Undergraduate Students' Attributions of Depicted Adult–Adolescent and Adolescent–Adolescent Sexual Interactions

PLEASE SCROLL DOWN FOR ARTICLE

This article was downloaded by: [Sherrill, Andrew]On: 24 March 2011Access details: Access Details: [subscription number 935355857]Publisher RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Child Sexual AbusePublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t792303988

Undergraduate Students' Attributions of Depicted Adult-Adolescent andAdolescent-Adolescent Sexual InteractionsAndrew Sherrilla; Kimberly Renka; Valerie K. Simsa; Anne Culpa

a University of Central Florida, Orlando, Florida, USA

Online publication date: 24 March 2011

To cite this Article Sherrill, Andrew , Renk, Kimberly , Sims, Valerie K. and Culp, Anne(2011) 'Undergraduate Students'Attributions of Depicted Adult-Adolescent and Adolescent-Adolescent Sexual Interactions', Journal of Child SexualAbuse, 20: 2, 157 — 181To link to this Article: DOI: 10.1080/10538712.2011.554342URL: http://dx.doi.org/10.1080/10538712.2011.554342

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial orsystematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply ordistribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss,actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directlyor indirectly in connection with or arising out of the use of this material.

Journal of Child Sexual Abuse, 20:157–181, 2011Copyright © Taylor & Francis Group, LLCISSN: 1053-8712 print/1547-0679 onlineDOI: 10.1080/10538712.2011.554342

CLINICAL RESEARCH

Undergraduate Students’ Attributions ofDepicted Adult–Adolescent and

Adolescent–Adolescent Sexual Interactions

ANDREW SHERRILL, KIMBERLY RENK, VALERIE K. SIMS,and ANNE CULP

University of Central Florida, Orlando, Florida, USA

The grayest areas of defining child sexual abuse appear to involvethe age and sex of the individuals involved, resulting in a poten-tial for different attributions regarding child sexual abuse acrossindividuals. As a result, this study examines the responses of 262male and female college student participants after viewing a seriesof hypothetical sexual abuse vignettes that depicted a 15-year-oldvictim that neither resisted nor encouraged the advances of a 15-,25-, or 35-year-old perpetrator’s actions. Gender roles and sexualattitudes were examined as potentially important covariates. Usinga series of analyses of covariance, female participants gave morepro-victim ratings than male participants, and younger perpetra-tors were viewed less negatively than older perpetrators. Genderroles and sexual attitudes served as significant covariates. Thesefindings emphasized the need to educate individuals about childsexual abuse and unwanted sexual contact involving individualsunder the age of consent.

KEYWORDS sexual abuse, perpetrator, gender role, sexualattitudes, ratings

Submitted 12 May 2010; revised 25 July 2010; accepted 28 July 2010.This manuscript is based on the honors thesis of the first author under the direction of

the second author and a committee composed of the third and fourth authors.Address correspondence to Kimberly Renk, Department of Psychology, University of

Central Florida, P.O. Box 161390, Orlando, Florida 32816. E-mail: [email protected].

157

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

158 A. Sherrill et al.

Child sexual abuse (CSA) has received much attention from researchers sincethe publication of Finkelhor’s (1979) book Sexually Victimized Children.Three decades after the publication of this book, however, defining CSAremains an ongoing task for researchers, lawmakers, and society in general.The grayest areas of defining CSA seem to involve the age and sex of theindividuals involved, resulting in different attributions for CSA across indi-viduals. To examine these attributions further, the current study investigatedhow contemporary undergraduate students perceive CSA by controlling forthe age and sex of individuals depicted in CSA vignettes. Such a study maybe particularly important, as researchers suggest that understanding pub-lic attitudes about CSA may be related significantly to the prevalence ofreporting incidences, attitudes toward community notification laws, the pros-ecution of perpetrators, and the provision of clinical services (Broussard,Wagner, & Kazelskis, 1991; Maynard & Wiederman, 1997; Pollard, 1992;Redlich, 2001; Waterman & Foss-Goodman, 1984). The clinical implicationsfor understanding attitudes about CSA also are particularly noteworthy, assupportive attitudes toward victims may be related to whether individualsidentify themselves as having been victimized (e.g., Lippert, Cross, Jones, &Walsh, 2009) and subsequently present for treatment services.

Stereotypically, CSA is thought to include an adult male fondling aprepubescent female. Previous research consistently suggests that femalesaccount for 83% of sexual abuse victims (Jones & McCurdy, 1992). As aresult, any incident of CSA that departs from such a description is less likelyto be subject to legal action (Williams & Farrell, 1990). To explore the bound-aries of CSA definitions, many researchers have individuals rate depictedvignettes in which variables are manipulated and controlled. Findings ofthese studies suggest that the age of the depicted perpetrator may be themost influential variable in defining CSA (Finkelhor & Redfield, 1984). In par-ticular, vignettes with perpetrators who are 25 years of age or older tend tobe rated as definitely sexual abuse regardless of the other variables that maybe manipulated. In contrast, those vignettes that depict younger perpetratorsare rated as less abusive (Finkelhor & Redfield, 1984). Thus, perpetrator ageappears to be particularly important in defining CSA interactions. In addition,the age and the sex of the victim, as well as other descriptive variables, mayprove collectively important in predicting individuals’ attributions about CSA.

THE ROLE OF AGE IN CSA INTERACTIONS

Given these findings, the approximate age of 25 years may be particularlyimportant for furthering the understanding of attributions of CSA. In fact,research shows that the mean age of the perpetrators of female victims is31.7 years of age and of male victims is 26.9 years of age (Finkelhor, 1979).More current research examining the typical age of perpetrators of CSA

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

Undergraduate Students’ Attributions 159

suggests relatively similar findings. For example, Cullen, Smith, Funk, andHaaf (2000) suggested that identified perpetrators of CSA in their sampletended to be older than 30 years of age (approximately 63% of their samplewas 30 years of age or older, with a mean age of 33.51 years). With theexception of Finkelhor and Redfield’s (1984) study, however, all previousCSA vignette studies either neglected to identify the ages of depicted per-petrators or depicted perpetrators as being 35 years old (Davies & Rogers,2006). Furthermore, previous CSA vignette studies tend to not entertain theidea that perpetrators could be under the age of consent.

