Gendered pathways from child sexual abuse to sexual aggression victimization and perpetration in adolescence and young adulthood

Abstract

This study aimed to examine the pathways from child sexual abuse to sexual assault victimization and perpetration in adolescence and early adulthood, considering risky sexual behavior and lowered sexual self-esteem as mediator variables. In a two-wave longitudinal study with 2251 college students in Germany, male and female participants provided reports of sexual aggression victimization and perpetration since age 14 (T1) and again a year later (T2), covering the last 12 months. In addition, child sexual abuse (CSA; before the age of 14), risky sexual behavior, and sexual self-esteem were assessed at T1, and risky sexual behavior and sexual-self-esteem were assessed again at T2. Experience of CSA was significantly associated with greater likelihood of sexual aggression victimization and perpetration, lower sexual self-esteem, and more risky sexual behavior in both gender groups at T1 and was directly related to victimization at T2 among male participants. In both gender groups, CSA indirectly contributed to a higher probability of sexual victimization at T2 via its impact on victimization T1. In males, the indirect path from CSA to T2 perpetration via T1 perpetration was also significant. Through its negative impact on sexual self-esteem, CSA indirectly increased the probability of sexual victimization among women and the probability of sexual aggression perpetration among men. Risky sexual behavior mediated the pathway from CSA to sexual victimization at T2 for men and women and the pathway from CSA to sexual aggression perpetration for women. The findings contribute to the understanding of gendered effects of CSA on revictimization and the victim-to-perpetrator cycle.

Introduction

Child sexual abuse (CSA) is a risk factor for a wide range of physical, mental, and sexual health problems. One of the consistently established sequelae of child sexual abuse is an increased vulnerability to revictimization in adolescence and adulthood, documented for both female and male survivors of CSA (Classen, Palesh, & Aggarwal, 2005; Desai, Arias, Thompson, & Basile, 2002). Moreover, there is ample evidence to demonstrate that the experience of CSA is a risk for sexual aggression perpetration, especially from studies with men (e.g., Abbey, Parkhill, BeShears, Clinton-Sherrod, & Zawacki, 2006; Casey, Beadnell, & Lindhorst, 2009), but increasingly also from studies including women (e.g., Aebi et al., 2015; Brousseau, Hébert, & Bergeron, 2012).

Several theoretical explanations have been offered to account for the impact of CSA on sexual assault victimization and perpetration in later developmental stages (see Pittenger, Huit, & Hansen, 2016, for a review). A particularly influential account is Finkelhor’s (1987) “traumagenic dynamics” model, which highlights the impact of CSA on survivors’ sexual self-esteem and sexual behavior in adolescence and early adulthood. Building on this theorizing, the current study focused on the role of sexual self-esteem and risky sexual behavior as process variables underlying the path from CSA to victimization by, and perpetration of, sexual aggression in adolescence and young adulthood. By including both male and female survivors of CSA and obtaining reports of both perpetration and victimization of sexual assault in adolescence and early adulthood, the study sought to examine differential consequences of CSA in the two gender groups. A large sample of male and female college students participated in a two-wave longitudinal study in which CSA, sexual self-esteem, risky sexual behavior, and sexual aggression victimization and perpetration in adolescence were assessed at T1 and used as prospective predictors of sexual victimization and perpetration in the following 12-month period.

Sexual abuse in childhood has been established as a risk factor for both sexual aggression victimization and perpetration in later life. Several reviews concluded that survivors of CSA have a substantially higher risk of experiencing sexual victimization in adolescence and early adulthood compared to individuals not sexually abused in childhood (Classen et al., 2005, Lalor and McElvaney, 2010, Messman-Moore and Long, 2003). A meta-analysis including 19 studies yielded an effect size of d = 0.59, indicating a substantially increased risk of sexual victimization by survivors of CSA (Roodman & Clum, 2001). All studies included in the meta-analysis examined revictimization in female CSA survivors, but recent evidence confirmed a parallel association for males (Aosved, Long, & Voller, 2011; D'Abreu & Krahé, 2016).

