Child maltreatment and perceived family environment as risk factors for adult rape: is child sexual abuse the most salient experience

Child Abuse & Neglect 28 (2004) 1019–1034

Child maltreatment and perceived family environment as riskfactors for adult rape: is child sexual abuse the most salient

experience?�

Terri L. Messman-Moore∗, Amy L. Brown

Department of Psychology, Benton Hall, Miami University, Oxford, OH 45056, USA

Received 21 August 2002; received in revised form 26 April 2004; accepted 12 May 2004

Abstract

Objective: Child maltreatment and family functioning were examined as predictors of adult rape in a sample of925 college women.Method: Information was obtained from retrospective self-report questionnaires. Child sexual abuse (CSA) wasassessed with the Life Experiences Questionnaire, child emotional abuse (CEA) and physical abuse (CPA) wereassessed with the Childhood Trauma Questionnaire, family functioning was assessed with the Family EnvironmentScale, and adult rape was assessed with the Sexual Experiences Survey.Results:Approximately 17% of women reported a history of any form of child abuse: 8.9% reported CSA, 4.2%reported CPA, and 8.6% reported CEA. Fifteen percent of participants reported a history of rape since age 17.CSA and CEA, but not CPA, were related to adult rape. Experiencing multiple forms of child abuse increased therisk of adult rape. Logistic regression analyses indicated that CSA predicted adult rape (OR = 1.9). Further, lowlevels of family cohesion predicted adult rape only in the absence of CEA (OR = 1.2), and low levels of emotionalexpressiveness in the family predicted adult rape only in the presence of CEA (OR = 1.4).Conclusions:Findings highlight the need to consider multiple forms of child abuse as well as family functioningin relation to the risk of adult sexual victimization. Findings suggest CSA is a risk factor for adult victimizationindependent of family functioning, and family dysfunction exerts an impact on rape only in relation to CEA.© 2004 Elsevier Ltd. All rights reserved.

Keywords:Revictimization; Rape; Sexual abuse; Physical abuse; Emotional abuse; Family functioning

� This project was supported by grants awarded to the first author from the Committee for Faculty Research and the College of Arts &Sciences at Miami University.

∗ Corresponding author.

0145-2134/$ – see front matter © 2004 Elsevier Ltd. All rights reserved.doi:10.1016/j.chiabu.2004.05.003

1020 T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034

Introduction

The experience of sexual abuse in childhood (CSA) has been identified as a risk factor for future sexualvictimization, the presence of which may increase the likelihood of rape in adulthood by as much as 11times (Fergusson, Horwood, & Lynskey, 1997). Sexual revictimization occurs for between 15% and 79%of CSA survivors (Roodman & Clum, 2001) and recently has been the subject of much study (for reviewsseeArata, 2002; Breitenbecher, 2001; Messman & Long, 1996; Roodman & Clum, 2001). However,much of the research on revictimization has focused on the experience of CSA survivors, with relativelyfewer studies examining the role of other forms of child abuse or trauma in relation to adult victimization.Examining only one form of child abuse may be problematic, however, as there is considerable overlapin the occurrence of different forms of childhood abuse (Higgins & McCabe, 2001). The combined orinteractive effects of multiple types of child abuse are not assessed in many studies focusing on the long-term effects of CSA, despite some evidence that childhood sexual and physical abuse are often associated.Furthermore, there is some indication that emotional difficulties once thought to be the sequelae of CSAare actually related to other traumatic childhood and adult experiences (Banyard, Williams, & Siegel,2001).

In addition to childhood maltreatment, other childhood factors, such as the family environment, affectthe risk of psychological dysfunction (Nash, Hulsey, Sexton, Harralson, & Lambert, 1993). Early researchtended to focus on CSA to the exclusion of other factors (e.g., other forms of childhood abuse, familydysfunction, or marital conflict). Therefore, although there is ample evidence that CSA experiencesare associated with psychological difficulties both during childhood and adulthood (Beitchman, Zucker,Hood, DaCosta, & Akman, 1991; Polusny & Follette, 1995), it is difficult to determine how much impactCSA alone has on later psychological distress.Briere (1988)identified two competing views regardingthe role of family factors and CSA in relation to psychological functioning. One suggests that disturbedfamily dynamics contribute to the occurrence of childhood abuse and neglect (as well as other negativeexperiences), and that these family dynamics, and not child abuse, cause later psychological problems. Theother perspective assumes that CSA has specific negative effects that impact psychological functioning,and that these effects occur beyond the effects of dysfunctional family dynamics. Disentangling theimpact of dysfunctional family functioning and childhood abuse is difficult and not without controversy(see discussion betweenNash et al., 1993andBriere & Elliott, 1993). Nevertheless, although it may bevery difficult to separate the impact of family dynamics and child abuse, it is also apparent that both factorsshould be assessed and examined in relation to subsequent psychological distress and related problemssuch as revictimization.

Child physical abuse and revictimization

Interestingly, although much research regarding revictimization focuses on CSA rather than CPA, manystudies that examine the relative importance of CSA and CPA in relation to adult victimization fail tosupport the individual role of CSA. The one exception occurred in a large sample of female Navy recruits,in which Merrill et al. (1999)examined the impact of CPA and CSA in relation to adult rape. In thatstudy, CSA predicted adult rape when CSA, CPA, and combined abuse were examined simultaneously, andwhen statistically controlling for CPA. Other studies found support for the role of CPA in revictimization.Cloitre, Tardiff, Marzuk, Leon, and Portera (1996)found that women reporting CPA or combined CSAand CPA were more likely to report adult sexual assault in a clinical sample, andSchaaf and McCanne

T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034 1021

(1998)found similar results with a group of college women. Authors of both studies concluded that theidentification of CSA as a risk factor for adult victimization in previous studies occurred because CSAoften occurs in the context of CPA. However, in at least one study, the two forms of child abuse wererelated to revictimization only in combination with one another (Wind & Silvern, 1992). The majority ofstudies did not discuss the identity of the perpetrator of the adult victimization experiences.Desai, Arias,Thompson, and Basile (2002)examined the relationship between child abuse and adult victimizationby both intimate partners and non-intimates among a nationally representative sample of 8,000 women.Only CPA and combined abuse were significant predictors of lifetime sexual victimization by an intimatepartner. However, CSA, CPA, and combined abuse predicted sexual victimization by acurrentpartner.

