less than] .001; sr = .24); frequency of intercourse, (F (1, 515) = 18.33, p [is less than] .001; sr = .18); and experience with intercourse, (F (1, 523) = 25.05, p [is less than] .001; sr = -.20); in women.

Does Self-Esteem Account for the Relation Between Abuse and Sexual Adjustment?

Note that the two sexual adjustment variables, sexual satisfaction and body image, are the only two variables that explicitly involve evaluative appraisals. It is reasonable, therefore, to speculate that the associations between abuse and these two variables may reflect more general self appraisal (i.e., self-esteem) rather than sexual appraisal per se. In the present sample, Rosenberg Self Esteem scores were significantly negatively related to frequency of physical abuse, (r (345) = -.16, p = .005), emotional abuse, (r (345) = -.32, p [is less than] .001), and neglect, (r (345) = -.20, p [is less than] .001), among females. The multiple correlation between self-esteem and the four abuse predictors, controlling for ethnicity, was significant in females, ([R.sup.2] = .14, F (5,355) = 11.4, p [is less than] .001); however, the only significant beta weight was for emotional abuse, (beta = -.36, t = -5.4, p [is less than] .001). Among males in the present sample, self-esteem scores were significantly negatively related to frequency of emotional abuse (r (155) = -.17, p [is less than] .05). Correlations between self-esteem scores and measures of physical abuse and neglect did not reach statistical significance (p [is greater than] .05). The multiple correlation of the four abuse variables, controlling for ethnicity, was not significant for males, ([R.sup.2] = .05, F (5,153) = 1.4, p [is greater than] .05). Note that although a significant association between sexual abuse and adult self-esteem has previously been reported (for review, see Browne & Finkelhor, 1986), in the present study associations between self-esteem and sexual abuse did not reach statistical significance in either males or females.

Follow-up analyses were conducted to examine whether relations between emotional abuse and body image, and between neglect and sexual satisfaction might be explained in terms of self-esteem in females. Self-esteem scores were significantly positively related to both (positive) body image, r(345) = .51, p [is less than] .001, and sexual satisfaction, r(345) = .35, p [is less than] .001, after the effects of ethnicity were controlled for. Neither the correlation between emotional abuse and body image, r(344) = -.05, p [is greater than] .05, or the correlation between neglect and sexual satisfaction, r(344) = -.09, p [is greater than] .05, remained significant after controlling for the effects of ethnicity and self-esteem. The results of these follow-up tests suggest that most of the association between emotional abuse and body image and between neglect and satisfaction in females can be accounted for by the association between these variables and self-esteem.

Follow-up analyses were conducted to examine whether relations between emotional abuse and body image and sexual satisfaction in males might also be explained in terms of self-esteem. Both (positive) body image, r(144) = .44, p [is less than] .001, and sexual satisfaction, r(144)= .20, p [is less than] .05, were significantly positively related to self-esteem scores after the effects of ethnicity were controlled for. After controlling for the effects of ethnicity and self-esteem the correlation between emotional abuse and body image, r(139) = -.05, p [is greater than] .05, did not remain significant. This suggests that most of the association between emotional abuse and body image in males can be accounted for by the association between these variables and self-esteem. The correlation between emotional abuse and sexual satisfaction, however, remained significant, r(139) = -.36, p [is less than] .001, after the effects of ethnicity and self-esteem were controlled for.

DISCUSSION

The present study examined relations between four relatively distinct forms of early abuse (physical, emotional, sexual, neglect) and measures of adult interpersonal sexual behavior, intrapersonal sexual behavior, sexual adjustment, and sexual orientation. Sexual abuse, independent of the effects of childhood physical abuse, emotional abuse, and neglect, was significantly related to all forms of interpersonal and intrapersonal sexuality variables in females. There were no significant relations between childhood physical abuse, emotional abuse, or neglect and any of the sexuality variables in females, independent of the effects of other forms of abuse. In males, frequency of emotional abuse was significantly negatively related to sexual satisfaction and body image, independent of the effects of physical abuse, neglect, and sexual abuse. Independent of other forms of abuse, there were no significant relations between childhood physical abuse, neglect, or sexual abuse and any of the sexuality variables in males.