Assessing judgments toward adolescents as sexual abuse perpetratorswould be particularly useful (Fergusson & Mullen, 1999) as adolescents maybe perpetrators in real cases (Zolondek, Abel, Northey, & Jordan, 2001). Forexample, Finkelhor (1979) reported that teenagers (who range in age from10 to 19 years) are the largest group of sexual perpetrators, followed byadults ranging in age from 20 to 29 years and then those ranging in agefrom 30 to 39 years. Consistent with Finkelhor (1979), the Federal Bureau ofInvestigation’s (2002) Uniform Crime Report suggested that individuals underthe age of 18 years account for 16.7% of all forcible rapes and 20.6% of othersexual offenses. Furthermore, juvenile sex offenders and their response tointervention often are viewed negatively (Sahlstrom & Jeglic, 2008). Althoughsome judicial systems might not consider adolescents to be subject to CSAlegislation, the psychological literature uses the term “CSA” between same-age adolescents when coercion or a power imbalance is involved (Rind,Tromovitch, & Bauserman, 1998). In addition, many policies developed foradult sex offenders (e.g., community notification) have been extended toadolescent sex offenders, with 39 states requiring adjudicated adolescentsex offenders to register as such (Szymanski, 2009). Thus, examining per-petrators who vary in age may be a particularly important contribution tothe research literature. As a result, the current study investigates participants’ratings of perpetrators who are of three different ages (15, 25, and 35 years).The youngest age (15 years old) is being used because the mean age foradolescent perpetrators is 14.8 years (Fehrenbach, Smith, Monastersky, &Deisher, 1986).

In addition, the age of the victim may be an important variable to con-sider. In this vignette study, the depicted adolescent victim is 15 years oldbecause nearly all American states (49 states) consider 15 years of age tobe below the age of consent (Leitenberg & Saltzman, 2003). Furthermore,some research suggests that the age of the child victim may be related to thedegree of responsibility that is attributed to the child victim. For example,Bottoms and Goodman (1994) suggested that children who are approach-ing adulthood are considered responsible for their abuse in the same wayas an adult victim might be considered responsible for their assault experi-ences. Thus, individuals’ attributions of CSA may be related to the age of thedepicted victim as well.

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

160 A. Sherrill et al.

THE ROLE OF SEX IN CSA INTERACTIONS

In addition to age, the sex of individuals depicted in CSA vignettes as wellas of those rating the vignettes may be important. For example, Pollard(1992) drew attention to the trend that women generally make more pro-victim judgments than do men. Furthermore, CSA vignette research suggeststhat male participants generally view CSA as less serious (Graham, Rogers, &Davies, 2007), show less sympathy toward victims (Davies, Pollard, & Archer,2001; Davies, Rogers, & Hood, 2009), view victims as less credible (McCauley& Parker, 2001), and attribute more culpability and responsibility to maleadolescent victims relative to that attributed to female adolescent victims(Back & Lips, 1998; Bornstein, Kaplan, & Perry, 2007; Broussard & Wagner,1988; Broussard et al., 1991; Davies & Rogers, 2009; Maynard & Wiederman,1997; Rogers & Davies, 2007; Rogers, Titterington, & Davies, 2009; Waterman& Foss-Goodman, 1984). In a study examining sex differences in sexualinteractions between an older adolescent and a high school teacher, femaleraters view these sexual interactions as having more negative effects on theadolescents in this scenario (Smith, Fromuth, & Morris, 1997). Thus, sexdifferences also may be pertinent to the individuals depicted in vignettestudies of CSA as well as to real life circumstances.

ATTRIBUTIONS OF BLAME

In the context of the age and sex of individuals depicted in CSA vignettes,understanding individuals’ ratings of who is to blame and who is responsiblefor interactions involving CSA may assist in identifying important definingfeatures of CSA. Unfortunately, children who have been abused sexuallyoften are held responsible for their own victimization (Davies & Rogers,2009; Pollard, 1992; Waterman & Foss-Goodman, 1984), resulting in vic-tim blaming. Such findings are unfortunate and may be discrepant withvictims’ actual experience, as victims may make more perpetrator-blameattributions and fewer self-blame attributions over time (Feiring & Cleland,2007). Furthermore, victim blaming often lowers victims’ chances for suc-cessful recovery (Frazier & Schauben, 1994). When children are blamed orare perceived to be responsible for their own abuse, those who hold suchbeliefs often think that CSA is not detrimental to these child victims, eventhough children cannot give consent legally (De Young, 1982).

Contrary to such ill-informed beliefs, it is supported firmly that holdingchildren responsible for abuse experiences is, in itself, detrimental to recov-ery (Frazier & Schauben, 1994). In fact, reluctance of children to discusstheir abuse experiences may have long-term effects on their functioning. Forexample, individuals who do not discuss their abuse experiences thoroughlywithin one year of their experiences of CSA tend to exhibit a significant

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

Undergraduate Students’ Attributions 161

number of mental health symptoms into adulthood (O’Leary, Coohey, &Easton, 2010). Such findings suggest that positive responses to children’sdisclosures of CSA are particularly important. Feiring and Cleland (2007)reported that victims’ initial levels of self-blame attributions predicted theirsubsequent experience of depression and intrusive experiences after con-trolling for demographic variables. Given such findings, Renk, Liljequest,Steinberg, Bosco, and Phares (2002) suggested that the responsibility for pre-venting CSA should be placed on perpetrators, thereby removing blame fromvictims. Thus, the current study is designed to contribute to the scientific lit-erature by examining participants’ ratings of depicted perpetrators’ blameand responsibility as well as their ratings of how abusive and representativeof CSA vignettes appear to be.

OTHER ATTITUDES THAT MAY BE RELATED TO CSAATTRIBUTIONS

In addition to the variables already described, other attitudes may be impor-tant in predicting individuals’ attributions regarding CSA vignettes. One suchattitude may be individuals’ gender role or beliefs about the responsibil-ities, privileges, and behaviors that are divided traditionally along genderlines (Spence & Hahn, 1997). Although one previous CSA perception studydid not find gender role attitudes to be important (Maynard & Wiederman,1997), the socialization of men and women is important in other studiesof CSA. For example, men’s endorsements of traditional gender role stereo-types are viewed to be one reason that men respond more negatively towardsexual assault victims (Davies et al., 2001). Some also suggest that mascu-line socialization, which plays a part in the male monopoly of CSA, alsomay contribute to sex differences in attributions of CSA (Finkelhor, 1979;Finkelhor & Redfield, 1984). Maynard and Wiederman (1997) also suggestedthat sexual beliefs, such as the perceived appropriateness of various sexualbehaviors, may be beneficial in the assessment of attributions of CSA. Thus,gender role and sexual attitudes also will be examined in the current study.

