Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Jun 1.
Published in final edited form as: Child Abuse Negl. 2010 Apr 18;34(6):429–437. doi: 10.1016/j.chiabu.2009.10.006

Sex differences in childhood sexual abuse characteristics and victims’ emotional and behavioral problems: Findings from a national sample of youth

Andrea Kohn Maikovich-Fong a,✉, Sara R Jaffee b
PMCID: PMC2913705  NIHMSID: NIHMS199440  PMID: 20400178

Abstract

Objective

The first objective of this study was to test for sex differences in four childhood sexual abuse characteristics---penetration, substantiation, perpetrator familial status, and multi-maltreatment---in a national sample of youth. The second objective was to test for sex differences in how these abuse characteristics were associated with victims’ emotional and behavioral problems.

Methods

The sample was drawn from the National Survey of Child and Adolescent Well-Being, a sample of children investigated by United States child welfare services. Youth in the current study (n = 573, including 234 adolescents) were investigated for alleged sexual abuse. Logistic regression and multivariate analysis of covariance were used to test for sex differences in abuse characteristics, and to determine whether sex moderated associations between abuse characteristics and emotional and behavioral problems.

Results

Girls were more likely than boys to have their abuse substantiated and to experience penetrative abuse (although differences in penetration status did not emerge among adolescents). Substantiation status and child age were positively associated with caregiver-reported internalizing and externalizing symptoms. Sex did not moderate the relationship between abuse characteristics and youth emotional and behavioral problems.

Conclusions

Sexual abuse characteristics might not be highly predictive factors when making decisions about services needs. Furthermore, there may not be a strong empirical basis for operating on the assumption that one sex is more vulnerable to negative consequences of abuse than the other, or that abuse affects girls and boys differently. The processes explaining why some victims exhibit more impairment than others are likely complex.

Introduction

Although childhood sexual abuse (CSA) was an unacknowledged and rarely studied phenomenon until approximately thirty years ago, research has now firmly established that it is a significant public health concern (e.g. Finkelhor, Hotaling, Lewis, & Smith, 1990). Many victims experience pervasive developmental problems such as enuresis, somatic complaints, sexually reactive behavior, and academic delays (Beitchman, Zucker, Hood, da Casta, & Ackman, 1991; Kendall-Tackett, Williams, & Finkelhor, 1993; Trickett & McBride-Chang, 1995), and sexual abuse victims are especially at risk for psychopathology (Putnam, 2003; Trickett & McBride-Chang, 1995). While estimates vary widely, it is likely that around 1 in 5 girls and 1 in 6 boys are victimized prior to age 18 (Centers for Disease Control, 1997). However, despite evidence that a substantial number of victims are boys, sexual abuse research findings are based disproportionately on female samples. Consequently, the extent to which findings generalize to male victims is unclear. In addition, the clinical utility of much of the existing literature is limited by heavy reliance on adult retrospective reports of childhood abuse, and on case studies or qualitative studies whose findings have not been empirically validated with quantitative research. The objectives of the present study are to briefly review these limitations, and then to test empirically in a national sample whether 1) there are sex differences in four childhood sexual abuse characteristics---penetration versus non-penetration, substantiation status, perpetrator familial status, and experiencing multiple forms of maltreatment, and 2) whether there are sex differences in how these sexual abuse characteristics are associated with victims’ emotional and behavioral problems (as measured by internalizing, externalizing, and trauma symptoms).

Limitations to the extant literature

Preponderance of female samples

Much of what is known about the developmental correlates and sequelae of childhood sexual abuse is based on samples of exclusively female victims (Bailey & McCloskey, 2005; Finkelhor, 1984; Valente, 2005). Due in large part to the historical under-reporting and subsequent lack of awareness of male CSA, it was not until the 1980s that researchers began to make concerted efforts to include male victims in their samples (Finkelhor, 1984). Most sexual abuse researchers agree that the sexual abuse of boys is still grossly under-reported (Briggs & Hawkins, 1995; Cermak & Molidor, 1996; Porter, 1986). At least three factors likely contribute to this under-reporting. First, when the abuser is male, boys may not report the abuse for fear they will be identified as gay (Cermak & Molidor, 1996; Valente, 2005). Second, if the abuser is female, boys might interpret the abuse as a culturally condoned sexual initiation experience for which they should feel “lucky” rather than victimized (Dimock, 1988; Hunter, 1990). Third, sexual offenders typically utilize more force and threats of violence with male than with female victims when warning them not to report the abuse (Pierce & Pierce, 1985), so boys might feel more intimidated than girls about reporting.

Although studies of sexually abused boys are becoming more common, exclusively male samples are rare, usually consist of fewer than thirty participants (Feiring, Taska, & Lewis, 1999), and typically comprise adult men reporting retrospectively on their childhood abuse experiences (Briggs & Hawkins, 1995; Dhaliwal, Gauzas, Antonowicz, & Ross, 1996; Etherington, 1995). Furthermore, due to the low frequency with which males report sexual abuse outside of a therapeutic or research setting (Valente, 2005), male sexual abuse samples often comprise specialized groups such as incarcerated pedophiles, prison inmates, and members of institutions such as the armed forces or boarding schools (Darves-Bornoz, Choquet, Ledoux, Gasquet, & Manfredi, 1998), which are not necessarily representative of the population of male victims.

Finally, due to small sample sizes of male victims, many current theories about how sexual abuse affects boys and how various features of the abuse experience affect girls and boys differently have been generated from case studies, anecdotal reports, and qualitative studies (Dimock, 1988; Durham, 2003; Gilgun & Reiser, 1990; Krug, 1989). This important, in-depth work has generated many useful observations and hypotheses; however, complementary quantitative research is needed to test these hypotheses empirically.