Of the studies reviewed above, three support the argument that CSA alone is not a sufficient predictorof adult victimization, whereas CPA alone is sometimes sufficient and the combination of both CSAand CPA consistently predicts adult victimization (Cloitre et al., 1996; Schaaf & McCanne, 1998; Wind& Silvern, 1992). These discrepant findings may be due to methodological differences. The definitionand assessment of adult victimization, including the type of adult victimization (e.g., sexual, physical,etc.) examined, may impact findings regarding the revictimization of CSA survivors. Many studies ofrevictimization in general focus on adult sexual victimization; however, bothWind and Silvern (1992)andSchaaf and McCanne (1998)defined revictimization in terms of adult sexualor physical abuse. It isunknown if either study would find a significant relationship between CSA and adult sexual victimizationwhen considered separately from physical assault, as didMerrill et al. (1999). Methods of assessing childand adult sexual victimization also may explain these findings. Asking respondents to self-identify assexually abused (Cloitre et al., 1996), or considering all unwanted sexual contact as adult sexual assault(Wind & Silvern, 1992) may affect findings regarding the revictimization of CSA survivors. Significantrates of revictimization might depend on more restrictive definitions; there tends to be a weaker, or evennonexistent, relationship between child and adult victimization when broader, more inclusive definitionsof CSA and/or adult victimization are used (Mayall & Gold, 1995; Merrill et al., 1999; Roodman & Clum,2001).

Although not all studies are in agreement, the few studies that examine the role of both CPA and CSAindicate the salience of physical, rather than sexual abuse, which emphasizes the importance of assessingmultiple forms of abuse in relation to revictimization. While there is evidence for the role of CPA inrevictimization, it is unclear if other forms of child maltreatment, such as emotional/psychological abuse,also would contribute to the risk for adult victimization. One study found that both CSA and psychologicalmaltreatment (psychological abuse and neglect) predicted later adult sexual victimization, although CPAdid not (Stermac, Reist, Addison, & Millar, 2002).

Family functioning and revictimization

The role of family functioning is also an important factor when considering the negative psychologicaleffects of child maltreatment as well as subsequent risk of adult victimization. There is evidence thatfamily dysfunction is associated with multiple forms of child maltreatment (Mullen, Martin, Anderson,Romans, & Herbison, 1996), although most studies have focused on the relationship between CSA andfamily functioning (seeDraucker, 1996, for a review). Regardless of whether maltreatment occurred withinthe home, poor family functioning may impact the level of support received following the initial CSAexperience, and lack of parental intervention may increase the risk of future abuse. Family dysfunction andpsychological problems in family members are more common in families of CSA survivors (Beitchman

1022 T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034

et al., 1991). Mullen, Martin, Anderson, Romans, and Herbison (1994)found that disorganized familysystems and high levels of marital distress were associated with CSA, andAlexander and Lupfer (1987)found that CSA survivors’ families were less cohesive, adaptable, and emotionally engaged.Long andJackson (1991)studied repeated childhood sexual victimization and found that, compared to the familiesof children victimized by only one perpetrator, the families of revictimized children were less cohesive,less expressive, more conflicted, and highly controlled. However, they found little difference betweenfamilies of nonvictims and victims of a single perpetrator, which could suggest that poor family functioningplaces victims at greater risk of revictimization.

A few studies have examined family environment in relation to child to adult sexual revictimization(as opposed to repeated sexual victimization during childhood).Mayall and Gold (1995)found that alack of maternal support among CSA survivors predicted sexual behavior, which in turn predicted sexualrevictimization. However,Fergusson et al. (1997)found the relationship between CSA and revictimizationto be independent of family background. Furthermore,Koverola, Proulx, Battle, and Hanna (1996)failedto find differences in family functioning between revictimized women and those with a history of CSAonly. These mixed findings make it difficult to determine the exact relationship between family functioningand revictimization. However, because some studies indicate differences in family functioning amongrevictimized individuals, and given that family functioning may play an important role in recovery fromthe effects of CSA and other forms of child abuse, it is important to assess as a possible protective factoror contributor to risk for adult victimization.

Most studies of revictimization have neglected to examine multiple forms of child abuse as well asfamily functioning. Given the frequency with which different forms of child abuse tend to overlap, itappears warranted to examine whether there is a cumulative impact of multiple types of child abuse inrelation to revictimization. Therefore, the purpose of this study was to examine childhood maltreatmentand family functioning as predictors of rape in adulthood. This study improves upon the current literatureby examining the role of perceived family environment as well as sexual, physical, and emotional abuse,including the impact of multiple forms of child abuse, in relation to adult sexual victimization.

Hypotheses.It was expected that all three forms of child abuse (sexual, physical, and emotional) wouldbe related to adult rape. It was also expected that child maltreatment would have a negative cumulativeimpact, such that experiencing multiple forms of child abuse would increase the risk of rape compared toexperiencing a single form of abuse or no abuse. The role of each form of child abuse (sexual, physical, andemotional) in the absence of any other form of child abuse was explored without specific hypotheses. A linkbetween family functioning (cohesion, expressiveness, and conflict) and all forms of victimization wasalso hypothesized. The family dysfunction perspective and the trauma perspective were both examinedin relation to rape during the study; we hypothesized that when all factors are considered simultaneously,child abuse, and not family dysfunction, would predict adult rape.

Method

Participants

Participants were 944 college women at a midsized public university in the Midwest recruited for a studyon “College Women’s Life Experiences.” Most participants participated as part of the course requirements

T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034 1023

for Introduction to Psychology; however, 23% (n= 217) received monetary payment for their participationin a related study. Participants were predominantly Caucasian (91.8%; African American = 2.1%; His-panic = .5%; Native American = .4%; Asian/Asian American = 1.9%; biracial = 1.9%; other = .6%) andcame from middle to upper class backgrounds (21.3% reported that they did not know their family’sincome; 58.6% reported an annual family income of $70,000 or higher; 11.2% $50,000–$69,000; 5.6%$30,000–$49,000; and 2.3% less than $29,000 a year). Ages ranged from 18 to 22 years (M= 18.88,SD= .98).