THE CURRENT STUDY

In the current study, participants were asked to read six CSA vignettesinvolving a 15-year-old victim. In the vignettes, the age (15, 25, or 35 yearsold) and the sex (male or female, with only heterosexual interactions beingused) of depicted perpetrators were manipulated. Participants’ ratings ofthese vignettes were measured using four separate questions that ask thedegree to which perpetrators should be blamed for the actions depicted inthe vignettes, the degree to which perpetrators are responsible for these

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

162 A. Sherrill et al.

actions, the degree to which each vignette depicts abuse, and the degree towhich each vignette represents CSA. This study also examined the impor-tance of perpetrators’ age and sex as well as participants’ sex in participants’ratings of these different facets of CSA. In addition, this study explored therelationships between participants’ ratings and their gender role and sexualattitudes.

Consistent with the literature (e.g., Broussard & Wagner, 1988; Davieset al., 2001; Finkelhor & Redfield, 1984; Jackson & Ferguson, 1983; Pollard,1992; Waterman & Foss-Goodman, 1984), it was hypothesized that femaleparticipants would endorse significantly higher levels of blame and respon-sibility for depicted perpetrators and significantly higher levels of abuse andCSA representation for each vignette. For both male and female participants,it was hypothesized that these ratings would be higher in vignettes depict-ing 25- and 35-year-old perpetrators relative to those depicting 15-year-oldperpetrators (e.g., Finkelhor & Redfield, 1984). It also was hypothesized thatwhen the perpetrator was male (and the victim was female), male partici-pants would endorse higher levels of perpetrator blame and responsibilityand higher levels of abusiveness and CSA representation for each vignette(e.g., Davies et al., 2001). In contrast, it was expected that female partici-pants would rate vignettes depicting male and female perpetrators similarly(e.g., Back & Lips, 1998; Bornstein et al., 2007; Broussard & Wagner, 1988;Broussard et al., 1991; Maynard & Wiederman, 1997; Rogers & Davies, 2007;Waterman & Foss-Goodman, 1984). Finally, it was expected that participants’gender role and sexual attitudes would be related to their ratings of perpetra-tor blame, perpetrator responsibility, the abusiveness of depicted vignettes,and the CSA representation of depicted vignettes.

METHOD

Participants

For this study, 262 individuals (136 males and 126 females) participated. Allparticipants were undergraduate students enrolled in psychology courses ata large Southeast university. Students ranged in age from 18 to 36 years,with a mean age of 19.96 years (SD = 2.35 years). The majority of thesample consisted of first-year students (43.9%), with the remainder beingsophomores (22.9%), juniors (21.0%), and seniors (12.2%). Students in thissample were predominantly Caucasian (63.4%), with the remainder identify-ing with another racial background (14.9% Hispanic American, 12.6% AfricanAmerican, 5.3% Asian American, and 3.8% “Other”). The majority of partici-pants in this sample did not have their own children (98.1%). The parents ofthe students in this sample were relatively well educated, with a large portionhaving a bachelor’s degree (27.8%). The remainder varied in their educa-tional background (3.1% doctoral degrees, 16.3% master’s degrees, 10.3%

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

Undergraduate Students’ Attributions 163

associate’s degrees, 16.2% some college, 9.2% high school diploma, 3.6%some high school, 0.8% less than high school, and 2.7% of parents’ educa-tion background was not known). The majority of participants (62.3%) wereraised in households with a yearly family income greater than $70,000.

Measures

DEMOGRAPHICS SHEET

A demographic questionnaire asked participants to describe their age, sex,class standing, socioeconomic status, and ethnicity.

THE SEXUAL ABUSE VIGNETTE

Each participant read and rated six vignettes (three levels of perpetrator ageby two levels of perpetrator sex). The vignettes were counterbalanced toavoid order effects. The vignettes used in the current study were adapteddirectly from the vignette used by Broussard and Wagner (1988; who hadadapted their vignette from Waterman & Foss-Goodman, 1984). The con-struct validity of the victim’s response depicted in the vignette was supportedby a pilot study in which the research participants were professionals work-ing in the field of CSA (Broussard & Wagner, 1988). In the current study,the age of the perpetrator (depicted as 15, 25, and 35 years old) and thesex of the perpetrator (and, consequently, the sex of the depicted victim, asonly heterosexual interactions are used) were manipulated for each vignette.Since perpetrator behavior often predicts attributions of blame, force was notdepicted in the vignettes used in the current study (Bornstein et al., 2007).

Previous studies suggest that there are three general responses exhibitedby CSA victims—being resistant, being encouraging, or acting ambigu-ously (nonresisting and nonencouraging)—and that victim responses predictresearch participants’ ratings (Broussard & Wagner, 1988; Broussard et al.,1991). In Waterman and Foss-Goodman’s (1984) study, the most commonreason for blaming victims was that they “should have resisted” (p. 340).Therefore, in order to control for the interpretation of victim response, anambiguous response (one that was neither resistant nor encouraging) wasused in each vignette. The relationship of the perpetrator to the victim waschosen to be a familiar neighbor, as previous studies support that childrenmore frequently are victimized sexually by an individual whom the childknows and trusts (Finkelhor, 1979; U.S. Department of Health and HumanServices, 2006). Although studies show that abuse by relatives is more trau-matic (Friedrich, Urquiza, & Beilke, 1986), research also suggests that abuseby a trusted neighbor may be more devastating than abuse by a distantrelative (Finkelhor & Browne, 1986).

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

164 A. Sherrill et al.

RESPONSE QUESTIONNAIRE

The response questionnaire for the vignettes was created for this study bythe authors and measured four constructs: participants’ ratings of perpetratorblame (“Who was to blame for events in the depicted scenario?”), perpe-trator responsibility (“Who was responsible for the events in the depictedscenario?”), the abusiveness of the depicted events (“Please rate the abusive-ness of the events in the depicted scenario.”), and the degree to which thevignettes represented CSA ( “To your knowledge, please rate the degree towhich the events in the depicted scenario represented child sexual abuse.”).Maynard and Wiederman (1997) indicated that blame and responsibility aretwo distinct concepts. Blame is the attribution made after the perceiverassesses and does not accept the validity of perpetrators’ justification orexcuses, whereas responsibility is applied to the outcome of a process(Shaver & Drown, 1986). Furthermore, blame conveys emotional negativity,whereas responsibility is more affectively neutral (Weiner, 1995). In addition,perceptions of abuse and of how representative something is of CSA arealso two distinct concepts. Abuse could be more of a personal opinion, andCSA is a societal construct. Each item on the response questionnaire used anine-point Likert-type scale. For the items regarding perpetrator blame andresponsibility, this scale ranged from 1 (name of the potential victim, asdepicted in the scenario) to 9 (the neighbor). For the items regarding theabusiveness of the depicted events and the degree to which the vignettesrepresented CSA, this scale ranged from 1 (not abusive/representative) to 9(abusive/representative).