Retrospective reporting of childhood sexual abuse

Many studies of CSA and its negative developmental sequelae have involved adults retrospectively reporting childhood abuse (Finkelhor et al., 1990). Although these studies are valuable because they may include individuals who never reported abuse as children (Kendall-Tackett & Becker-Blease, 2004), Widom and others have highlighted the importance of also conducting research with youth samples because findings often differ from those of retrospective studies (Raphael, Widom, & Lange, 2001; Widom, Weiler, & Cottler, 1999). Moreover, Hardt and Rutter (2004) identified significant measurement error in retrospective studies, particularly when individuals were asked to recall how they felt about events at the time they were happening. As many studies aim to identify associations between the experience of CSA and psychological well-being, there is clearly a need for additional studies in this field that utilize youth samples.

Current research needs

Some evidence suggests that sexual abuse might affect boys and girls differently, and that the prevalence of certain characteristics of the sexual abuse experience might differ for boys and girls (Bauserman & Rind, 1997; Darves-Bornoz et al., 1998; Feiring et al., 1999; Finkelhor et al., 1990; Fontanella, Harrington, & Zuravin, 2000; Friedrich, Urquiza, & Beilke, 1986; Gold, Elhai, Lucenko, & Swingle, 1998; Kendall-Tackett & Simon, 1992). For example, female adolescent sexual abuse victims have been shown to display more somatic complaints and mood disorders than male victims, whereas males have been shown to display more behavioral problems than females (Darves-Bornoz et al., 1998). However, some researchers have not found sex differences in the effects of CSA (Calam, Horne, Glasgow, & Cox, 1998; Young, Bergandi, & Titus, 1994), and at least one study has found that male adolescent victims tend to exhibit more emotional, behavioral, and suicidal problems than their female counterparts (Garnefski & Diekstra, 1997).

Characteristics of the abuse experience have also been shown to differ for girls and boys. One review paper found that the estimated percentage of male victims’ perpetrators who are themselves male ranges from 18 to 97, depending on the study, and that the estimated percentage of male perpetrators for female victims ranges from 80 to 100 (Dhaliwal et al., 1996). Another study of early childhood victims found that boys were more likely to experience fondling and oral intercourse than girls, whereas girls were more likely to experience penetrative abuse (Fontanella et al., 2000). Finally, girls may be more likely to be sexually abused by family members, whereas boys may be more likely to be abused by strangers (Finkelhor et al., 1990; Gold et al., 1998). Existing studies suggest that certain characteristics of the abuse experience, including the use of force and coercion, penetration, familial perpetrators, and longer duration of abuse, are associated with more negative outcomes (Beitchman et al., 1991; Browne & Finkelhor, 1986; Estes & Tidwell, 2002; Friedrich et al., 1986; Molnar, Buka, & Kessler, 2001).

In short, given how little is known about male sexual abuse victims in general, and how little is known about sex differences in the experience and consequences of childhood sexual abuse, it is important to test empirically whether potentially important abuse characteristics differ in their rates and consequences for males and females in a national sample of victims still in their youth. Social workers and clinicians working with sexually abused children within the confines of limited time and resources need a solid body of evidence to which they can refer in order to implement the most empirically-grounded assessments and treatments possible. Furthermore, some researchers and clinicians continue to operate on the assumption that boys are, in general, less adversely affected by childhood sexual abuse than are girls, but this may be a potentially dangerous assumption to operate on without solid empirical supporting evidence.

While it is important to examine how abuse characteristics affect boys and girls differently across childhood, it may be especially important to focus on the subpopulation of adolescents, as it is during this stage of life that youth experience the normative emergence of sexuality and sexual identity. These processes may be compromised and complicated by sexual abuse, leading to symptoms of psychopathology (Berliner & Conte, 1990; Durham, 2003). Furthermore, adolescence is a time of emerging self-identity and self-awareness (Erikson, 1968), and adolescents may question their own motives more and experience more self-blame for abuse than younger children (Celano, 1992; Myers, 1989).

The current study addresses the questions of whether the prevalence of different abuse characteristics differs for male and female victims, and whether there are sex differences in the association between sexual abuse characteristics and youth emotional and behavioral problems. We first examine these questions in a sample of children ranging in age from 4 to 16 years, and then focus on youth aged 11 years and older. In light of the existing literature, we predict that girls will experience higher rates of penetrative abuse and will more frequently have familial perpetrators than boys, and that because these are the characteristics most associated with psychopathology, girls will have higher rates of psychopathology symptoms than boys. However, we also predict that child sex will not moderate the association between different abuse characteristics and children’s difficulties (i.e., that these characteristics will be equally associated with psychopathology in boys and girls). Nonetheless, given the limitations to the existing literature in this field, this study is largely exploratory.

Method

Participants

The National Survey of Child and Adolescent Well-Being (NSCAW) is a nationally-representative sample of United States children who have had contact with Child Protective Services (Dowd et al., 2004). The full cohort includes 5,501 children (50% female), less than 1 year to 16 years of age when sampled, who were subjects of child abuse or neglect investigations conducted by CPS from October 1999 to December 2000. Participants in the NSCAW study gave informed consent to enroll in the study, and the study procedures were approved by the participating universities’ Institutional Review Boards. Additional information about sample composition is available from Dowd and colleagues (Dowd, Kinsey, Wheeless, & NSCAW Research Group, 2004). The present study’s sample is restricted to children who, according to caseworker reports, were investigated as alleged victims of sexual abuse. Interviews were collected 2–6 months following the close of the CPS investigations. Children who were members of the same household as a previously selected child were not eligible to participate in the study, in order to limit the burden on families. Therefore, there are not any siblings included in the NSCAW sample.