Measures

Life Experiences Questionnaire (LEQ).The LEQ (Long, 2002) is a retrospective, self-report instrumentthat assesses demographic information and childhood sexual experiences. CSA was screened with a seriesof eight questions assessing sexual experiences ranging in severity. If these are answered affirmatively,the participant is asked additional questions to obtain specific information regarding the identity of theperpetrator, age of perpetrator and victim, nature of sexual activity, and method of coercion. CSA wasdefined as contact abuse prior to age 17 and must have met at leastoneof the following criteria: (a)abuse perpetrated by a relative; (b) more than 5-year age difference between victim and perpetrator; or(c) if less than 5-year age difference between victim and perpetrator, threat or force was involved. TheLEQ has demonstrated reliability and validity in previous studies with college women (Messman-Moore& Long, 2000), which reported good internal consistency for the eight screening questions (Cronbach’salpha .89).Messman-Moore and Long (2000)reported kappas and percentage agreement on items relatedto the identity of the perpetrator (intrafamilial vs. extrafamilial: .86, 94%), duration of abuse (less than ormore than 1 year: 1.0, 100%), the nature of the sexual contact (penetration vs. no penetration: .91, 97%),and presence or absence of force (.39, 69%). Interclass correlation coefficients were also high for itemssuch as age of onset of abuse (.99), the age of the perpetrator (.96), and the age difference between victimand perpetrator (.95).

Childhood TraumaQuestionnaire (CTQ).The CTQ (Bernstein & Fink, 1998) is a 28-item inventory thatconsists of five scales assessing different forms of childhood maltreatment. Each scale consists of fiveitems, and responses are made on a 5-point Likert-type scale ranging from (1) Never True to (5) VeryOften True. Scales are scored by summing the responses within each subscale, producing a possible rangeof 5 to 25. Respondents’ scores are then classified into one of four categories of abuse severity basedon published recommendations. For the purposes of this study, these categories were collapsed and onlythose women reporting moderate to extreme abuse were considered abused. The CTQ has been shownto be internally consistent with a college student sample (Bernstein & Fink, 1998). For purposes of thisstudy, only two subscales were used; both demonstrated high internal consistency in the current sample:physical abuse Cronbach’s alpha = .81 and emotional abuse Cronbach’s alpha = .88.

Family Environment Scale (FES).The FES (Moos & Moos, 1986) was used to examine individuals’perception of their family characteristics and functioning. Respondents were instructed to answer ques-tions based upon the childhood environment in which they were raised. The FES consists of 90 true-falsequestions covering dimensions of personal growth, family relationship, and family organization. Tensubscales can be calculated with the FES; however, for purposes of this study, only three subscales wereexamined: conflict, cohesion, and expressiveness. The family conflict subscale reflects the amount of open

1024 T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034

aggression and anger that is typical of the family; the family cohesion subscale indicates the degree ofconcern and support that occurs between family members; the family expressiveness subscale describesthe extent to which family members feel free to act openly and to express their feelings.Moos and Moos(1986)reported 8-week test-retest reliabilities for each of the subscales, which are in the acceptable range(.61 to .86). The FES also has demonstrated internal consistency, with coefficients ranging from .61 to.78 (Moos & Moos, 1986). For the current study, internal consistency for the three subscales of interestwere as follows: cohesion,α = .61, expressiveness,α = .62, and conflict,α = .79.

Sexual Experiences Survey.A modified version of the 10-item Sexual Experiences Survey (SES;Koss &Gidcyz, 1985) assessed adult sexual assault and rape after age 17. The SES asks a series of yes/no questionsassessing whether specific types of sexual activities have been completed. The SES was modified to obtainmore specific information regarding the types of coercion experienced and the identity of the perpetrator(Messman-Moore & Long, 2000). The current version extends the number of questions from 10 to 18,increasing the number of questions regarding kissing and fondling, oral-genital contact, and unwantedsexual intercourse. The modified SES assesses four methods of coercion: arguments and pressure, misuseof authority, alcohol or drug use by the respondent, and threat or use of physical force. Phrasing ofquestions regarding alcohol and drug use were modified and modeled after those used byMuehlenhard,Powch, Phelps, and Giusti (1992), for example, “Have you ever had vaginal or anal intercourse when youdidn’t want to because you were incapable of giving consent or resisting due to alcohol or drugs?” Theset of 18 questions were asked twice, first to assess victimization by acquaintances or dating partners,and second to assess victimization by strangers. The modified SES demonstrated internal consistency inthis sample, coefficient alpha for sexual victimization by acquaintances .80, and by strangers .78. Rapewas defined specifically as unwanted oral, anal, or vaginal penetration obtained through threats or the useof force, or because the victim was unable to resist or consent due to alcohol or drug use.

Procedure

All data were anonymous and were collected in large group sessions using self-report questionnaires.Informed consent was obtained prior to the study, and participants’ rights were explained to the participantsby the experimenter and on separate consent forms. After completing the consent form, participants weregiven the surveys to complete. Consent forms and surveys were stored separately to ensure anonymityof data. Upon completing the session, participants were given either payment or research credit, andwere provided with contact information for the researchers as well as for local counseling and rape crisisservices. All procedures were approved by the university’s committee on human subjects in research.

Data analysis

Several sets of analyses were conducted to examine the relationships among child abuse, familyfunctioning, and adult rape. The association of different forms of child abuse and adult rape was examinedwith χ2 analyses, and the strength of these relationships was assessed with Phi correlation coefficients.Chi Square analyses were used to examine the relationship between the accumulation of child abuseand adult rape. The relationship between the three family functioning scales (i.e., conflict, cohesion andexpressiveness) and victimization was examined with multivariate analyses of variance (MANOVAs),with child abuse (and adult rape) as independent variables and FES subscales as dependent variables.

T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034 1025

Table 1Prevalence of child abuse

Type of abuse Prevalence (%) n Combinations of abuse Prevalencea n

Sexual abuse 8.9 83 All three forms of child abuse .8%/4.5% 7Physical abuse 4.2 40 Two forms of child abuse 3.4%/19.7% 31Emotional abuse 8.6 81 One form of child abuse 12.9%/75.8% 119Sexual abuse only 6.6 62 No child abuse 83.0% 768Physical abuse only 1.1 10Emotional abuse only 5.2 47Sexual and physical abuse .5 5Sexual and emotional abuse 1.0 9Physical & emotional abuse 1.8 17Sexual, physical and emotional abuse .8 7Any child abuse 17.0 157

a The first percentage reflects the prevalence of combination of abuse within the entire sample (N= 925), the second percentagereflects the prevalence of combination of abuse among only those women reporting child abuse (n= 157).