ATTITUDES TOWARD WOMEN SCALE (AWS)

Because sex differences are documented in previous studies, a gender roleattitude scale was used in this study. To measure gender role attitudes, theshort version of the AWS (Spence, Helmreich, & Stapp, 1973), which assessesthe construct of pro-feminist, egalitarian attitudes (versus traditional, conser-vative attitudes), was used. This scale includes 25 items. Each item is ratedon a scale ranging from 0 to 3, with 0 representing the most traditionalresponse and 3 representing the most pro-feminist response. The total scorewas used in this study.

The AWS (Spence & Helmreich, 1978) is the most frequently used mea-sure of attitudes toward women (Beere, 1990). Previous studies indicate thatinternal consistency coefficients range from .81 to .90 and provide evidencefor criterion and convergent validity (Beere, 1990; Yoder, Rice, Adams, Priest,& Prince, 1982). To construct the short version of the AWS, an item analysiswas performed on data from 527 individuals who completed the original ver-sion. The 25 items that had score distributions that maximally discriminatedamong quartiles for both sexes and had the highest biserial correlations were

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

Undergraduate Students’ Attributions 165

selected from the original 55 items. As a result, the short and full AWS scalesare correlated highly (r = .95; Spence et al., 1973). Furthermore, in a pre-vious study, the short version of the AWS has a high Cronbach’s alpha (.89)and Spearman-Brown split-half reliability (.86; Daugherty & Dambrot, 1986).

BRIEF SEXUAL ATTITUDE SCALE (BSAS)

Maynard and Wiederman (1997) suggested that sexual role beliefs may bebeneficial in the assessment of CSA. Therefore, participants’ sexual attitudeswere examined in this study using the BSAS (Hendrick, Hendrick, & Reich,2006). This scale has four attitudinal subscales. These subscales includePermissiveness, a measure of participants’ attitudes about engaging in casualsex, with items such as “Casual sex is acceptable”; Instrumentality, a measureof participants’ attitudes about the purpose and function of sex, with itemssuch as “The main purpose of sex is to enjoy oneself”; Communion, a mea-sure of participants’ attitudes about the characteristics of an ideally mutualengagement in sex, with items such as “Sex is the closest form of commu-nication between two people”; and Birth Control, a measure of participants’attitudes about using birth control, with items such as “Birth control is partof responsible sexuality” (Hendrick et al., 2006). The BSAS includes 23 itemsthat are scored using a scale ranging from 0 to 4, with 0 representing aresponse that most endorses the applicable attitudinal subscale, 2 represent-ing a neutral response, and 4 representing a response that most opposes theapplicable attitudinal subscale.

The original Sexual Attitudes Scale (Hendrick & Hendrick, 1987)was used widely; however, it needed to be revised because the struc-ture of its components changed somewhat over time (Hendrick et al.,2006). The brief version of the scale (BSAS) contains 23 of the original43 items in identical form and has comparable psychometric propertiesto the longer 43-item scale, including high convergent validity (Hendricket al., 2006). In a previous study, the BSAS has high internal con-sistency with the following Cronbach’s alphas: Permissiveness = .93,Instrumentality = .77, Communion = .71, and Birth Control = .84 (Hendricket al., 2006). Additionally, five of the six correlations among the four sub-scales were below r = .19, with the correlation between Permissivenessand Instrumentality being the highest correlation at r = .41 (Hendrick et al.,2006). Last, one-month test-retest correlations for the subscales were as fol-lows: Permissiveness = .92, Instrumentality = .75, Communion = .86, andBirth Control = .57 (Hendrick et al., 2006).

PROCEDURE

After the institutional review board approved this study, an online researchsystem was used to recruit potential participants and assign extra credit in

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

166 A. Sherrill et al.

an undergraduate psychology course of participants’ choosing in return fortheir participation. During data collection sessions, participants reviewed andwere asked to sign a consent form if they agreed to participate. Participantsalso were told that their responses were anonymous, as their consent formswould be stored separately from their completed surveys (which containedno identifying information). After the consent forms were signed and ques-tions had been addressed, participants completed their survey packets inapproximately 20 minutes. A debriefing form that described the purposeof the study was provided following the completion of the survey packet.Contact information for the university’s counseling center also was pro-vided as a referral source for those who may have experienced distressor discomfort as a result of their participation in this study.

RESULTS

Descriptive Statistics for Participants’ Attributions

On average, participants’ ratings supported the victim and validated thevignettes as abusive and representative of CSA. Across all vignettes, par-ticipants endorsed relatively high ratings of perpetrator blame (total sample:M = 7.17, SD = 1.40; male sample: M = 6.80, SD = 1.86; female sample:M = 7.58, SD = 1.48) and perpetrator responsibility (total sample: M = 7.30,SD = 1.26; male sample: M = 7.02, SD = 1.71; female sample: M = 7.62,SD = 1.52). Furthermore, female participants endorsed significantly higherlevels of perpetrator blame, t (df = 1569) = −9.27, p < .001, and perpetratorresponsibility, t (df = 1568) = −7.28, p < .001, relative to male participants.As previously mentioned, blame and responsibility are two separate con-structs (Shaver & Drown, 1986; Weiner, 1995); however, in this study, thetwo were correlated highly (overall r = .63).

In addition, participants provided relatively high ratings of the vignettes’abusiveness (total sample: M = 6.69, SD = 1.67; male sample: M = 6.11,SD = 2.55; female sample: M = 7.32, SD = 1.86) and perceived representa-tiveness of CSA (total sample: M = 6.66, SD = 1.36; male sample: M = 6.24,SD = 2.78; female sample: M = 7.12, SD = 2.54). Further, female partici-pants endorsed significantly higher levels of abusiveness, t (df = 1569) =−10.85, p < .001, and representativeness of CSA, t (df = 1569) = −6.56,p < .001, than male participants for the vignettes used in this study. Thesetwo constructs also were correlated highly (overall r = .78).

On the AWS, participants endorsed generally pro-feminist ratings (totalsample: M = 3.23, SD = .38; male sample: M = 3.05, SD = .35; femalesample: M = 3.44, SD = .26). Female participants endorsed significantlyhigher levels of pro-feminist ratings relative to male participants, t (df =239) = −10.24, p < .001. On the BSAS, participants scored moderately onPermissiveness (total sample: M = 3.32, SD = 0.91; male sample: M = 2.85,

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

Undergraduate Students’ Attributions 167

SD = .83; female sample: M = 3.82, SD = .71) and Instrumentality (totalsample: M = 3.28, SD = 0.77; male sample: M = 3.11, SD = .76; femalesample: M = 3.45, SD = .74) but relatively low on Birth Control (total sam-ple: M = 1.52, SD = 0.67; male sample: M = 1.65, SD = .69; female sample:M = 1.37, SD = .63) and Communion (total sample: M = 2.38, SD = 0.74;male sample: M = 2.34, SD = .73; female sample: M = 2.42, SD = .75).Female participants endorsed significantly higher levels of Permissiveness,t (df = 260) = −10.11, p < .001, and Instrumentality, t (df = 257) = −3.62,p < .001. In contrast, female participants endorsed significantly lower levelsof Birth Control, t (df = 260) = 3.45, p < .001, relative to male partici-pants. There were no differences between male and female participants onCommunion, t (df = 260) = −.84, p < .40.