Regarding the victim-to-perpetrator cycle, the traumagenic dynamics model proposes that traumatic sexualization, identified as a unique effect of CSA compared to other forms of childhood abuse, involves the use of sexual behavior to manipulate others (Finkelhor, 1987). Many studies confirming a link between CSA and subsequent sexual aggression perpetration are based on samples of (primarily male) sex offenders (Thomas & Fremouw, 2009). A smaller number of studies with community or student samples also showed that survivors of CSA have higher odds of sexual aggression perpetration in adolescence and young adulthood (Tharp et al., 2013). In a longitudinal study of a nationally representative sample of men, experience of CSA was a significant prospective predictor of later sexually coercive behavior toward an intimate partner (Casey et al., 2009). A longitudinal study with male college students found that CSA predicted adolescent sexual coercion, which in turn predicted sexually coercive behaviors in college (White & Smith, 2004). In a large sample of Swiss adolescents, a significant association between CSA and coercive sexual behaviors was found in both males and females. Among males, victims of contact CSA were almost four times more likely to report coercive sexual behavior than nonvictimized participants. Among females, CSA victims were almost six times more likely to report sexually coercive behavior (Aebi et al., 2015).

Low self-esteem has been established as a risk factor for both victimization by, and perpetration of, sexual assault. Self-esteem is typically measured as a global construct (e.g., French, Bi, Latimore, Klemp, & Butler, 2014), but a few studies employed more specific measures of sexual self-esteem (e.g., Van Bruggen, Runtz, & Kadlec, 2006). Sexual self-esteem (sometimes also referred to as sexual esteem) is conceptualized as an individual’s self-evaluation of worth as a sexual being (Buzwell & Rosenthal, 1996) and forms a critical part of the overall sexual self-concept (Deutsch, Hoffman, & Wilcox, 2014). In the traumagenic dynamics associated with CSA, stigmatization is a consequence of CSA that affects victims’ self-esteem (Finkelhor, 1987). Two further dynamics are feelings of betrayal and powerlessness, which are linked to depression and lowered self-efficacy. Consistent with the traumagenic dynamics model, the association between sexual assault victimization and lowered global and sexual self-esteem has been established by several studies, both assessed within a short time after the victimization experience (e.g., Feiring, Taska, & Lewis, 1998; Turner, Finkelhor, & Ormrod, 2010) and retrospectively reported in adulthood (e.g., Van Bruggen et al., 2006).

Although many studies linking CSA to lowered self-esteem included both male and female CSA survivors, research associating low sexual self-esteem with revictimization has largely focused on women (e.g., Kelley and Gidycz, 2015, Van Bruggen et al., 2006). In fact, one of the first measures for assessing sexual self-esteem was explicitly directed at women (Zeanah & Schwarz, 1996). However, the conceptual model of sexual self-esteem presented by Deutsch et al. (2014) applies to women and men. Studies including both gender groups suggest that self-esteem may yield different relationships with aspects of sexual behavior in males and females. For example, although there were no gender differences in sexual self-esteem in their study, Maas and Lefkowitz (2014) found that for males, sexual self-esteem was higher among those who had never used contraceptives during recent sexual intercourse than among those who had done so, whereas the difference was reversed for females. Therefore, the present study analyzed the potential differences in the role of sexual self-esteem as a mediator in the pathways from CSA to sexual aggression victimization and perpetration among both women and men.

As noted above, a unique dynamic identified by Finkelhor (1987) as resulting from CSA is traumatic sexualization, which is manifested at the behavioral level in precocious sexual activity, promiscuity, preoccupation with sex and compulsive sexual behavior, and sexual aggression perpetration. Operational definitions of risky sexual behavior include the number lifetime sexual partners, number of sexual partners in the past year, number of hookups, or drinking alcohol in the context of having sex (Fargo, 2009, French et al., 2014; Testa, Hoffman, & Livingston, 2010). A review of the literature by Senn, Carey, and Vanable (2008) found consistent evidence for an association of CSA with sexual risk behavior in women in community, college student, and adolescent samples. The evidence for males is more limited but also shows greater sexual risk behavior among CSA survivors. Of the different criteria of risky sexual behavior, earlier age at first intercourse and having a higher number of sexual partners showed the most consistent relationships with CSA. Senn et al. (2008) concluded that future research on the link between CSA and risky sexual behavior should include both women and men and examine the role of gender as a moderator of the association.