The predictive utility of all child abuse variables and family functioning variables in relation to adultrape was examined with two logistic regression analyses, with the three forms of child abuse and threeFES subscales as independent variables. The moderating impact of family functioning on child abusewas examined via logistic regression.

Results

Child abuse and rape

Prevalence rates of individual and combined forms of child abuse are listed inTable 1. Of 925 par-ticipants with complete data, 17% reported a history of any form of child abuse (sexual, physical, oremotional), and 15% reported adult rape. It is noteworthy that CPA rarely occurred in isolation. Therelationship between adult rape and each form of child abuse was examined (regardless of overlap acrosstypes of child abuse) withχ2 analyses, and the strength of these associations was assessed with Phicorrelation coefficients. Most forms of child abuse were related to adult rape with the exception of CPA(seeTable 2), although the magnitude of these relationships was relatively small (Φ ranged from .08 to.12).

Because of the overlap between the different types of abuse, the cumulative impact of child abuse onadult rape was examined. Women were categorized into one of four groups reflecting experience of childabuse: all three types of child abuse, two forms of child abuse, one form of child abuse and no child abuse(seeTable 1). There was a significant effect for the accumulation of abuse,χ2 (3, n= 900) = 18.27,p <.001 (seeFigure 1). Forty-three percent of women reporting all three forms of abuse, 35.5% of womenreporting two of the three forms of abuse, and 20.9% of women reporting one form of abuse were rapedcompared to 13.5% of women without a history of child abuse. Follow-upχ2 analyses indicated thatwomen reporting three forms of abuse, two forms of abuse, or one form of abuse were more likely toreport adult rape compared to women without a history of child abuse. No other groups were significantlydifferent atp < .05.

1026 T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034

Table 2Comparisons of the percentage of child abuse victims and nonvictims who reported adult rape

Type of abuse Percent rape victims Chi Square Phi coefficientp

Childhood victim (%) Childhood nonvictim (%)

Any child abuse 24.5 13.5 11.91 .12 .001Child sexual abuse 26.6 14.3 8.45 .10 .004Child physical abuse 22.5 15.0 1.64 .04 .20Child emotional abuse 30.9 13.8 16.46 .13 .001CSA only 20.7 14.9 1.39 .04 .24CPA only .0 15.0 1.77 −.05 .18CEA only 25.5 13.8 4.92 .08 .03

Family environment

The relationship between different types of child abuse and family environment was examined withmultivariate analysis of variance (MANOVA). Four separate analyses were conducted for each of the fourtypes of victimization: child sexual abuse, child physical abuse, child emotional abuse and adult rape,with the three subscales of the FES (i.e., cohesion, expressiveness and conflict) as dependent variables.It was hypothesized that women who experienced abuse as children would also have experienced familyenvironments lower in cohesion and expressiveness and higher in conflict than women who were notabused in childhood. The relationship between adult rape and family functioning was explored withoutany a priori hypotheses.Table 3provides the means and standard deviations for each family factor forboth abused and non-abused women and Wilks’Λ for each analysis. For all analyses, univariate ANOVAsare statistically significant unless noted. Compared to nonvictims, CPA and CEA victims reported poorerfamily functioning (lower levels of cohesion and expressiveness and higher levels of conflict); however,CSA was not related to family functioning. Adult rape victims also reported lower levels of cohesion andhigher levels of family conflict compared to nonvictims.

Figure 1. Accumulation of different types of child abuse and percentage of women reporting adult rape.

T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034 1027

Table 3Comparison of levels of family cohesion, emotional expressiveness, and conflict based on victimization history

Cohesion Expressiveness Conflict F Wilks’ Λ

Victims Nonvictims Victims Nonvictims Victims Nonvictims

Child sexual abuse 6.04 (1.97) 6.44 (1.84) 4.84 (2.40) 5.29 (2.13) 3.77 (2.36) 3.16 (2.40) 2.06 .993ns

Child physical abuse 4.54 (2.32) 6.48 (1.79) 3.05 (2.16) 5.33 (2.11) 6.33 (2.26) 3.09 (2.31) 30.57 .909***

Child emotional abuse 4.11 (2.23) 6.62 (1.67) 3.26 (1.92) 5.42 (2.09) 6.41 (1.99) 2.91 (2.20) 82.89 .787***

Adult rape 5.76 (2.11) 6.50 (1.81) 4.94 (2.19) 5.28a (2.17) 3.87 (2.63) 3.12 (2.34) 6.73 .978***

Note:Multivariate analysis of variance analyses conducted for four forms of victimization. All univariate tests significant atp<.001 unless noted. Higher scores on the cohesion and expressiveness scales of the FES indicate more adaptive family functioning;lower scores on the conflict scale of the FES indicates more adaptive family functioning.

a Univariate test,p= .09ns p= .10.

∗∗∗ p < .001

The cumulative impact of multiple types of child abuse was also examined in relation to familyfunctioning. It was hypothesized that women reporting multiple forms of child abuse (two or moreforms) would report more family dysfunction than those women reporting a single form of abuse or noabuse. MANOVA was used to examine differences between the four groups: all three forms of childabuse, two forms of child abuse, one form of child abuse and no child abuse in relation to the three FESsubscales, family cohesion (COH), family expressiveness (EXP), and family conflict (CON). Results weresignificant; Wilks’Λ = .830,F (9, 2195.38) = 19.35,p < .001 (seeTable 4). For all three FES subscales,women reporting two or all three forms of child abuse reported greater family dysfunction compared towomen reporting one form or no forms of child abuse. Women reporting two forms of abuse did notdiffer from those that reported all three; women reporting two forms of abuse reported poorer familyfunctioning compared to women reporting only one type of child abuse, and women reporting only onetype of child abuse reported greater family dysfunction compared to nonvictims.

Child abuse and family environment

Logistic regression analysis was used to examine multiple predictors of adult rape. Although conditionsof multicollinearity existed, partialling procedures using step-wise regression were not used becauseseparating different types of child abuse from one another and from the family context appeared to threaten

Table 4The accumulation of different types of child abuse in relation to family cohesion, emotional expressiveness, and conflict

Three types ofabuse (n= 6)

Two types ofabuse (n= 31)

One type ofabuse (n= 119)

No abuse (n= 752) UnivariateF p

COH 3.50a,e (1.87) 4.68b,d (2.32) 5.38c,d,e (2.28) 6.66a,b,c (1.64) 34.11 .001EXP 2.00a,e (1.41) 3.19b,d (2.07) 4.52c,d,e (2.32) 5.48a,b,c (2.04) 22.88 .001CON 6.83a,e (1.94) 6.48b,d (2.20) 4.48c,d,e (2.47) 2.85a,b,c (2.19) 46.70 .001

Note:Mean (SD); groups with the same superscripts are significantly different atp < .001; COH, family cohesion; EXP, familyexpressiveness; CON, family conflict. Higher levels of cohesion and expressiveness are more adaptive; lower levels of conflictare more adaptive.