Analyses were conducted with two subsets of youth. The first subset, the “child and adolescent sample” (n = 573; 72% female), comprised children 4 years or older (M = 9.46 years; SD = 3.28). Fifty percent of children in the child and adolescent sample were White (non-Hispanic), 23% were Black (non-Hispanic), 18% were Hispanic, and 8% were of other races or ethnicities. The second subset, the “adolescent sample” (n = 234; 82% female), consisted of youth 11 years or older (M = 12.82 years; SD = 1.28). Forty-eight percent of children in the adolescent sample were White (non-Hispanic), 26% were Black (non-Hispanic), 20% were Hispanic, and 7% were of other races or ethnicities. 90% of caregivers in the sample were female, and the average annual income of families was $19,000.

Measures

Internalizing and externalizing symptoms

Caregivers were administered the Child Behavior Checklist (CBCL), which includes an externalizing scale comprising delinquent and aggressive behavior domains, and an internalizing scale comprising withdrawn behavior, somatic complaints, and anxious/depressed domains (Achenbach, 1991a). Caregivers were asked 113 questions on a 3-point Likert-type scale (0 = not true, 1 = somewhat or sometimes true, 2 = very true or often true). Internal consistency reliabilities were high for both externalizing (α = .92) and internalizing (α = .90) scales. Children’s age- and gender-standardized T scores were used in all analyses. T scores at or above 65 are considered clinically significant because this cut-off has been shown to significantly discriminate between children referred for mental health treatment and matched children who are not referred (Achenbach & Rescorla, 2001).

Youth in the adolescent sample were administered the Youth Self-Report (Achenbach, 1991b), which contains 113 questions similar to those in the CBCL. Internal consistency reliability was high (α = .90 externalizing, α = .90 internalizing). Youth’s age- and gender-standardized T scores were used in all analyses. The correlation between caregiver and youth reports of externalizing symptoms in the adolescent sample was 0.40, and the correlation for internalizing symptoms was 0.39 (both significant at the p < .01 level). This magnitude of agreement among informants is standard (Achenbach, McConaughy, & Howell, 1987). Given that parent and child reports of psychopathology do not correlate perfectly, and that adolescents may be in a better position to report on their own emotional and behavioral functioning than caregivers, it is important to examine both youth and caregiver reports of psychological functioning when available.

Trauma symptoms

Youth in the adolescent sample were administered the Post Traumatic Stress Disorder section of a version of the Trauma Symptom Checklist for Children adapted for NSCAW (Briere, 1996). The measure included ten questions asking children to describe how often they experienced various symptoms (1 = never; 2 = sometimes; 3 = lots of times; 4 = almost all of the time). For example, youth were asked how often they have “bad dreams or nightmares,” and how often “scary ideas or pictures just pop into [their] head.” Internal consistency reliability on this measure was adequate (α = 0.84). Again, youth’s age- and gender-standardized T scores were used in all analyses, and T scores at or above 65 are considered clinically significant (Briere, 1996).

Abuse characteristics

Caseworkers indicated whether or not the alleged abuse was substantiated (0 = no; 1 = yes). They also categorized the nature of the child’s sexual abuse experience, which we transformed into a dichotomous variable indicating whether the abuse was penetrative (1) or non-penetrative (0). Penetrative abuse included vaginal/anal intercourse, digital penetration of the vagina/anus, and oral copulation of an adult if the perpetrator was male. Non-penetrative abuse included fondling/molestation, masturbation, oral copulation of the child, oral copulation of an adult if the perpetrator was female, and “other less severe types.” Finally, caseworkers indicated other types of maltreatment children experienced. We transformed this information into a dichotomous variable indicating whether children experienced multiple forms of maltreatment (physical, neglect, or emotional in addition to sexual; 1) or only sexual abuse (0). Caseworkers also indicated who the alleged perpetrator was, from which we classified them as familial (1) or non-familial (0). Familial perpetrators included mothers, fathers, step-mothers, step-fathers, grandmothers, grandfathers, aunts, uncles, brothers, sisters, and other relatives. Non-familial perpetrators included mothers’ boyfriends, neighbors, strangers, friends, and child care providers.

Analysis approach

Logistic regression was used to test the hypotheses that girls were more likely than boys to have experienced penetrative abuse and to have familial perpetrators. Two multivariate analyses of covariance (MANCOVAs) were conducted to test whether there were sex differences in the association between abuse characteristics and youth emotional and behavioral problems in the child and adolescent sample and the adolescent sample. Statistical approaches that require omnibus test significance before examining specific contrasts are most appropriate when the research question has yielded limited and/or mixed results in the extant literature (Burchinal & Clarke-Stewart, 2007), such as the current research question. This type of conservative analysis approach minimizes the risk of Type I error, unlike alternative a priori comparison approaches such as regression (Kirk, 1982). Given that the NSCAW weights are highly variant whole sample weights, they are not appropriate to use with small subsamples, and so were not used in these analyses. All analyses were conducted using SPSS 14.0 (2005).

Results

Missing data

Missing data ranged from less than 1% for measures of emotional and behavioral problems to 26% for characteristics of the sexual abuse. Because analyses based on listwise-deleted data have been shown to generate biased and inefficient parameter estimates (Schafer & Graham, 2002), multiple imputation was used to generate a set of complete observations for all sample members. Five multiply imputed data sets were created using the Stata 9.0 (StataCorp, 2005) user-written add-on program ICE (Imputation by Chained Equations) (Royston, 2005). ICE imputes missing values using an iterative regression switching procedure (Royston, 2004; Royston, 2005). By default, ICE uses linear regression to estimate values for any incomplete continuous variable, and logistic regression to estimate values for any incomplete dichotomous variable. The imputed values are obtained by sampling from the distribution of the incomplete variable, given the observed values and explanatory variables included in the predictive model. One advantage of ICE is that it does not assume normality of the joint multivariate distribution of variables, so different types of variables (e.g., continuous, categorical) can be imputed simultaneously. For a description of how ICE combines the estimates and obtains standard errors, see Carlin, Li, Greenwood, and Coffey (2003).