A relation between childhood sexual abuse and adult sexual behavior in females has been previously reported. In the present study, sexual abuse was associated with more liberal sexual attitudes and behavior, as evidenced by a higher frequency of intercourse and masturbation, a greater range of sexual experience: and fantasies, and a greater likelihood of engaging in unrestricted sexual behavior. These findings are consistent with past research (e.g., Courtois, 1979; deYoung, 1982; Herman, 1981; Meiselman, 1978; Laumann et al., 1994; Potterat, Rothenberg, Muth, Darrow, & Phillips-Plummer, 1998.; Seidner & Calhoun, 1984; Tsai, Feldman-Summers, & Edgar, 1979; Wisniewski, 1989) which has also noted that a sexualization of relationships and increased sexual activity were associated with a history of childhood sexual abuse. These findings may, in part, be explained within the framework of Finkelhor and Browne's (1985) theory of sexual traumatization. Briefly, this theory posits that, through a variety of means, childhood sexual abuse shapes sexuality in a way that may foster confusion and misconceptions about sexual self-concepts, unusual emotional associations to sexual activities, and an inappropriate repertoire of sexualized behaviors (Finkelhor & Browne, 1985). For example, the child may be rewarded for sexual behavior that is developmentally inappropriate and learn to use sexual behavior as a strategy for manipulating others (see Footnote 1). While this theory may help explain why childhood sexual abuse in the present study was related to increased sexual activity, it cannot explain why childhood sexual abuse was unrelated to measures of sexual adjustment.

A number of researchers have reported that the relation between childhood sexual abuse and later adult adjustment (broadly defined) appears to be accounted for, to a large extent, by family background rather than sexual abuse per se. For example, using path analysis to examine child sexual abuse-adjustment relations among female students across 32 U.S. colleges, Wisniewski (1989) concluded that factors such as family violence had a greater impact on current emotional adjustment than did the specific effects of sexual abuse. In a recent review of seven studies on childhood sexual-abuse-symptom relations which used national probability samples, Rind and Tromovitch (1997) concluded that "the results from psychological adjustment measures imply that, childhood sexual abuse is related to poorer adjustment in the general population, the magnitude of this relation is small ... [and] cannot safely be assumed to reflect causal effects of the childhood sexual abuse" (p. 253). The authors suggested that differences in adjustment between sexually abused persons and controls observed in national samples may be attributable to larger differences in social environment, rather than to the sexual abuse. In a later meta-analyses of 59 studies based on college samples, Rind et al. (1998) similarly concluded that students with childhood sexual abuse histories were slightly less well adjusted than controls, but that this was more likely attributable to family environment than abuse experiences per se. These assertions are consistent with other studies that have noted when other abuse (i.e., physical, emotional) experiences are held constant, childhood sexual-abuse-symptom relations frequently disappear (e.g., Cole, 1987; Eckenrode, Laird, & Doris, 1993; Higgins & McCabe, 1994; Ney et al., 1994). In the present study, when other forms of abuse were statistically controlled for, relations between sexual abuse and measures of inter- and intrapersonal sexuality remained significant. Although we did not directly measure family background in the present study, and measures of physical abuse, emotional abuse, and neglect may serve as only proxy measures, if it were the case that relations between childhood sexual abuse and increased sexual activity were accounted for primarily by social environment one would expect such relations to decline substantially when other, nonsexual forms of abuse were controlled for. The fact that these relations remained significant suggests that, unlike relations between sexual abuse and adjustment measures, relations between sexual abuse and liberal sexual behavior may be specifically related to consequences of the sexual act per se. Perhaps in some way the early sexual abuse experience awakened sexual interests which, over time, led to the development of more sexually liberal values. This is not to say, of course, that social environment would not play an important role in this sexual development. A disruptive family background may foster conflicting messages about sex, may be associated with a lack of supervision which may open up more opportunities for sexual experiences, and may create anxiety over abandonment which may encourage sexualized ties (for review of childhood sexual abuse risk factors see Higgins & McCabe, 1994).