1028 T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034

Table 5Logistic regression analysis predicting adult rape from childhood abuse and family factors

Predictor variable B SE Wald statistic p Odds ratio 95% CI

Model �2(6) = 26.91 .001CSA .63 .29 �2(1) = 4.80 .03 1.9 1.07–3.30CPAa −.41 .47 �2(1) = .78 .38 1.4 .61–3.33CEA .59 .33 �2(1) = 3.30 .07 1.8 .95–3.44COHa −.14 .06 �2(1) = 5.68 .02 1.1 1.02–1.30EXP .02 .05 �2(1) = .23 .63 1.0 .93–1.13CON .04 .05 �2(1) = .60 .44 1.0 .94–1.14

CSA, child sexual abuse; CPA, child physical abuse; CEA, child emotional/psychological abuse; COH, family cohesion; EXP,family expressiveness; CON, family conflict.

a The inverse of odds ratios and confidence intervals were calculated for variables negatively correlated with adult rape toreflect the increased risk associated with the absence of these factors.

construct validity (seeBriere, 1988). To determine the strongest predictors of adult rape, all variables wereentered on one step, an approach more conservative than the step-wise method. Six variables were enteredon step one: CSA, CPA, CEA, COH, EXP, and CON. Two factors emerged as significant predictors ofadult rape: CSA and COH (seeTable 5). After controlling for other forms of child abuse and familyfunctioning, CSA victims were 1.9 times more likely than women who did not report CSA to report anadult rape experience. Women reporting lower levels of family cohesion were more likely to report adultrape. Results of logistic regression equations using stepwise procedures were identical.

An additional logistic regression analysis was conducted to examine whether certain types of childabuse in the context of family functioning impacts risk of rape to a greater extent than either child abusevariables or family functioning variables alone. For this analysis, all three forms of child abuse and thethree family variables were entered in the first block using stepwise procedures; the interactions betweenthese variables were then entered in the second block (seeTable 6). In the first block, there were twosignificant main effects, CSA and COH. In the second block, there was one significant main effect, CSA,and two significant interactions, CEA× COH and CEA× EXP.

Post hoc analysis.In order to interpret the two significant interactions, four additional logistic regressionanalyses were conducted, examining cohesion (COH) and expressiveness (EXP) at each level of emo-tional abuse (presence = 1 and absence = 0). Results of these analyses (seeTable 6) indicated that familycohesion was not a significant predictor of adult rape when emotional abuse was present. However, whenemotional abuse was absent, risk of rape increased significantly as family cohesion decreased. Further-more, emotional expressiveness within the family was not a significant predictor of adult rape whenemotional abuse was absent, however the risk for rape increased significantly as family expressivenessdecreased when emotional abuse was also present.

Discussion

Given the state of the current literature, it is difficult to determine whether vulnerability for revictim-ization occurs due to effects common to all forms of child abuse, due to specific effects of CSA (e.g.,

T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034 1029

Table 6Stepwise logistic regression analysis testing moderation effects of family functioning on child abuse variables predicting adultrape

Predictor variable B SE Wald statistic p Odds ratio 95% CI

Block 1 model �2(2) = 22.19 .001CSA .65 .28 .02 1.9 1.10–3.30COHa −.19 .05 .001 1.3 1.10–1.32

Block 2 model �2(4) = 31.77 .001CSA .58 .29 .04 1.8 1.02–3.12COHa −.10 .06 .08 1.1 .99–1.23CEA × COH .29 .09 .002 1.3 1.12–1.60CEA × EXPa −.20 .09 .03 1.3 1.02–1.47

Post hoc analysesCEA × COH

Emotional abuse = 1 COH −.02 .11 �2(1) = .04 .85 1.0 .82–1.27Emotional abuse = 0 COH −.18 .05 �2(1) = 10.56 .001 1.2 1.07–1.32

CEA × EXPEmotional abuse = 1 EXP −.32 .15 �2(1) = 4.60 .05 1.4 1.03–1.85Emotional abuse = 0 EXP .00 .05 �2(1) = .00 .98 1.0 .91–1.10

CSA, child sexual abuse; CPA, child physical abuse; CEA, child emotional/psychological abuse; COH, family cohesion; EXP,family expressiveness; CON, family conflict. Block 1 variables: CSA, CPA, CEA, COH, EXP, CON entered forward stepwise.Block 2 variables: CSA× COH, CSA× EXP, CSA× CON, CPA× COH, CPA× EXP, CPA× CON, CEA× COH, CEA×EXP, CEA× CON entered forward stepwise.

a The inverse of odds ratios and confidence intervals were calculated for variables negatively correlated with adult rape toreflect the actual increased risk associated with the absence of these factors.

traumatic sexualization,Finkelhor & Browne, 1985), or due to a cumulative or interactive impact of mul-tiple types of child abuse. Unfortunately, most studies examining sexual revictimization focus on CSAto the exclusion of other forms of child abuse or family functioning.

As in previous research (for reviews seeArata, 2002; Breitenbecher, 2001; Messman & Long, 1996), inthe present investigation women reporting CSA were almost twice as likely as women without histories ofCSA to report rape during adulthood. CSA was a predictor of adult victimization even when consideringthe impact of the family environment and was the only form of child abuse to predict rape independentof family functioning. Emotional abuse was also a significant predictor of adult rape. The present studysuggests that emotional abuse by itself is an important experience in relation to increased vulnerabilityfor adult victimization, and that family cohesion and emotional expressiveness increase the risk of adultrape only in relation to emotional abuse. Fewer studies have examined the impact of psychologicalforms of abuse or neglect. The present findings are consistent with another study in which psychologicalmaltreatment (psychological abuse or neglect) predicted adult sexual victimization experiences amongwomen in a clinical sample (Stermac et al., 2002). Contrary to previous findings (e.g.,Cloitre et al.,1996), CPA did not emerge as a risk factor for later rape. The present findings suggest that CPA is not asstrong a predictor ofsexualvictimization as other forms of child abuse unless it occurs in combinationwith other forms of abuse. Perhaps CPA is a stronger predictor of physical rather than sexual abuse inadulthood (Schaaf & McCanne, 1998; Wind & Silvern, 1992), or it may be that dysfunctional family

1030 T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034

dynamics mediate the relationship between CPA and revictimization found in other studies. However, thefailure of CPA to predict sexual victimization has been found in other studies (Stermac et al.).