Descriptive data and correlations among variables

Tables 1 and 2 provide descriptive statistics for abuse characteristics and children’s psychopathology symptoms. Table 3 provides a correlation matrix displaying relationships among all analysis variables. Statistically significant correlations were small in magnitude, indicating that multicollinearity was not a problem.

Table 1.

Sexual Abuse Characteristics: Descriptive and Inferential Statistics

Sample subset (n) Percent substantiated (n) Percent penetrative (n) Percent familial perpetrator (n) Percent experiencing multiple maltreatment (n)
Child and adolescent (423) 66.9 (283) 43.5 (184) 67.1 (284) 27.4 (116)
 Boys only (117) 50.4 (59) 30.8 (36) 65.8 (77) 23.9 (28)
 Girls only (306) 73.2 (224) 48.4 (148) 67.6 (207) 28.8 (88)
OR [95% confidence interval]1 2.47** [1.69, 3.61] 2.05** [1.31, 3.20] 1.09 [0.65, 1.82] 1.26 [0.87, 1.85]
Adolescent (176) 73.3 (129) 50.6 (89) 65.9 (116) 29.0 (51)
 Boys only (30) 63.3 (19) 53.3 (16) 63.3 (19) 36.7 (11)
 Girls only (146) 75.3 (110) 50.0 (73) 66.4 (97) 27.4 (40)
OR [95% confidence interval]1 2.45* [1.22, 4.92] 1.05 [0.50, 2.24] 1.37 [0.55, 3.38] 0.74 [0.37, 1.50]
Open in a new tab

Note: Percentages are based on non-imputed data; odds ratios are based on imputed data.

1

Odds ratios reflect the odds of experiencing each abuse characteristic for girls relative to boys.

*

p < .05;

**

p < .01

Table 2.

Emotional and Behavioral Problems: Descriptive Statistics

Sample subset (n) Mean externalizing score1 (SD)/% clinically significant Mean internalizing score1 (SD)/% clinically significant Mean trauma score1 (SD)/% clinically significant
Child and adolescent, caregiver report (423) 60.35 (12.22)/38 57.26 (11.44)/28 n/a
 Boys only (117) 60.72 (12.17)/39 58.96 (11.52)/36
 Girls only (306) 60.20 (12.25)/38 56.60 (11.36)/25
Adolescent, caregiver report (176) 61.67 (12.49)/44 56.71 (11.50)/25 n/a
 Boys only (30) 60.33 (12.88)/33 57.20 (11.58)/30
 Girls only (146) 61.95 (12.44)/46 56.61 (11.53)/24
Adolescent, youth report (176) 55.41 (11.76)/27 51.64 (11.82)/13 51.10 (10.96)/12
 Boys only (30) 53.57 (11.74)/27 53.40 (12.99)/13 52.47 (11.07)/10
 Girls only (146) 55.79 (11.76)/27 51.27 (11.58)/12 50.81 (10.96)/12
Open in a new tab
1

Externalizing, internalizing, and trauma symptoms scores are gender- and age-standardized T scores.

Table 3.

Correlation Matrix of All Variables in MANCOVAs (N = 423)

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.
1. Child sex
2. Age 0.22**
3. Penetrative Abuse 0.16** 0.16**
4. Substantiated 0.22** 0.14** 0.25**
5. Familial Perpetrator 0.02 −0.02 0.09 0.14**
6. Multiple Maltreatment 0.05 0.05 0.13** 0.05 0.08*
7. Caregiver-rated internalizing −0.09 0.04 −0.01 0.10 −0.01 −0.03
8. Caregiver-rated externalizing −0.02 0.12** 0.05 0.11* −0.03 0.03 0.59**
9. Youth-rated internalizing −0.07 0.06 −0.06 0.02 0.05 0.05 0.46** 0.26**
10. Youth-rated externalizing 0.07 0.15 0.03 0.03 0.04 −0.01 0.40** 0.49** 0.61**
11. Youth-rated trauma −0.04 −0.08 0.05 0.09 0.08 0.10 0.33** 0.19** 0.63** 0.45**
Open in a new tab

Note: Correlations displayed between dichotomous variables are phi correlations.

*

p < .05;

**

p <.01

Sex differences in the likelihood of experiencing different abuse characteristics

Logistic regression analyses revealed some sex differences in the likelihood of experiencing different sexual abuse characteristics. Girls in the child and adolescent sample had significantly higher odds of having their abuse substantiated and of experiencing penetrative abuse (Table 1). Girls and boys were equally likely to have a familial perpetrator and to experience multiple forms of maltreatment. In the adolescent sample, girls were more likely than boys to have their abuse substantiated, but did not differ from boys on any of the other three abuse characteristics.

Sex differences in associations between abuse characteristics and youth emotional and behavioral problems

The MANCOVA conducted with the child and adolescent sample included the dependent variables of caregiver-rated internalizing and externalizing symptoms. The child’s age was entered as a covariate, and five variables were entered as fixed effects: child sex, penetration status, substantiation status, perpetrator familial status, and multiple maltreatment status. The multivariate test (Wilks’ Lambda) was significant for substantiation status and child age (Table 4). The univariate tests of substantiation and age were significant for externalizing and internalizing symptoms. Children with substantiated abuse cases had higher externalizing (M = 60.75, SD = 12.09) and internalizing (M = 57.96, SD = 11.35) scores than children with unsubstantiated cases (externalizing M = 57.84, SD = 12.64; internalizing M = 55.31, SD = 11.74). Older children had more externalizing and internalizing symptoms than younger children. No other multivariate test for main or interactive effects was significant.