Correlational Analyses

To examine the relationships among participants’ ratings of the vignettes,their AWS scores, and their BSAS scores, correlational analyses were con-ducted separately for male and female participants (given the significantdifferences noted previously). It should be noted that, although manycorrelations were significant, they represented relatively small effect sizes.

For female participants, AWS scores were related significantly and posi-tively to ratings of perpetrator blame and responsibility. In addition, females’ratings of instrumentality (from the BSAS) were related significantly andpositively to their ratings of how abusive the depicted vignettes were.Furthermore, females’ ratings of communion (from the BSAS) were relatedsignificantly and positively to their ratings of perpetrator blame, perpetra-tor responsibility, and how representative of CSA the depicted vignetteswere. Finally, their ratings of birth control (from the BSAS) were relatedsignificantly and positively to perpetrator responsibility (see Table 1).

TABLE 1 Correlations

1 2 3 4 5 6 7 8 9

1. Attitude TowardWomen

— −.19∗∗∗ −.30∗∗∗ .07∗ −.06 .06 .10∗∗ .06 .01

2. Permissive −.09∗ — .15∗∗∗ −.05 .38∗∗∗ .04 −.03 .06 .023. Birth Control −.12∗∗∗ .23∗∗∗ — −.15∗∗∗ .09∗∗ −.09∗∗ −.02 −.07∗ −.034. Communion −.04 −.14∗∗∗ .18∗∗∗ — .06 .11∗∗∗ .03 .07 .09∗∗

5. Instrumental .01 .46∗∗∗ .14∗∗∗ .02 — .02 −.06 −.02 −.036. Blame? .08∗ .01 .05 .11∗∗ −.04 — .69∗∗∗ .51∗∗∗ .44∗∗∗

7. Responsible? .15∗∗∗ −.01 .07∗ .12∗∗∗ −.03 .57∗∗∗ — .54∗∗∗ .48∗∗∗

8. Abusive? .06 .06 .07 .03 .10∗∗ .47∗∗∗ .50∗∗∗ — .78∗∗∗

9. Representative? .01 .02 .03 .08∗ .05 .46∗∗∗ .50∗∗∗ .71∗∗∗ —

Note: Scores for male participants are above the diagonal, whereas scores for female participants arebelow the diagonal.∗p < .05 ∗∗p < .01 ∗∗∗p < .001.

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

168 A. Sherrill et al.

For male participants, AWS scores were related significantly and posi-tively to ratings of perpetrator responsibility. In addition, males’ ratings ofcommunion (from the BSAS) were related significantly and positively to theirratings of perpetrator blame and how representative of CSA the depictedvignettes were. Finally, their ratings of Birth Control (from the BSAS) wererelated significantly to victim blame and the abusiveness of the vignettes.

Analyses of Covariance

Analyses of covariance were used to determine whether the age and sexof the depicted perpetrator and the participants’ sex predicted participants’ratings of who was to blame for the depicted interactions, who was respon-sible for the interactions, how abusive the depicted interactions were, andhow representative of CSA the interactions were. In these analyses, theage and sex of the depicted perpetrator and the participants’ sex servedas independent variables. In addition, all interactions among these variables(perpetrator age and perpetrator sex, perpetrator age and participant sex,perpetrator sex and participant sex, and perpetrator age and perpetratorsex and participant sex) were examined in these analyses. Finally, partic-ipants’ ratings—who was to blame for the depicted interactions, who wasresponsible for the interactions, how abusive the depicted interactions were,and how representative of CSA the interactions were—served as depen-dent variables. Significant effects were examined further using Tukey’s LeastSignificant Difference post hoc analyses. These findings are summarized inTable 2.

WHO WAS TO BLAME?

In the analysis examining participants’ ratings of who was to blame for thedepicted interactions, AWS scores (p < .003) and Permissiveness (p < .003),Communion (p < .001), and Instrumentality (p < .008) scores from the BSASwere significant covariates. In addition, the age of the perpetrator accountedfor a significant amount of variance (p < .007). In particular, Tukey’s LeastSignificant Difference post hoc analyses suggested that there were signifi-cant differences in participants’ ratings based on the age of the perpetrator,F (2, 1488) = 112.31, p < .001, with participants rating 15-year-old per-petrators (M = 6.35, SD = .07) as being less to blame than 25-year-old(M = 7.52, SD = .07) and 35-year-old (M = 7.70, SD = .07) perpetrators.Participants’ sex also accounted for a significant amount of variance in partic-ipants’ ratings of who was to blame for the depicted interactions (p < .04).In particular, Tukey’s Least Significant Difference post hoc analyses sug-gested that female participants (M = 7.44, SD = .07) rated the neighbor asbeing more to blame relative to male participants (M = 6.94, SD = .07),

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

Undergraduate Students’ Attributions 169

TABLE 2 Analyses of Covariance

Analysis/Main effectsand interactions

Meansquare df F p

Blame?Attitude Toward Women 21.37 1 8.98 .003Permissiveness 20.80 1 8.74 .003Birth Control 1.03 1 .43 .51Communion 73.06 1 30.69 .001Instrumentality 16.96 1 7.13 .008Perpetrator Age 267.33 2 131.26 .007Perpetrator Sex 35.88 1 7.40 .22Participant Sex 47.04 1 10.21 .04Perpetrator Age X Perpetrator Sex .10 2 1.37 .42Perpetrator Age X Participant Sex 2.01 2 26.78 .04Perpetrator Sex X Participant Sex 4.82 1 64.24 .02Perpetrator Age X Perpetrator Sex

X Participant Sex.08 2 .03 .97

Responsible?Attitude Toward Women 66.34 1 30.19 .001Permissiveness .55 1 .25 .62Birth Control 19.46 1 8.86 .003Communion 24.65 1 11.22 .001Instrumentality 12.32 1 5.61 .02Perpetrator Age 266.88 2 412.69 .03Perpetrator Sex 27.93 1 5.51 .23Participant Sex 26.75 1 7.89 .11Perpetrator Age X Perpetrator Sex .76 2 2.18 .31Perpetrator Age X Participant Sex .24 2 .68 .60Perpetrator Sex X Participant Sex 4.66 1 13.39 .07Perpetrator Age X Perpetrator Sex