The present findings support the idea that experiencing multiple forms of abuse during childhoodincreases the risk of adult rape and suggest that in some cases vulnerability for revictimization maybe related to the cumulative impact of child abuse in general, rather than CSA specifically. Forty-twopercent of women reporting all three forms of child abuse and 35% of women reporting two of the threeforms of child abuse were raped as adults, rates significantly higher than those without child abuse, andwomen reporting two forms of abuse were somewhat more likely to report rape compared to womenreporting only one form of abuse. These findings are consistent with another study that noted a strongassociation between CSA and other forms of traumatic childhood experiences (Banyard et al., 2001), andmay lend more support to the argument that interpersonal violence, including violence in childhood, hasa cumulative impact (Follette, Polusny, Bechtle, & Naugle, 1996; Messman-Moore, Long, & Siegfried,2000).

The interactive effects of child abuse variables and family functioning variables on adult rape werealso explored. Although there is evidence that CSA victims come from dysfunctional families (Draucker,1996), the two variables were not correlated in the present study. The fact that CSA was the only form ofchild abuse not associated with family dysfunction may be because CSA does not always occur withinthe family. Moreover, the link between family dysfunction and the other forms of child abuse (CPA andCEA) was expected given that (a) in most cases these forms of maltreatment occur within the family(see alsoMullen et al., 1996), and (b) in the present study CPA and CEA were assessed as behaviors bya caregiver or other family member. Interestingly, family functioning impacted the risk of rape only inrelation to CEA. When emotional abuse was not present, the risk of rape increased as levels of cohesiondecreased; however, low levels of emotional expressiveness only affected risk when CEA was present.This suggests that family cohesion serves as a protective factor only when child abuse, namely CEA,is absent, and implies that CEA alone increases the risk of rape to the extent that cohesion does notaffect vulnerability when CEA is present. However, family expressiveness only affects the risk of rapewhen CEA is present, suggesting that family environments in which emotional expression and opencommunication is discouraged increase the vulnerability for adult rape only when emotional abuse isinvolved as well.

These findings call into question the conclusions of many studies that examine the impact of anyone form of child abuse, regardless of type. It appears critical to assess for multiple forms of childhoodmaltreatment as well as other types of aversive childhood experiences (Banyard et al., 2001; Higgins &McCabe, 2001). It is also important to recognize that certain forms of child abuse, particularly CPA, arelikely to exist in relation to other forms of child abuse, whereas other forms of abuse may be less likely toco-occur (Higgins & McCabe, 2001). Our study also raises more questions regarding the appropriatenessof attempting to separate different types of child abuse from one another if they co-occur, as well asseparating the impact of child abuse and family functioning. This is an important issue to address becauseresults in the current investigation differed depending on whether abuse was considered isolated from otherforms. These differences in findings may simply reflect lower statistical power resulting from isolatingabuse variables, but it may also speak to the importance of multiple forms of abuse and family functioningas risk factors for adulthood victimization. In the future, researchers should strongly consider conceptualand theoretical models of child abuse and family dysfunction when designing studies and selecting astatistical approach to data analysis (e.g., choosing to isolate the impact of different forms of child abuseor the impact of child abuse and family functioning).

T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034 1031

The majority of current findings are consistent with previous research. However, these findings donot help clarify the mixed literature regarding the impact of CPA on adult rape (Cloitre et al., 1996;Desai et al., 2002; Schaaf & McCanne, 1998; Stermac et al., 2002). Although the rates of child abusein this sample are comparable to other similar studies (for review seeHiggins & McCabe, 2001), re-sults should be interpreted with caution given the small number of women reporting CPA, particularlythose who experienced CPA alone, which confounded the impact of CPA with the cumulative impactof abuse. Although there were significant relationships found between child abuse and adult sexual vic-timization, the magnitude of these associations, although significant, were rather small, which suggeststhat other factors not measured in the present study likely impact the risk of rape. Future research mustcontinue to examine potential mediators of this relationship, which may include psychological sequelae ofprior trauma, coping strategies, social support, and intervention following abuse disclosure, among otherfactors.

Although CSA and CEA both increase the risk of victimization, the impact may occur via differentmechanisms. For example, the effects of traumatic sexualization (Finkelhor & Browne, 1985) may impactrevictimization via risky sexual behavior, as CSA survivors may come to believe that their only value liesin their sexuality or that they can gain a sense of power or control via their sexual behavior. CEA mayincrease vulnerability indirectly by impacting self-esteem or assertiveness in relationships, especiallywhen such abuse is experienced in a family environment that discourages direct expression of feelings. Itmay also be that both forms of child abuse as well as family functioning contribute to vulnerability becausegirls and young women are seeking validation and love in potentially exploitative relationships. Finally,perhaps dysfunctional family dynamics impact vulnerability for victimization by impacting women’sunderstanding of and expectations for interpersonal and intimate relationships.

The present findings should be considered in the context of some limitations. First, the sample wascomprised of college women, who were predominantly Caucasian, young, from middle to upper classbackgrounds, and with generally good psychological functioning, which may affect results and gener-alization of findings. It is unknown if the childhood experiences found important in relation to latervictimization will continue to have a strong impact as these women mature. More research should beconducted with larger, more diverse samples and ideally with prospective or longitudinal designs. Sec-ond, although the overall sample was large, some groups, particularly those reporting all three types ofchild abuse, were very small. Research aiming to dissect the relative contribution of different types andcombinations of child abuse will necessitate very large community samples. The small sample prohibitedexamining abuse-specific factors that may impact revictimization, including identity of the perpetratorin the case of CSA. Further, use of a screener to assess CEA and CPA did not allow for examination ofthe extent or severity of physical or emotional abuse reported. The FES measures only certain aspects offamily functioning, and less is known about the impact of parental substance use, psychological disorder,divorce, or witnessing parental conflict or domestic violence. Finally, the retrospective manner in whichchildhood experiences were assessed may impact these results. It may be that remembering negativechildhood experiences influences memories of adult experiences so that individuals with histories ofchild abuse or dysfunctional families are more likely to label negative adult experiences as victimiza-tion. It is also possible that the experience of adult rape may influence or prime memories of childhoodexperiences. Because the measures assessing childhood abuse were always included before the measureassessing adult victimization, it is unlikely that recalling adult experiences influenced recall for childhoodexperiences; however, the reverse remains a possibility.