Table 4.

MANCOVA: Association Between Abuse Characteristics and Caretaker-Reported Emotional and Behavior Symptoms (Child and Adolescent Sample)

Source Multivariate Test (Wilks’ Lambda) Multivariate F Effect Size1 Univariate F: Extern. Univariate F: Intern.
Child age 0.98 5.33** 0.02 10.27** 6.17**
Child sex 0.99 2.45 0.01
Penetration 1.00 0.49 0.00
Substantiation 0.98 4.38* 0.02 5.31* 8.32**
Familial perpetrator 1.00 0.50 0.00
Multiple maltreatment 0.99 1.66 0.01
Child gender × Penetration 0.99 2.83 0.01
Child gender × Substantiation 1.00 0.38 0.00
Child gender × Familial perpetrator 1.00 0.90 0.00
Child gender × Multiple maltreatment 1.00 1.05 0.00
Open in a new tab
1

Effect size is partial eta squared.

Note:

*

p < .05

A second MANCOVA was conducted to investigate sex differences in the association between abuse characteristics and youth emotional and behavioral problems within the adolescent sample. The dependent variables were caregiver-reported internalizing and externalizing symptoms and youth-reported internalizing, externalizing, and trauma symptoms. Covariates and fixed effects were identical to those in the analysis of the child and adolescent sample. No multivariate test for main or interactive effects was significant, with the exception of one uninterpretable result. Specifically, the multivariate test (Wilks’ Lambda) was significant for penetration. However, the univariate test of penetration was not significant for any outcome measure (full analyses available upon request).

Discussion

This study contributes to the limited body of quantitative research that examines sex differences in the experience and psychological sequelae of childhood sexual abuse in youth samples. The study had several methodological strengths. First, the samples were relatively large by sexual abuse sample standards, and were drawn from a national sample of children involved with the United States CPS system. Second, the samples included enough male victims to allow for tests of sex differences in the rates and correlates of specific abuse characteristics. Third, the NSCAW dataset included information about a number of potentially important sexual abuse characteristics. Fourth, the sample was large enough to allow for tests of sex differences in the effects of abuse specifically among adolescents, for whom issues of sexual identity are especially salient. Youth reports of their own emotional and behavioral functioning were also available for adolescents in this sample.

Implications for future research

The analyses revealed sex differences in the prevalence of two abuse characteristics: girls were more likely than boys to have their abuse substantiated, and girls in the child and adolescent sample were more likely to experience penetrative abuse. The sex differences in substantiation rates is interesting, and might reflect either a true sex difference, or a bias in what types of cases end up substantiated by CPS. While differentiating between these two potential explanations is outside the scope of the present paper, it is an important question to examine in future studies because there may be a subset of victims, such as boys, particularly at-risk for having their abuse unsubstantiated. Furthermore, there may be a subset of victims at risk for having their abuse undetected and/or unreported in the first place. For example, perhaps young boys who experience penetrative abuse are particularly at-risk for not having their abuse detected and reported to CPS, due in part to societal lack of awareness of sexual abuse of young boys, and/or perpetrators’ more concerted efforts to hide their abuse of boys.

Despite some sex differences in the prevalence of abuse characteristics, there were not any sex differences in the severity of internalizing, externalizing, or trauma symptoms. Although other studies have shown that males and females respond differently to sexual abuse (e.g., Bauserman & Rind, 1997; Friedrich et al., 1986), many of those studies identified sex differences in specific symptoms like nightmares and somatic complaints (Darves-Bornoz et al., 1998), whereas the present study examined broad categories of mental health symptoms. Thus, there may be sex differences in the association between sexual abuse and specific psychopathology symptoms, but not many differences at more global levels. Methodological differences among studies could also account for discrepant findings. Many studies that have found sex differences either utilized child samples with smaller age ranges than the present sample (Fontanella et al., 2000), or used samples of adults reporting retrospectively on childhood abuse (Bauserman & Rind, 1997; Dhaliwal et al., 1996). Although these methodological differences make cross-study comparisons difficult, the present results, combined with those from other studies that failed to find sex differences in how sexual abuse affects children’s mental health (Calam et al., 1998; Young et al., 1994), suggest that further study is needed before researchers, social workers, and clinicians can conclude that sexual abuse is more psychologically damaging for one sex than the other.

Furthermore, it is possible that small sample sizes have detracted researchers from publishing studies about sex differences in the experience and consequences of sexual abuse. For example, if analyses that yield null results have been attributed to small sample sizes and have remained unpublished, and if the few studies that have yielded sex differences (true differences or those produced as a result of Type I error) have been published, then the literature base in this field may be biased. In other words, it is possible that the file drawer problem, in which researchers fail to publish non-significant findings (Rosenthal, 1979), may be operating within this field due to the small sizes of samples that include males. It is important, then, that more studies such as the present one enter the literature base not only for the information they contain individually, but also to inform future meta-analyses.

This study did not examine the role of perpetrator sex in sexually abused girls’ and boys’ symptomatology, because information on perpetrator sex was missing for many children and the number of youth with known female perpetrators was very small. However, researchers with access to samples that include more youth with female perpetrators should test the qualitatively-derived hypothesis that having a male perpetrator may be particularly damaging for male adolescent victims because they are in a stage of development in which issues related to sexuality are normatively expected to emerge (Durham, 2003; Erikson, 1968). For example, in one practitioner study, male victims sexually abused by men during adolescence expressed significant concerns about their sexuality, and fear that their peers would learn of the abuse and consider them gay as a result of it (Durham, 2003). Given that there is a natural biological increase in sexual drives and sexual awareness during adolescence (Moore & Rosenthal, 1993), male victims sexually abused by men often report feeling some physical pleasure during the abuse, which is associated with extreme shame and confusion about the experience, their sexuality, and their sexual identities (Durham, 2003). Furthermore, it is possible that boys sexually abused by women may be particularly unlikely to report their abuse. Thus, this is an important domain of study for future research efforts.