The striking gender difference in the relation between sexual abuse and measures of both inter- and intrapersonal sexual behavior warrants comment. In females, childhood sexual abuse was related to all measures of inter- and intrapersonal sexual behaviors; in males it was related to none. First, it should be noted that there was a substantially greater (i.e., 25-40% females; 11% males) incidence of childhood sexual abuse among female participants than among male participants in the present study, which may statistically account for some of this gender difference in equivalence. In both a meta-analytic review of childhood sexual-abuse-symptom national probability studies (Rind and Tromovitch, 1997), and a meta-analyses of 59 college sample studies (Rind, Tromovitch, & Bauserman, 1998), it was reported that the relation between sexual abuse and adjustment problems was considerably stronger for women than for men. Whereas a majority of females (approximately two thirds) reported negative effects of sexual abuse, only a minority (approximately one third) of males did so. The authors suggested that these differences may be accounted for by the fact that males and females typically do not have the same type of childhood sexual abuse experiences. Females are more likely than males to report being coerced during the sexual abuse experience, females are generally younger than males when they experienced the sexual abuse, and female childhood sexual abuse experiences are more likely to involve an incestuous act than are male childhood sexual abuse experiences. Because we did not assess coercion or perpetrator relations in the present study, we unfortunately are unable to assess whether such factors may account for this gender difference.

A sex-role perspective might explain the gender difference in relations between childhood sexual abuse and adult sexuality in terms of a persisting double standard in the gender appropriateness of sexual constraint (i.e., females should be more restrained than males). The self-evaluative consequences of childhood sexual abuse could be expected to differ between women and men because of this norm. For example, abused women may come to see themselves as sexually atypical (i.e., unconstrained, seductive, provocative, "damaged") with respect to the norm of constraint, and that self-image could disinhibit later sexual constraint via the phenomenon of self-verification (Swan, Wenzlaff, & Tafarodi, 1992). Abused men are unlikely to develop such a self-perception if being sexually indiscriminate is what successful men (i.e., agentic, powerful, competent) do.

In males, after the effects of other forms of abuse were controlled for, frequency of emotional abuse remained significantly negatively related to (positive) body image and sexual satisfaction. As was the case with relations between nonsexual forms of abuse and sexual adjustment measures in females, the relation between emotional abuse and body image in males was primarily accounted for by the effects of self-esteem. This suggests that such relations may be explained largely in terms of global, negative self-appraisal. Global self-esteem did not, however, account for the relation between emotional abuse and sexual satisfaction among males. The relation was also not due to a narrow association involving specific content within either the sexual satisfaction or emotional abuse scales. We examined individual correlations between emotional abuse items and sexual satisfaction. There was no discernible pattern between the content of emotional abuse items and the strength of their relationship with sexual satisfaction. The largest correlation was for the emotional abuse item "At home I was criticized and made to feel worthless," r(146) = .32, p [is less than] .001. The satisfaction composite consists of four internally consistent subscales derived via factor analysis (Trapnell & Meston, 1995), namely competence, communication, compatibility, and contentment. The strength of the association with emotional abuse did not vary meaningfully between scales. Partial correlations, controlling for ethnicity, ranged from r(337) = .14, p = .01, for competency, to r(337) = .24, p [is less than] .001, for contentment.

If the relation between emotional abuse and sexual satisfaction in males is a general one that involves multiple facets of sexual satisfaction and emotional abuse, but is not due to general negative self-appraisal (i.e., low self-esteem), what accounts for this association and why is it specific to males? One potential explanation is differences in sex roles governing courtship initiation. If males need to be the initiator to get involved in a sexual relation, it might require a certain amount of dating efficacy. If one assumes that the experience of repeated emotional abuse (i.e., belittlement, criticism) impairs the development of social efficacy, such as dating efficacy, then gender differences in courtship initiation could potentially explain the association between emotional abuse and sexual satisfaction in males and not females. This speculation would seem to be ruled out however, by the fact that emotional abuse was not related to sexual experience levels in males. This interesting gender difference, if replicable, may warrant future research attention.