A study by Testa et al. (2010) found that risky sexual behavior in the first semester of college predicted sexual victimization at the end of the first year. Using a person-centered approach to identify groups of adult victims of sexual assault, French et al. (2014) found that membership in the “poly-victimized” group, reporting multiple forms of sexual aggression victimization, was not only predicted by CSA, but participants in this group also had lower self-esteem and were more likely to engage in sexual risk-taking than the non-victim group and two groups experiencing only one form of sexual assault.

Regarding perpetration, the review by Tharp et al. (2013) found consistent evidence that having multiple partners, engaging in impersonal sex, and early initiation of sex are risk factors for sexual violence perpetration. Abbey et al. (2006) found a significant cross-sectional association between engaging in casual sexual relationships and the number of sexual assaults committed in their male sample.

Prevalence rates of CSA are higher for girls than for boys (see the worldwide review by Krug, Dahlberg, Mercy, Zwi, & Lozano, 2002), but the evidence is inconclusive regarding gender differences in the psychological impact of CSA. In some studies, abused girls reported having experienced greater distress and engaged in more self-blame compared to boys (Ullman & Filipas, 2005), but other studies found the consequences of CSA to be similar for male and female survivors (Banyard, Williams, & Siegel, 2004).

Because studies examining revictimization have mostly studied women (e.g., Messman-Moore & Long, 2003) and studies examining the risk of perpetration by CSA survivors have mostly focused on men, evidence is limited on whether revictimization and victim-to-abuser patterns hold similarly for both men and women. For the impact of CSA on revictimization, the few studies that included both female and male survivors suggest that the pathways show both parallels and differences. Although both male and female survivors were found to have an increased risk of revictimization in some studies (Werner et al., 2016), there is also evidence that the effects may differ by gender. For example, Desai et al. (2002) found that the overall odds of sexual revictimization were higher for male than for female survivors of CSA, but only women had an increased risk of sexual revictimization by an intimate partner. In a recent longitudinal study, childhood sexual trauma predicted sexual risk-taking behavior in women, but not in men (Braje, Eddy, & Hall, 2016). Regarding the path from CSA to sexual aggression perpetration, the evidence is inconsistent so far. Aebi et al. (2015) and Glasser et al., 2001 found significant associations in both gender groups, whereas another study found that sexual abuse predicted sexually coercive behavior in women, but not in men (Schatzel-Murphy, Harris, Knight, & Milburn, 2009).

Because victimization and perpetration are rarely studied in the same individuals and in samples comprising both males and females, little is known about the possibility of differential pathways from CSA to subsequent victimization and perpetration in the two gender groups. An exception is the study by Brousseau et al. (2012), which included data from 209 couples. They found that in women, CSA was unrelated to sexual aggression victimization by the current or a former relationship partner but predicted a higher likelihood of sexual aggression perpetration. In men, CSA was unrelated to either victimization or perpetration.

From a theoretical perspective, differences in gender role socialization with regard to sexual behavior suggest that male and female CSA survivors might respond differently to the undermining effect of the abuse experience on their sexual self-esteem. According to the predominant sexual script, men are socialized into taking the role of initiator of sexual interactions (Morrison et al., 2015). Lowered sexual self-esteem may lead them to pursue their sexual interest through the use of aggressive tactics in order to achieve sexual gratification, suggesting a significant path from lowered sexual self-esteem to sexual aggression perpetration. By contrast, women are socialized to be gatekeepers of male sexual initiatives (Jozkowski & Peterson, 2013), and lowered sexual self-esteem may weaken their ability to reject unwanted sexual advances, suggesting a path from lowered sexual self-esteem to sexual assault victimization.