Implications for clinical intervention

Although victims in the child and adolescent sample whose abuse was substantiated had more externalizing and internalizing symptoms than victims with unsubstantiated abuse, these results did not replicate within the adolescent sample. Furthermore, neither familial perpetrator status nor multiple maltreatment predicted differences in emotional or behavioral problems in either sample, despite the relatively large sample sizes. These findings suggest that it may be problematic to assume that a child is a low priority for intervention services because he/she did not have his/her abuse substantiated (especially among adolescents), or because the abuse did not fit a profile of what is typically perceived of as more “severe.” Assuming that an adolescent is less at risk for behavioral, emotional, and social difficulties because their sexual abuse was not substantiated, for example, may result in “missing” multiple survivors in need of treatment.

The present study also tested whether the association between sexual abuse characteristics and emotional and behavioral problems differed in males and females. In the child and adolescent sample and in the adolescent sample, there were no significant interactions between child sex and sexual abuse characteristics. Thus, although certain characteristics of sexual abuse are more common in females than males, these characteristics appear to affect girls and boys similarly. In sum, this study suggests that it may not be particularly straightforward to design abuse severity profiles for risk assessments based on abuse characteristics or victim sex alone or in combination. Furthermore, it is important for clinicians to assess for various types of psychopathology as comprehensively in males as in females, given that child sex does not appear to be a moderator of the relationship between abuse characteristics and child outcomes. Finally, although age was not a focal variable of this study, age did emerge as a significant predictor of children’s psychopathology symptoms. This suggests that it may be important to ensure that young adolescents’ mental health needs are not neglected, and that age-appropriate treatments for older youth continue to be developed and implemented.

Limitations

Although the sample sizes were relatively large by previous study standards, power to detect sex differences in emotional and behavioral problems, abuse characteristics, and the association between abuse characteristics and emotional and behavioral problems was still limited in the adolescent subsample. Second, while this study considered four characteristics that can differ among sexual abuse experiences, there are other potentially important characteristics that were not examined, such as duration, age-of-onset of abuse, and perpetrator sex. Third, it is likely that the children in this sample were also experiencing other types of stressors, such as poverty and community violence, that may also have been contributing to their emotional and behavioral problems, and that may have been differentially predictive for males and females. This study does not tease apart the unique contribution of sexual abuse or shed any light on how children’s contexts further moderate experiences of abuse on their psychological well-being. Fourth, this study examined children’s emotional and behavioral symptoms, not explicit diagnoses.

Fifth, there was not enough statistical power to consider how girls and boys of different races and ethnicities were affected by characteristics of the abuse experience, and the only variable related to race or ethnicity in the NSCAW dataset was the child’s race, identified by caregivers as one of a limited set of forced choices. The cultures in which children of different races and ethnicities grow up may shape very different understandings of sexual abuse experiences, and the balance of risk and protective factors that characterize sexually abused boys and girls may vary greatly as a function of race and ethnicity. Thus, it is not clear that our findings will generalize to racial and ethnic subgroups. Whereas child sex did not emerge as a moderator in this study, it is possible that other demographic and contextual factors---such as the child’s ethnicity and/or culture---may serve as moderators. Sixth, the imputation procedure we used assumes that data are missing completely at random (MCAR). To the extent that this is not true, our results may be biased. Seventh, one of our hypotheses (that child sex will not moderate the association between abuse characteristics and children’s difficulties) was a null hypothesis. One possibility is that this hypothesis was confirmed due to limited power to detect a moderating effect, rather than a true lack of moderation.

Despite these limitations, the present study provides new quantitative data on sex differences in the likelihood of experiencing different sexual abuse characteristics, and the ways in which sex moderates the association between these characteristics and children’s emotional and behavioral problems. The national nature of the sample, the sample’s size, and the fact that victims in the sample were still in their youth contribute to the novelty and strength of the study. Nonetheless, significantly more research on sex differences in children’s responses to sexual abuse (and other types of abuse and trauma) is needed before clinicians and intervention specialists can design and implement the most empirically-grounded assessments and treatments possible for this vulnerable population of youth. Given the inherent dangers of the file drawer problem, in which researchers fail to publish findings that show few significant differences (Rosenthal, 1979), it is crucial that studies such as the present one enter the literature base.

Acknowledgments

This work was supported by grant HD050691 from the National Institute of Child Health and Human Development to S. Jaffee. This document includes data from the National Survey of Child and Adolescent Well-Being, which was developed under contract with the Administration on Children, Youth, and Families, U.S. Department of Health and Human Services (ACYF/DHHS). The data have been provided by the National Data Archive on Child Abuse and Neglect.