1032 T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034

Conclusions

The present investigation suggests that there exist multiple pathways that lead to vulnerability forlater sexual victimization. As in previous studies, CSA appears to play an important role in adult sexualvictimization. This study highlights the potential importance of CEA, a form of childhood abuse that hasnot been examined extensively in relation to adult rape. The present study also helps to clarify the role offamily functioning in relation to risk for adult rape. Low levels of family cohesion are a significant riskfactor for rape when emotional abuse has not occurred, and low levels of emotional expressiveness withinthe family increase the risk of rape when emotional abuse is also present. Without longitudinal studies thatexamine these factors prior to and during abuse, it is difficult to determine whether abuse preceded familydysfunction or vice versa. Regardless, it is probably not ecologically valid or even useful to separate theimpact of child abuse and family dysfunction. Rather, it is important to recognize that within childhoodexperience there exist multiple pathways to vulnerability for adult rape experiences, and future researchshould strive not only to identify those types of adverse childhood experiences that predispose women toincreased risk of victimization, but also to identify and examine the underlying mechanisms that mightexplain these relationships.

Acknowledgments

The authors would like to acknowledge numerous research assistants, without whom this project wouldnot be possible: Stephanie Calmes, Diana Hickey, Lori Koelsch, Ryan McLaughlin, Sara Nelson, RobPace, Allison Scheer, Jaclyn Tooley, and Kyleigh Turner. We are grateful to David DiLillo, Rose Ward,and anonymous reviewers for their feedback on earlier versions of this manuscript.

References

Alexander, P. C., & Lupfer, S. L. (1987). Family characteristics and long-term consequences associated with sexual abuse.Archives of Sexual Behavior, 16, 235–245.

Arata, C. M. (2002). Child sexual abuse and revictimization.Clinical Psychology: Science& Practice, 9(2), 135–164.Banyard, V. L., Williams, L. M., & Siegel, J. A. (2001). The long-term mental health consequences of child sexual abuse:

An exploratory study of the impact of multiple traumas in a sample of women.Journal of Traumatic Stress, 14(4), 697–715.

Beitchman, J. E., Zucker, K. J., Hood, J. E., DaCosta, G. A., & Akman, D. (1991). A review of the short-term effects of childsexual abuse.Child Abuse& Neglect, 15, 537–556.

Bernstein, D. P., & Fink, L. (1998).Childhood TraumaQuestionnairemanual. San Antonio, TX: The Psychological Corporation.Breitenbecher, K. H. (2001). Sexual revictimization among women: A review of the literature focusing on empirical investigations.Aggression& Violent Behavior, 6(4), 415–432.

Briere, J. (1988). Controlling for family variables in abuse effects research.Journal of Interpersonal Violence, 3(1), 80–89.

Briere, J., & Elliott, D. M. (1993). Sexual abuse, family environment and psychological symptoms: On the validity of statisticalcontrol.Journal of Consulting and Clinical Psychology, 61(2), 284–288.

Cloitre, M., Tardiff, K., Marzuk, P. M., Leon, A. C., & Portera, L. (1996). Childhood abuse and subsequent sexual assault amongfemale inpatients.Journal of Traumatic Stress, 9(3), 473–482.

Desai, S., Arias, I., Thompson, M. P., & Basile, K. C. (2002). Childhood victimization and subsequent adult revictimizationassessed in a nationally representative sample of women and men.Violence& Victims, 17(6), 639–653.

T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034 1033

Draucker, C. B. (1996). Family-of-origin variables and adult female survivors of childhood sexual abuse: A review of the research.Journal of Child Sexual Abuse, 5(4), 35–63.

Fergusson, D. M., Horwood, L. J., & Lynskey, M. T. (1997). Childhood sexual abuse, adolescent sexual behaviors and sexualrevictimization.Child Abuse& Neglect, 21(8), 789–802.

Finkelhor, D., & Browne, A. (1985). The traumatic impact of child sexual abuse: A conceptualization.American Journal ofOrthopsychiatry, 55(4), 530–541.

Follette, V. M., Polusny, M. A., Bechtle, A. E., & Naugle, A. E. (1996). Cumulative trauma: The impact of child sexual abuse,adult sexual assault, and spouse abuse.Journal of Traumatic Stress, 9, 25–35.

Higgins, D. J., & McCabe, M. P. (2001). Multiple forms of child abuse and neglect: Adult retrospective reports.Aggression andViolent Behavior, 6, 547–578.

Koss, M. P., & Gidcyz, C. A. (1985). Sexual experiences survey: Reliability and validity.Journal of Consulting and ClinicalPsychology, 53, 422–423.

Koverola, C., Proulx, J., Battle, P., & Hanna, C. (1996). Family functioning as predictors of distress in revictimized sexual abusesurvivors.Journal of Interpersonal Violence, 11(2), 263–280.

Long, P. J. (2002).Psychometric validation of the Life Experiences Questionnaire (LEQ). Unpublished manuscript, OklahomaState University.

Long, P. J., & Jackson, J. L. (1991). Children sexually abused by multiple perpetrators.Journal of Interpersonal Violence, 6,147–159.

Mayall, A., & Gold, S. R. (1995). Definitional issues and mediating variables in the sexual revictimization of women sexuallyabused as children.Journal of Interpersonal Violence, 10(1), 26–42.

Merrill, L. L., Newell, C. E., Thomsen, C. J., Gold, S. R., Milner, J. S., Koss, M. P., & Rosswork, S. G. (1999). Child-hood abuse and sexual revictimization in a female Navy recruit sample.Journal of Traumatic Stress, 12(2), 211–225.

Messman, T. L., & Long, P. J. (1996). Child sexual abuse and its relationship to revictimization in adult women: A review.Clinical Psychology Review, 16(5), 397–420.