Associations between early abuse and adult sexual behavior did not differ significantly between persons of Southeast Asian and European ancestry. Among a comparable sample of 702 undergraduate students, Meston et al. (1996; 1998) found that Asian students were significantly more sexually conservative or restrained than their non-Asian counterparts on virtually every measure of inter- and intrapersonal sexual behavior, and held significantly more conservative sexual attitudes. Given such strong cultural differences, one might have expected, for example, cultural pressures to have inhibited sexually abused females from acting in a sexually unrestrained manner. The fact that none of 96 associations between abuse and sexuality measures differed significantly between Asian and non-Asian persons emphasizes the strength of these relationships and their potential universality across certain ethnic groups.

In summary, the results of this study suggest an independent relation between childhood sexual abuse and sexual behavior in females, and between emotional abuse and measures of body image and sexual adjustment in males. These findings highlight the importance of examining relationships between early abuse and adult sexuality separately, within gender, and within specific subcategories of abuse. Several factors limit interpretation of the present results and their generalizability to other populations. First, with regard to the measures used in this study, the coefficient alphas for the neglect composite were relatively low (i.e., .44 to .65) across ethnic groups. This suggests that the particular items used in the present study may not form a reliable composite indicator of neglect. The items referred primarily to insufficient food or clothing, lack of cleanliness, and being left home alone; lack of supervision, a commonly used indicator of neglect, was not well represented.

Second, the sample used in the present study was comprised exclusively of an undergraduate student population. This worked to an advantage in that (a) students have better recall than older populations (Finkelhor, 1979); (b) emerging sexual patterns can be examined within a relatively homogeneous group with respect to age, intelligence, and socioeconomic status; and (c) student populations may be more representative of the community than are clinical or criminal populations (Higgins & McCabe, 1994). However, student samples exclude most persons who are severely troubled, below average intelligence, or from deviant subcultures (Finkelhor, 1979). Hence, the present sample may have excluded persons who have been most negatively affected by early abuse experiences.

A third limitation of this study is that the findings are based strictly on retrospective, self-report data. Several measures were taken to maximize honest self-disclosure in the present study, such as the administration of questionnaires in a confidential laboratory setting, the use of same-sex testing scenarios, and frequent reminders within the questionnaire for honesty and accuracy. Using the same student sample, Meston et al. (1999) found no significant relation between measures of socially desirable responding and reports of early physical, emotional, or sexual abuse or neglect, and Meston, Heiman, Trapnell, and Paulhus (1998) found only minimal evidence for the role of social desirability in self-reported sexuality variables under anonymous testing conditions. Despite these facts, however, the possibility that participants may have either over or under reported abuse experiences cannot be ruled out. Several researchers (e.g., Finkelhor, 1984; Watkins & Bentovim, 1992) have noted a tendency for males to under report sexual abuse experiences. To the extent that this may have occurred in the present study, this gender-specific reporting bias may have masked potential relations between childhood abuse and adult sexuality in men.

The correlational nature of the present study further limits the interpretation of results. While it is tempting to speculate that the early abuse experiences impacted later adult sexual behavior, such causal inferences must be made with caution. Given that the age criterion for abuse was 18 years, it may be the case that a number of the sexual behaviors measured in the present study occurred prior to the reported abuse. Hence, being sexually active at an early age may have, in some way, led subjects to become more vulnerable to abuse, or alternatively, the relation between abuse and sexuality variables found in the present study may both be attributable to a third, unrelated factor. Clearly, prospective studies with long-term follow-up are needed to circumvent such limitations.

The final consideration regarding interpretation of the present findings is that participants with multiple abuse histories (e.g., sexual, physical, and emotional) were not distinguished from those with a history of one specific form of abuse. It may be the case that certain combinations of abuse experiences impact sexuality in ways that differ from those seen in the present study. Clearly, this study represents a preliminary examination of the influence of early physical, sexual, and emotional abuse and neglect on adult sexuality. Futures studies are needed to examine the effects of compound abuse experiences on sexuality, and to explore potential moderators that might affect the direction and/or strength of these relations.

(1) We are grateful to an anonymous reviewer for this comment.