X Participant Sex.35 2 .16 .85

Abusive?Attitude Toward Women 38.65 1 10.46 .001Permissiveness 43.25 1 11.70 .001Birth Control .28 1 .08 .78Communion 21.77 1 5.89 .02Instrumentality .60 1 .16 .69Perpetrator Age 858.18 2 35.89 .02Perpetrator Sex 164.88 1 3.64 .30Participant Sex 117.84 1 3.32 .20Perpetrator Age X Perpetrator Sex 2.41 2 1.37 .42Perpetrator Age X Participant Sex 23.26 2 13.19 .07Perpetrator Sex X Participant Sex 44.70 1 25.35 .04Perpetrator Age X Perpetrator Sex

X Participant Sex1.76 2 .48 .62

Representative of CSA?Attitude Toward Women 2.99 1 .88 .35Permissiveness 14.01 1 4.12 .04Birth Control .03 1 .01 .93

(Continued)

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

170 A. Sherrill et al.

TABLE 2 (Continued)

Analysis/Main effectsand interactions

Meansquare df F p

Communion 83.01 1 24.42 .001Instrumentality 3.34 1 .98 .32Perpetrator Age 2671.73 1 367.44 .001Perpetrator Sex 139.71 1 4.95 .23Participant Sex 83.87 1 5.13 .19Perpetrator Age X Perpetrator Sex 3.87 2 3.56 .22Perpetrator Age X Participant Sex 4.49 2 4.13 .20Perpetrator Sex X Participant Sex 25.44 1 23.40 .04Perpetrator Age X Perpetrator Sex

X Participant Sex1.09 2 .32 .73

F (1, 1488) = 19.76, p < .001. More important, there were significant inter-actions between the age of the perpetrator and participants’ sex (p < .04)and between the sex of the perpetrator and participants’ sex (p < .02).

Post hoc analyses of the perpetrator age and participant sex interaction,F (5, 1565) = 62.96, p < .001, suggested that there were several significantdifferences in participants’ ratings of who was to blame for the interaction.In particular, male participants rated the 15-year-old perpetrator as beingsignificantly less to blame for the depicted interaction relative to all othervignettes. Similarly, female participants rated the 15-year-old perpetrator asbeing significantly less to blame relative to male and female participants’ratings of 25- and 35-year-old perpetrators. Male participants rated 25-year-old perpetrators as being less to blame relative to female participants’ ratingsof 25- and 35-year-old perpetrators. Male participants also rated 35-year-oldperpetrators as being less to blame relative to female participants’ ratings of25- and 35-year-old perpetrators.

Post hoc analyses of the perpetrator sex and participant sex interaction,F (3, 1567) = 33.38, p < .001, suggested that male participants rated femaleperpetrators as being significantly less to blame for the interaction relative tomale participants who rated male perpetrators and female participants whorated male and female perpetrators. In addition, male participants rated maleperpetrators as being significantly less to blame for the interaction relativeto female participants who rated male and female perpetrators.

WHO WAS RESPONSIBLE?

In the analysis examining participants’ ratings of who was responsible forthe depicted interactions, AWS scores (p < .001) and Birth Control (p <

.003), Communion (p < .001), and Instrumentality (p < .02) scores fromthe BSAS were significant covariates. In addition, the age of the perpetra-tor accounted for a significant amount of variance in participants’ ratings

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

Undergraduate Students’ Attributions 171

of who was responsible (p < .03). In particular, Tukey’s Least SignificantDifference post hoc analyses suggested that there were significant differ-ences in participants’ ratings based on the age of the perpetrator, F (2, 1487)= 121.47, p < .001, with participants rating 15-year-old perpetrators (M =6.47, SD = .07) as being less to blame than 25-year-old (M = 7.64, SD = .07)and 35-year-old (M = 7.82, SD = .07) perpetrators. More important, therewas a marginally significant interaction between the sex of the perpetratorand participants’ sex (p < .07). Post hoc analyses of this interaction, F (3,1566) = 21.91, p < .001, suggested that male participants rated female per-petrators as being significantly less responsible for the interaction relative tomale participants who rated male perpetrators and female participants whorated male and female perpetrators. In addition, male participants rated maleperpetrators as being significantly less responsible for the interaction relativeto female participants who rated male and female perpetrators.

HOW ABUSIVE?

In the analysis examining participants’ ratings of the abusiveness of thedepicted interactions, AWS scores (p < .001) and Permissiveness (p < .001)and Communion (p < .02) scores from the BSAS were significant covari-ates. In addition, the age of the perpetrator accounted for a significantamount of variance in participants’ ratings of how abusive the depictedinteractions were perceived to be (p < .02). In particular, Tukey’s LeastSignificant Difference post hoc analyses suggested that there were signifi-cant differences in participants’ ratings based on the age of the perpetrator,F (2, 1488) = 232.16, p < .001. Participants rated the vignettes depicting15-year-old perpetrators (M = 5.21, SD = .09) as being less abusive thanthose with 25-year-old (M = 7.27, SD = .09) and 35-year-old (M = 7.64,SD = .09) perpetrators. In addition, participants rated vignettes depicting25-year-old perpetrators as being less abusive than those with 35-year-oldperpetrators. More important, there was a significant interaction between thesex of the perpetrator and participants’ sex (p < .04) as well as a marginallysignificant interaction between the age of the perpetrator and participants’sex (p < .07).

Post hoc analyses of the perpetrator sex and participant sex interaction,F (3, 1567) = 53.85, p < .001, suggested that males rated vignettes depictingfemale perpetrators as being significantly less abusive relative to males whorated vignettes depicting male perpetrators and females who rated vignettesdepicting male and female perpetrators. In addition, males rated vignettesdepicting male perpetrators as being significantly less abusive relative tofemales who rated vignettes depicting male and female perpetrators.

Post hoc analyses of the perpetrator age and participant sex interaction,F (5, 1565) = 121.74, p < .001, suggested that there were several significantdifferences in participants’ ratings of the vignettes’ abusiveness. In particular,

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

172 A. Sherrill et al.

males rated vignettes depicting 15-year-old perpetrators as being signifi-cantly less abusive relative to all other vignettes. Similarly, females ratedvignettes depicting 15-year-old perpetrators as being significantly less abu-sive relative to male and female participants’ ratings of vignettes depicting25- and 35-year-old perpetrators. Males rated vignettes depicting 25-year-old perpetrators as being less abusive relative to males’ ratings of vignettesdepicting 35-year-old perpetrators and females’ ratings of vignettes depict-ing 25- and 35-year-old perpetrators. Males also rated vignettes depicting35-year-old perpetrators as being less abusive relative to females’ ratings ofvignettes depicting 25- and 35-year-old perpetrators.