Footnotes

The information and opinions expressed herein reflect solely the position of the authors. Nothing herein should be construed to indicate the support or endorsement of its content by ACYF/DHHS.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Achenbach TM. Manual for the child behavior checklist 4–18 and 1991 profile. Burlington: Department of Psychiatry, University of Vermont; 1991a. [Google Scholar]
  2. Achenbach TM. Manual for the youth self-report and 1991 profile. Burlington: Department of Psychiatry, University of Vermont; 1991b. [Google Scholar]
  3. Achenbach TM, McConaughy SH, Howell CT. Child/adolescent behavioral and emotional problems: Implication of cross-informant correlations for situational specificity. Psychological Bulletin. 1987;101:213–232. [PubMed] [Google Scholar]
  4. Achenbach TM, Rescorla LA. Manual for ASEBA school-age forms & profiles. Burlington, VT: University of Vermont, Research Center for Children, Youth, & Families; 2001. [Google Scholar]
  5. Bailey JA, McCloskey LA. Pathways to adolescent substance use among sexually abused girls. Journal of Abnormal Child Psychology. 2005;33:39–53. doi: 10.1007/s10802-005-0933-0. [DOI] [PubMed] [Google Scholar]
  6. Bauserman R, Rind B. Psychological correlates of male child and adolescent sexual experiences with adults: A review of the nonclinical literature. Archives of Sexual Behavior. 1997;26:181–210. doi: 10.1023/a:1024581610658. [DOI] [PubMed] [Google Scholar]
  7. Beitchman JH, Zucker KJ, Hood JE, da Casta GA, Ackman D. A review of the short-term effects of child sexual abuse. Child Abuse & Neglect. 1991;15:537–556. doi: 10.1016/0145-2134(91)90038-f. [DOI] [PubMed] [Google Scholar]
  8. Berliner L, Conte JR. The process of victimization: The victims’ perspective. Child Abuse & Neglect. 1990;14:29–40. doi: 10.1016/0145-2134(90)90078-8. [DOI] [PubMed] [Google Scholar]
  9. Briere J. Trauma symptom checklist for children: Professional manual. Florida: Psychological Assessment Resources, Inc; 1996. [Google Scholar]
  10. Briggs F, Hawkins RMF. Protecting boys from the risk of sexual abuse. Early Child Development and Care. 1995;110:19–32. [Google Scholar]
  11. Browne A, Finkelhor D. The impact of child sexual abuse: A review of the research. Psychological Bulletin. 1986;99:66–77. [PubMed] [Google Scholar]
  12. Burchinal MR, Clarke-Stewart KA. Maternal employment and child cognitive outcomes: The importance of analytic approach. Developmental Psychology. 2007;43:1140–1155. doi: 10.1037/0012-1649.43.5.1140. [DOI] [PubMed] [Google Scholar]
  13. Calam R, Horne L, Glasgow D, Cox A. Psychological disturbance and child sexual abuse: A follow-up study. Child Abuse & Neglect. 1998;22:901–913. doi: 10.1016/s0145-2134(98)00068-4. [DOI] [PubMed] [Google Scholar]
  14. Carlin JB, Li N, Greenwood P, Coffey C. Tools for analyzing multiple imputed datasets. Stata Journal. 2003;3:226–244. [Google Scholar]
  15. Celano MP. A developmental model of victims’ internal attributions of responsibility for sexual abuse. Journal of Interpersonal Violence. 1992;7:57–69. [Google Scholar]
  16. Centers for Disease Control. Adverse Childhood Experiences Study. 1997 Retrieved February 22, 2009, from http://www.cdc.gov/nccdphp/ace/prevalence.htm.
  17. Cermak P, Molidor C. Male victims of child sexual abuse. Child & Adolescent Social Work Journal. 1996;13:385–400. [Google Scholar]
  18. Darves-Bornoz JM, Choquet M, Ledoux S, Gasquet I, Manfredi R. Gender differences in symptoms of adolescents reporting sexual assault. Social Psychiatry and Psychiatric Epidemiology. 1998;33:111–117. doi: 10.1007/s001270050030. [DOI] [PubMed] [Google Scholar]
  19. Dhaliwal GK, Gauzas L, Antonowicz DH, Ross RR. Adult male survivors of childhood sexual abuse: Prevalence, sexual abuse characteristics, and long-term effects. Clinical Psychology Review. 1996;16:619–639. [Google Scholar]
  20. Dimock PT. Adult males sexually abused as children: Characteristics and implications for treatment. Journal of Interpersonal Violence. 1988;3:203–221. [Google Scholar]
  21. Dowd K, Kinsey S, Wheeless S NSCAW Research Group. National survey of child and adolescent well-being (NSCAW): Wave 1 data file user’s manual. Research Triangle Park, NC: Research Triangle Institute; 2004. [Google Scholar]
  22. Dowd K, Kinsey S, Wheeless S, Thissen R, Richardson J, Suresh R, Mierzwa F, Biemer P, Johnson I, Lytle T. National survey of child and adolescent well-being (NSCAW): Combined waves 1–4 data file user’s manual. Ithaca, NY: National Data Archive on Child Abuse and Neglect; 2004. [Google Scholar]
  23. Durham A. Young men living through and with child sexual abuse: A practitioner research study. British Journal of Social Work. 2003;33:309–323. [Google Scholar]
  24. Etherington K. Adult male survivors of childhood sexual abuse. Counselling Psychology Quarterly. 1995;8:233–241. [Google Scholar]
  25. Erikson E. Identity, youth, and crisis. New York: Norton; 1968. [Google Scholar]
  26. Estes LS, Tidwell R. Sexually abused children’s behaviours: Impact of gender and mother’s experience of inra- and extra-familial sexual abuse. Family Practice. 2002;19:36–44. doi: 10.1093/fampra/19.1.36. [DOI] [PubMed] [Google Scholar]
  27. Feiring C, Taska L, Lewis M. Age and gender differences in children’s and adolescents’ adaptation to sexual abuse. Child Abuse & Neglect. 1999;23:115–128. doi: 10.1016/s0145-2134(98)00116-1. [DOI] [PubMed] [Google Scholar]