REFERENCES

Beitchman, J. H., Zucker, K. J., Hood, J. E., DaCosta, G. A., Akman, K., & Cassavia, E. (1992). A review of the long-term effects of child sexual abuse. Child Abuse and Neglect, 16, 101-118.

Boney-McCoy, S., & Finkelhor, D. (1995). Psychosocial sequelae of violent victimization in a national youth sample. Journal of Consulting and Clinical Psychology, 63, 726-736.

Briere, J. (1992). Child abuse trauma: Theory and treatment of the lasting effects. Newbury Park, CA: Sage.

Briere, J., & Runtz, M. (1988). Symptomatology associated with childhood sexual victimization in a non-clinical adult sample. Child Abuse and Neglect, 12, 51-59.

Briere, J., & Runtz, M. (1990). Differential adult symptomatology associated with three types of child abuse histories. Child Abuse and Neglect, 14, 357-364.

Browne, A., & Finkelhor, D. (1986). The impact of child sexual abuse: A review of the research. Psychological Bulletin, 99, 66-77.

Carlin, A. S., Kemper, K., Ward, N. G., Sowell, H., Gustafson, B., & Stevens, N. (1994). The effects of differences in objective and subjective definitions of childhood physical abuse on estimates of its incidence and relationship to psychopathology. Child Abuse and Neglect, 18, 393-399.

Carlin, A. S., & Ward, N. G. (1992). Subtypes of psychiatric inpatient women who have been sexually abused. The Journal of Nervous and Mental Disease, 180, 392-397.

Cole, C. B. (1987). The specificity of long-term effects of sexual abuse and factors mediating outcome: A comparison of sexually and physically abused young adults. Unpublished doctoral dissertation, University of Washington.

Courtois, C. (1979) The incest experience and its aftermath. Victimology: An International Journal, 4, 337-437.

Crittenden, P. M. (1988). Relationships at risk. In J. Belsky & R. Nezworski (Eds.), Clinical implications of attachment (pp. 1336-174). Hillsdale, NJ: Lawrence Erlbaum.

Cunningham, R. M., Stiffman, A. R., & Dore, P. (1994). The association of physical and sexual abuse with HIV risk behaviors in adolescence and young adulthood: Implication for public health. Child Abuse and Neglect, 18, 233-245.

Derogatis, L. R. (1978). Derogatis Sexual Functioning Inventory (Rev. ed.) Baltimore MD: Clinical Psychometrics Research.

De Young, M. (1982). The sexual victimization of children. Jefferson, NC: McFarland.

Eckenrode, J., Laird, M., & Doris, J. (1993). School performance and disciplinary problems among abused and neglected children. Developmental Psychology, 29, 53-62.

Finkelhor, D. (1979). Sexually victimized children. New York: Free Press.

Finkelhor, D. (1984). Child sexual abuse: New theory and research. New York: Free Press.

Finkelhor, D., & Browne, A. (1985). The traumatic impact of child sexual abuse: A conceptualization. American Journal of Orthopsychiatry, 55, 530-541.

Gilmartin, P. (1994). Rape, incest, and child sexual abuse. New York: Garland Publishing.

Hart, S. N., Brassard, M. R., & Karlson, H. C. (1996). Psychological maltreatment. In J. Briere, L. Berliner, J. A. Bulkley, C. Jenny, & T. Reid (Eds.), The APSAC handbook on child maltreatment (pp. 72-89). London: Sage Publications.

Hart, S. N., Germain, R. B., & Brassard, M. R. (1987). The challenge: To better understand and combat psychological maltreatment of children and youth. In M. R. Brassard, R. Germain, & S. N. Hart (Eds.), Psychological maltreatment of children and youth (pp. 3-24). New York: Pergamon Books.

Herman, J. L. (1981). Father-daughter incest. Cambridge. MA: Harvard University Press.

Higgins, D. J., & McCabe, M. P. (1994). The relationship of child sexual abuse and family violence to adult adjustment: Toward an integrated risksequelae model. The Journal of Sex Research, 31, 255-266.

Kinsey, A. C., Pomeroy, W. B., & Martin, C. E. (1948). Sexual behavior in the human male. Philadelphia: Saunders.