HOW REPRESENTATIVE OF CSA?

In examining participants’ ratings of how representative of CSA the depictedinteractions were, Permissiveness (p < .04) and Communion (p < .001)scores from the BSAS were significant covariates. In addition, the age of theperpetrator accounted for a significant amount of variance (p < .001). Inparticular, Tukey’s Least Significant Difference post hoc analyses suggestedthat there were significant differences in participants’ ratings based on theage of the perpetrator, F (2, 1488) = 785.96, p < .001. Participants ratedvignettes depicting 15-year-old perpetrators (M = 3.99, SD = .08) as beingless representative of CSA than those with 25-year-old (M = 7.80, SD =.08) and 35-year-old (M = 8.16, SD = .08) perpetrators. In addition, partici-pants rated vignettes depicting 25-year-old perpetrators as being less abusivethan those with 35-year-old perpetrators. More important, there was a sig-nificant interaction between the sex of the perpetrator and participants’ sex(p < .04).

Post hoc analyses of this interaction, F (3, 1567) = 22.60, p < .001,suggested that males rated vignettes depicting female perpetrators as beingsignificantly less representative of CSA relative to males’ ratings of vignettesdepicting male perpetrators and females’ ratings of vignettes depicting maleand female perpetrators. In addition, males rated vignettes depicting maleperpetrators as being significantly less representative of CSA relative tofemales’ ratings of vignettes depicting male perpetrators.

DISCUSSION

The results of the current study highlight the need for public educa-tion regarding the potential applicability of CSA to adult–adolescent andadolescent–adolescent sexual interactions. Education programs are impor-tant for the general public and for professionals who are likely to encounterCSA. By providing such programs, more incidents of CSA may be reported,

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

Undergraduate Students’ Attributions 173

and appropriate services then can be provided to those who require assis-tance in dealing with experiences of CSA. When defining CSA, it is commonfor individuals to disagree on the severity of a nonconsensual sexual interac-tion as well as on their opinions of who is to blame and who is responsiblefor such an interaction, particularly when the individuals involved arebetween childhood and adulthood. The findings of this study suggest thatthere are particular factors that may contribute to individuals’ attributionsof CSA.

One factor that appears to be particularly important in attributions thatare made regarding CSA is the age of the perpetrator. In particular, in thisstudy, the age of the perpetrator was important in predicting individuals’ratings of CSA, even in the context of gender role and sexual attitudesand other characteristics of research participants and depicted perpetrators.Across the questions asked about the CSA vignettes, older depicted perpetra-tors prompted more pro-victim ratings by college student participants. Whenthe perpetrator was depicted to be either 10 or 20 years older than the 15-year-old victim, participants rated the depicted perpetrator as having moreblame and responsibility for the depicted interaction. Furthermore, partic-ipants rated the vignettes with these perpetrators as being more abusiveand more representative of CSA. Given previous literature (e.g., Finkelhor &Redfield, 1984), these findings were expected. There also were differences inthe ratings given to vignettes with 25- versus 35-year-old perpetrators, how-ever. Although these findings suggested that 35-year-old perpetrators wereheld more accountable for the interactions that were depicted, it is unlikelythat CSA legislation will discern between these two ages. Nonetheless, thisfinding has implications for those individuals who may be working withchildren and adolescents who are disclosing experiences of CSA. Theseindividuals should be reminded that, even when perpetrators are youngadults, perpetrators should still be held accountable for their victimization ofchildren and adolescents. The significant difference between the depictionsof 15-year-old perpetrators versus 25- and 35-year-old perpetrators requiresgreater attention from future research. Overall, these results indicate thatperpetrators’ age may be a factor in participants’ attributions regarding CSA.

With regard to who was to blame for the depicted interactions and howabusive these interactions were rated, the significant differences between25- and 35-year-old perpetrators were more pronounced for male partici-pants. As mentioned earlier, previous studies indicate that females are lessinfluenced by demographic variables of depicted perpetrators and victims(Back & Lips, 1998; Bornstein et al., 2007; Broussard & Wagner, 1988;Broussard et al., 1991; Maynard & Wiederman, 1997; Rogers & Davies, 2007;Waterman & Foss-Goodman, 1984). The current study supports this find-ing. Although it may be the case that males and females are significantlydifferent in the attributions that they make about CSA, it also may be thecase that females are not as responsive to vignette studies. In order to better

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

174 A. Sherrill et al.

understand the influence of the age of the perpetrator on attributions ofCSA, future research should consider investigating a wider range of ages fordepicted perpetrators in the context of participants’ sex.

These findings with regard to perpetrator age may be particularlyimportant, as research shows that victims who are sexually abused byperpetrators who are younger than 26 years of age report lower levelsof trauma than those victimized by perpetrators ranging in age from 26to 50 years (Russell, 1986). Briere and Runtz (1988) indicated that abuse-related symptoms are correlated positively with the age of the perpetrator.Nonetheless, in both of these previous studies, victims of younger per-petrators also report trauma, albeit less than victims of older perpetrators(Briere & Runtz, 1988; Russell, 1986). These findings suggest that clinicalinterventions should be planned to address victims’ attributions regardingtheir abuse experiences as well as the symptoms exhibited by victims fol-lowing these experiences, regardless of the characteristics of the perpetrator.These findings appear to be consistent with those of the current study, inthat participants’ ratings of blame and responsibility were higher for olderperpetrators.

Future research also should investigate individuals’ perceptions of ado-lescent perpetrators, as the prosecution of such perpetrators has becomea controversial issue (Barbaree & Marshall, 2006). In the current study, forexample, 43.55% of participants perceived the vignettes with 15-year-oldperpetrators to be at least “somewhat” representative of CSA. Barbaree andMarshall (2006) claimed that, prior to the early 1980s, sexual crimes commit-ted by male adolescents were minimized in courtrooms. By the early 2000s,the pendulum swung in the opposite direction, resulting in severe punish-ments for these perpetrators. Given these inconsistent attitudes over time,individuals’ perceptions of adolescent perpetrators deserve further study.Such research may assist in better informing the courts about what typesof interventions and punishments may be most appropriate for adolescentperpetrators.

Beyond perpetrator age, another factor that is important in the predic-tion of attributions of CSA is the sex of the perpetrator. In the current study,males viewed the depicted female perpetrators as being significantly less toblame and less responsible than male perpetrators. The depicted events withfemale perpetrators were rated as significantly less abusive and less repre-sentative of CSA. Similarly, females viewed the depicted female perpetratorsas significantly less to blame and less responsible and the depicted eventswith female perpetrators as significantly less abusive and less representativeof CSA. Despite the current study’s results that depicted male victims areseen as generally more responsible and culpable, it is irrefutable that thevictimization of males does exist. In fact, it is estimated that 3% to 13% ofthe national adult male population was abused sexually during childhood

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

Undergraduate Students’ Attributions 175

(Bolen & Scannapieco, 1999). Others suggest that rates of male sexual abuseare higher, comprising 9% to 25% of all victims of CSA (Rogers & Terry,1984).