  28. Finkelhor D. Child sexual abuse: Theory and research. New York: Free Press; 1984. [Google Scholar]
  29. Finkelhor D, Hotaling G, Lewis IA, Smith C. Sexual abuse in a national survey of adult men and women: Prevalence, characteristics, and risk factors. Child Abuse & Neglect. 1990;14:19–28. doi: 10.1016/0145-2134(90)90077-7. [DOI] [PubMed] [Google Scholar]
  30. Fontanella C, Harrington D, Zuravin SJ. Gender differences in the characteristics and outcomes of sexually abused preschoolers. Journal of Child Sexual Abuse. 2000;9:21–40. [Google Scholar]
  31. Friedrich WN, Urquiza AJ, Beilke RL. Behavior problems in sexually abused young children. Journal of Pediatric Psychology. 1986;11:47–57. doi: 10.1093/jpepsy/11.1.47. [DOI] [PubMed] [Google Scholar]
  32. Garnefski N, Diekstra RFW. Child sexual abuse and emotional and behavioral problems in adolescence: Gender differences. Journal of the American Academy of Child & Adolescent Psychiatry. 1997;36:323–329. doi: 10.1097/00004583-199703000-00010. [DOI] [PubMed] [Google Scholar]
  33. Gilgun JF, Reiser E. The development of sexual identity among men sexually abused as children. Families in Society. 1990;71:515–523. [Google Scholar]
  34. Gold SN, Elhai JD, Lucenko BA, Swingle JM. Abuse characteristics among childhood sexual abuse survivors in therapy: A gender comparison. Child Abuse & Neglect. 1998;22:1005–1012. doi: 10.1016/s0145-2134(98)00085-4. [DOI] [PubMed] [Google Scholar]
  35. Hardt J, Rutter M. Validity of adult retrospective reports of adverse childhood experiences: Review of the evidence. Journal of Child Psychology and Psychiatry. 2004;45:260–273. doi: 10.1111/j.1469-7610.2004.00218.x. [DOI] [PubMed] [Google Scholar]
  36. Hunter M. Abused boys: The neglected victims of sexual abuse. New York: Fawcett Columbine; 1990. [Google Scholar]
  37. Kendall-Tackett K, Becker-Blease K. The importance of retrospective findings in child maltreatment research. Child Abuse & Neglect. 2004;28:723–727. doi: 10.1016/j.chiabu.2004.02.002. [DOI] [PubMed] [Google Scholar]
  38. Kendall-Tackett KA, Simon AF. A comparison of the abuse experiences of male and female adults molested as children. Journal of Family Violence. 1992;7:57–62. [Google Scholar]
  39. Kendall-Tackett KA, Williams LM, Finkelhor D. Impact of sexual abuse on children: A review and synthesis of recent empirical studies. Psychological Bulletin. 1993;113:164–180. doi: 10.1037/0033-2909.113.1.164. [DOI] [PubMed] [Google Scholar]
  40. Kirk RE. Experimental design: Procedures for the behavioral sciences. 2. Monterey, CA: Brooks/Cole Publishing Co; 1982. [Google Scholar]
  41. Krug RS. Adult male report of childhood sexual abuse by mothers: Case descriptions, motivations, and long-term consequences. Child Abuse & Neglect. 1989;13:111–119. doi: 10.1016/0145-2134(89)90034-3. [DOI] [PubMed] [Google Scholar]
  42. Molnar BE, Buka SL, Kessler RC. Child sexual abuse and subsequent psychopathology: Results from the National Comorbidity Survey. American Journal of Public Health. 2001;91:753–760. doi: 10.2105/ajph.91.5.753. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Moore S, Rosenthal D. Sexuality in adolescence. London: Routeledge; 1993. [Google Scholar]
  44. Myers MF. Men sexually assaulted as adults and sexually abused as boys. Archives of Sexual Behavior. 1989;18:203–215. doi: 10.1007/BF01543195. [DOI] [PubMed] [Google Scholar]
  45. Pierce R, Pierce LH. The sexually abused child: A comparison of male and female victims. Child Abuse & Neglect. 1985;9:191–199. doi: 10.1016/0145-2134(85)90011-0. [DOI] [PubMed] [Google Scholar]
  46. Porter E. Treating the young male victim of sexual assault: Issues and intervention strategies. Syracuse, NY: Safer Society Press; 1986. [Google Scholar]
  47. Putnam FW. Ten-year research update review: Child sexual abuse. Journal of the American Academy of Child & Adolescent Psychiatry. 2003;42:269–278. doi: 10.1097/00004583-200303000-00006. [DOI] [PubMed] [Google Scholar]
  48. Raphael KG, Widom CS, Lange G. Childhood victimization and pain in adulthood: A prospective investigation. Pain. 2001;92:283–293. doi: 10.1016/s0304-3959(01)00270-6. [DOI] [PubMed] [Google Scholar]
  49. Rosenthal R. The “file drawer problem” and tolerance for null results. Psychological Bulletin. 1979;86:638–641. [Google Scholar]
  50. Royston P. Multiple imputation of missing values: Update of ice. Stata Journal. 2005;5:527–536. [Google Scholar]
  51. Royston P. Multiple imputation of missing values. Stata Journal. 2004;4:227–241. [Google Scholar]
  52. StataCorp. Stata statistical software: Release 9. College Station, TX: StataCorp LP; 2005. [Google Scholar]
  53. Trickett PK, McBride-Chang C. The developmental impact of different forms of child abuse and neglect. Developmental Review. 1995;15:311–337. [Google Scholar]
  54. Valente SM. Sexual abuse of boys. Journal of Child and Adolescent Psychiatric Nursing. 2005;18:10–16. doi: 10.1111/j.1744-6171.2005.00005.x. [DOI] [PubMed] [Google Scholar]
  55. Widom CS, Weiler BL, Cottler LB. Childhood victimization and drug abuse: A comparison of prospective and retrospective findings. Journal of Consulting and Clinical Psychology. 1999;67:867–880. doi: 10.1037//0022-006x.67.6.867. [DOI] [PubMed] [Google Scholar]
  56. Young RE, Bergandi TA, Titus TG. Comparison of the effects of sexual abuse on male and female latency-aged children. Journal of Interpersonal Violence. 1994;9:291–306. [Google Scholar]

RESOURCES