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

Malinosky-Rummell, R., & Hansen, D. J. (1993). Long-term consequences of childhood physical abuse. Psychological Bulletin, 114, 68-79.

Meiselman, K. (1978). Incest. San Francisco: Jossey-Bass.

Meston, C. M., Heiman, J. R., Trapnell, P. D., & Carlin, A. S. (1999). Ethnicity, desirable responding, and self reports of abuse: A comparison of European- and Asian-ancestry undergraduates. Journal of Consulting and Clinical Psychology, 67, 139-144.

Meston, C. M., Heiman, J. R., Trapnell, P. D., & Paulhus, D. L. (1998). Socially desirable responding and sexuality self-reports, The Journal of Sex Research, 35, 148-157.

Meston, C. M., Trapnell, P. D., & Gorzalka, B. B. (1996). Ethnic and gender differences in sexuality: Variations in sexual behavior between Asian and Non-Asian university students. Archives of Sexual Behavior, 25, 33-72.

Meston, C. M., Trapnell, P. D., & Gorzalka, B. B. (1998). Ethnic, gender, and length of residency influences on sexual knowledge and attitudes. The Journal of Sex Research, 35, 176-188.

Ney, P. G., Fung, T., & Wickett, A. R. (1994). The worst combinations of child abuse and neglect. Child Abuse and Neglect, 18, 705-714.

Okami, P. (1991). Self-reports of "positive" childhood and adolescent sexual contacts with older persons: An exploratory study. Archives of Sexual Behavior, 20, 437-457.

Potterat, J. J., Rothenberg, R. B., Muth, S. Q., Darrow, W. W., & Phillips-Plummer, L. (1998). Pathways to prostitution: The chronology of sexual and drug abuse milestones. The Journal of Sex Research, 35, 333-340.

Rind, B. (1995). An analysis of human sexuality textbook coverage of the psychological correlates of adult-nonadult sex. The Journal of Sex Research, 32, 219-245.

Rind, B., & Tromovitch, P. (1997). A meta-analytic review of findings from national samples on psychological correlates of child sexual abuse. The Journal of Sex Research, 32, 219-233.

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, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.

Seidner, A. L., & Calhoun, K. S. (1984). Child sexual abuse: Factors related to differential adult adjustment. Paper presented at the Second National Family Violence Research Conference, Durham, NH.

Simpson, J. A., & Gangestad, S. W. (1991). Individual differences in sociosexuality: Evidence for convergent and discriminant validity. Journal of Personality and Social Psychology, 60, 870-883.

Swan, W. B., Jr., Wenzlaff, R. M., & Tafarodi, R. W. (1992). Depression and the search for negative evaluations: More evidence of the role of self-verification strivings. Journal of Abnormal Psychology, 101, 314-317.

Symons, D. (1987). An evolutionary approach: Can Darwin's view shed light on human sexuality? In J. H. Geer & W. O'Donahue (Eds.), Theories of human sexuality (pp. 91-122). New York: Plenum Press.

Trapnell, P. D., & Meston, C. M. (1995). The Sexual Experiences Inventory (SEI). Unpublished manuscript.

Tsai, M., Feldman-Summers, S., & Edgar, M. (1979). Childhood molestation: Differential impacts on psychosexual functioning. Journal of Abnormal Psychology, 88, 407-417.

Watkins, B., & Bentovim, A. (1992). The sexual abuse of male children and adolescents: A review of current research. Journal of Child Psychology and Psychiatry, 33, 197-248.

Wisniewski, N. (1989). A path analytic model of the aftereffects of childhood sexual victimization. Unpublished doctoral dissertation, Kent State University, Kent, OH.

Manuscript accepted July 22, 1999

This research was assisted by a postdoctoral fellowship to Cindy M. Meston from the Sexuality Research Fellowship from the Social Science Research Council with funds provided by the Ford Foundation.

Address correspondence to Cindy M. Meston, Ph.D., Department of Psychology, Mezes Hall 330, University of Texas at Austin, Austin, TX 78712; e-mail: [e-mail].

Cindy M. Meston University of Texas at Austin

Julia R. Heiman University of Washington

Paul D. Trapnell Ohio State University at Mansfield