Messman-Moore, T. L., & Long, P. J. (2000). Child sexual abuse and revictimization in the form of adult sexualabuse, adult physical abuse, and adult psychological maltreatment.Journal of Interpersonal Violence, 15(5), 489–502.

Messman-Moore, T. L., Long, P. J., & Siegfried, N. J. (2000). The revictimization of child sexual abuse survivors: An examinationof the adjustment of college women with child sexual abuse, adult sexual abuse and adult physical abuse.Child Maltreatment,5, 18–27.

Moos, R. H., & Moos, B. S. (1986).Family Environment Scale manual(2nd ed.). Palo Alto, CA: Consulting Psychologists Press.Muehlenhard, C. L., Powch, I. G., Phelps, J. L., & Giusti, L. M. (1992). Definitions of rape: Scientific and political implications.Journal of Social Issues, 48, 23–44.

Mullen, P. E., Martin, J. L., Anderson, J. C., Romans, S. E., & Herbison, G. P. (1994). The effect of child sexual abuse on social,interpersonal and sexual function in adult life.British Journal of Psychiatry, 165, 35–47.

Mullen, P. E., Martin, J. L., Anderson, J. C., Romans, S. E., & Herbison, G. P. (1996). The long-term impact of the physical,emotional, and sexual abuse of children: A community study.Child Abuse& Neglect, 20, 7–21.

Nash, M. R., Hulsey, T. L., Sexton, M. C., Harralson, T. L., & Lambert, W. (1993). Long-term sequelae of childhood sexualabuse: Perceived family environment, psychopathology and dissociation.Journal of Consulting and Clinical Psychology,61(2), 276–283.

Polusny, M. A., & Follette, V. M. (1995). Long-term correlates of child sexual abuse: Theory and review of the empiricalliterature.Applied& Preventive Psychology, 4, 148–166.

Roodman, A. A., & Clum, G. A. (2001). Revictimization rates and method variance: A meta-analysis.Clinical PsychologyReview, 21(2), 183–204.

Schaaf, K. K., & McCanne, T. R. (1998). Relationship of childhood sexual, physical and combined sexual and physical abuseto adult victimization and posttraumatic stress disorder.Child Abuse& Neglect, 22(11), 1119–1133.

Stermac, L., Reist, D., Addison, M., & Millar, G. M. (2002). Childhood risk factors for women’s sexual victimization.Journalof Interpersonal Violence, 17(6), 647–670.

Wind, T. W., & Silvern, L. (1992). Type and extent of child abuse as predictors of adult functioning.Journal of Family Violence,7(4), 261–281.

1034 T.L. Messman-Moore, A.L. Brown / Child Abuse & Neglect 28 (2004) 1019–1034

Resume

Objectif: On a examine les mauvais traitements et le fonctionnement familial en tant que predicteurs du viola l’ageadulte aupres d’unechantillon de 925etudiantes de niveau universitaire.Methode:Les renseignements ontete obtenus par le biais d’un questionnaire retrospectif auto-administre. L’abussexuel subi dans l’enfance (CSA) aeteevalue avec le “Life Experience Questionnaire”, l’abus psychologique (CEA)et l’abus physique (CPA) ontete evalues avec le “Childhood Trauma Questionnaire”. Le fonctionnement familiala ete evalue avec le “Family Environment Scale” et le viol chez l’adultea l’aide de l’enquete “Sexual ExperiencesSurvey”.Resultats:A peu pres 17% des femmes ont rapporte avoir subi dans leur passe une forme d’abus, selon la distributionsuivante: 8.9% CSA, 4.2% CPA, et 8.6% CEA. 15% des participantes ont rapporte avoir subi un viol depuis l’agede 17 ans. CSA et CEA, mais pas CPA ont montre etre en relation avec le viola l’age adulte. L’experience subiede formes multiples de mauvais traitements a augmente le risque de subir un viola l’age adulte. Les analyses parregression logistique ont indique que CSA pouvait predire le viol de l’adulte (OR = 1.9). De plus, un bas niveaude cohesion familiale predit le viol de l’adulte seulement en l’absence de CEA (OR = 1.2), et de bas niveauxd’expression desemotions dans la famille predisent le viol de l’adulte en presence de CEA (OR = 1.4).Conclusions:Les resultats montrent qu’il est necessaire de considerer les formes multiples de mauvais traitementsenvers les enfants tout comme le fonctionnement familial commeetant en relation avec le risque de victimisationchez l’adulte. Les resultats suggerent que l’abus sexuel constitue un facteur de risque de victimisation chez l’adulteindependant du fonctionnement familial et que le dysfonctionnement familial exerce un impact sur le viol uniquements’il est en relation avec des abus psychologiques.

Resumen

Objetivo: Se examinaron el maltrato infantil y el funcionamiento familiar como predictores de la violacion enadultos en una muestra de 925 mujeres universitarias.Metodo: Se obtuvo informacion de cuestionarios de auto reporte retrospectivos. El abuso sexual infantil (CSA)fue evaluado con el Cuestionario de Experiencias de la Vida, el abuso emocional infantil (CEA) y el abuso fısico(CPA) fueron evaluados con el Cuestionario de Trauma Infantil, y el funcionamiento familiar fueron evaluados conla Escala de Contexto Familiar, y la violacion en la adultez fue evaluada con la Encuesta de Experiencias Sexuales.Resultados:Aproximadamente el 17 % de las mujeres reportaron una historia de alguna forma de abuso infantil:8.9% reportaron CSA, 4.2% reportaron CPA, y el 8.6% reporto CEA. Quince por ciento de los participantesreportaron una historia de violacion desde la edad de 17. El CSA y el CEA, pero no el CPA, fueron relacionadoscon la violacion en la adultez. El sufrir multiples formas de abuso infantil aumento el riesgo de la violacion en laadultez. Los analisis de regresion logıstica indicaron que el CSA predecıa la violacion en la adultez (OR = 1.2), ybajos niveles de expresividad emocional en la familia predecıa la violacion en la adultez solo en presencia del CEA(OR = 1.4).Conclusiones:Los hallazgos resaltan la necesidad de considerar las multiples formas de abuso infantil ası como elfuncionamiento familiar en relacion con el riesgo de victimizacion sexual en la adultez. Los hallazgos sugieren queel CSA es un factor de riesgo para la victimizacion adulta independiente del funcionamiento familiar, y la disfuncionfamiliar ejerce un impacto sobre la violacion solo en relacion al CEA.