Since CSA often is publicized as a feminist issue, male victims have beenneglected by the public and by clinical research (Mezey & King, 1989). It isestimated that research and support for male victims is more than 20 yearsbehind that for female victims (Rogers, 1998). An enhanced educationalcurriculum provided to the general public may assist in reducing the stigmafelt by male victims. Specifically, it will be important to emphasize that malescan suffer the same psychological trauma as females (Ferguson & Mullen,1999; Nielsen, 1983; Rogers & Terry, 1984) and that males might requirespecial treatment (Dhaliwal, Gauzas, Antonowicz, & Ross, 1996; Summit,1983). Thus, further research that examines male victims relative to femalevictims as well as treatments that are meant to be effective for male victimsis needed.

A final factor that appears to be important in predicting individuals’attributions of CSA is the sex of these individuals. Consistent with all CSAvignette research, the male and female participants in this study perceivedthe depicted CSA vignettes differently, with females endorsing more pro-victim ratings, as noted by the results of t-tests as well as by the significantinteractions in the ANCOVAs conducted for this study. This sex difference ispronounced even more when the victim is depicted as a male. To explainsimilar differences, Finkelhor (1984) suggested that CSA should be describedas “a problem of masculine socialization” (p. 12). Although the differences inthe socialization of the sexes are used to account for the distribution of per-petrators (males constitute 95% of perpetrators in cases of the abuse of girlsand 80% in cases of the abuse of boys; Finkelhor, 1984), these differencesalso may explain why male participants generally exhibit less sympathy dur-ing CSA vignette studies. In particular, differences in the socialization of thesexes suggest that women learn to distinguish between sexual and non-sexual forms of affection, that men see heterosexual success as much moreimportant to their gender identities, and that men are socialized to focus theirsexual interest around sexual acts isolated from the context of a relationship(Finkelhor, 1984).

As covariates in this study, individuals’ gender role and sexual atti-tudes appeared to be important for individuals’ ratings of the CSA vignettesused in this study. Based on the findings of this study, the egalitarianattitudes held by both male and female participants appeared importantfor their attributions of perpetrator responsibility for depicted interactions.Although this relationship occurred across male and female participants,other relationships between gender role, sexual attitudes, and attributions ofCSA occurred differentially across male and female participants. Certainly,as noted in the research literature (e.g., Maynard & Wiederman, 1997),

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

176 A. Sherrill et al.

continued examination of male and female gender role and sexual attitudesin the context of attributions of CSA is warranted. In particular, it may be thecase that individuals’ biological sex, in conjunction with their gender rolesand sexual attitudes, are important for predicting individuals’ attributionsof CSA.

The findings of the current study should be examined within the con-text of its limitations. The generalizability of these results is limited by itsconvenience sample of undergraduate students. The few studies that haveused samples other than undergraduate students and those that use samplesof convenience show consistent findings, but it is important not to assumethat the findings of this study could be generalized to different samples ofindividuals (Davies & Rogers, 2006). Furthermore, it must be rememberedthat this study used hypothetical vignettes and may lack generalizability toreal CSA cases (Davies & Rogers, 2006). In particular, the vignettes maynot have been as salient as real individuals describing experiences of CSA. Italso may be the case that the vignettes departed from real life experiences inthat participants were able to review and possibly compare several potentialexamples of CSA. A design in which participants review only one vignettemay resemble the legal system more closely, as jurors are usually exposed toonly one case at a time. Finally, other demographics (e.g., participants’ ageand parental status) and trauma-related variables (e.g., participants’ experi-ence of victimization in their own lives) were not examined in this study;however, such variables may prove fruitful in future research endeavors anddeserve examination.

Even in the context of these limitations, this study suggests several valu-able lines of future research. First, this study used Likert-type questions tomeasure individuals’ ratings of depicted perpetrators and victims with theassumption that all participants would view these ratings as occurring onone continuum. It might be the case that some participants did not concep-tualize these attributions on a continuum (e.g., a participant might perceiveneither depicted individual to be responsible for CSA). Therefore, it mightbehoove future studies to use separate questions for perpetrators and vic-tims. Second, the current study used vignettes that depict only heterosexualinteractions (male victims with female perpetrators and female victims withmale perpetrators). Homosexual interactions were not used because previ-ous research suggests that negative public attitudes toward homosexualitymay be a confounding variable (Kite & Whitley, 1996). Nonetheless, moreresearch is needed regarding ratings of adolescent–adolescent homosexualvignettes in which there is no overt consent. Third, instead of exposing eachparticipant to all vignettes, using a between-subjects design that exposeseach participant to only one vignette might increase generalizability. Finally,future research should include self-perception measures as covariates, assuch measures may be important in the context of individuals’ attributionsof CSA. In particular, the ages of most participants in this study were close

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

Undergraduate Students’ Attributions 177

to the ages of the depicted individuals in the vignettes, potentially prompt-ing a different level of identification with the individuals depicted in thevignettes.

In conclusion, results of this study suggest that there are significantdifferences between male and female participants in their attributions of CSA,with females rating perpetrators as having more blame and responsibility fordepicted CSA interactions, rating the depicted interactions as being moreabusive and more representative of CSA, and rating themselves as higherin pro-feminist and some sexual attitudes. Results of analyses of covariancefurther suggest that the age of the depicted perpetrator is important in par-ticipants’ ratings of perpetrators’ blame and responsibility as well as for howabusive and representative of CSA depicted vignettes are. Although the find-ings of this study contribute to the understanding of attributions of CSA,these findings also raise further research questions. Nonetheless, the find-ings of this study suggest the importance of research examining factors thatpredict individuals’ ratings of CSA. These findings also emphasize the needto educate individuals about CSA and unwanted sexual contact involvingindividuals under the age of consent.

REFERENCES

Back, S., & Lips, H. M. (1998). Child sexual abuse: Victim age, victim gender, andobserver gender as factors contributing to attributions of responsibility. ChildAbuse & Neglect, 22, 1239–1252.

Barbaree, H. E., & Marshall, L. M. (2006). An introduction to the juvenile sexoffender: Terms, concepts, and definitions. In H. E. Barbaree & L. M. Marshall(Eds.), The juvenile sex offender (pp. 1–8). New York, NY: The Guilford Press.

Beere, C. (1990). Gender roles: A handbook of tests and measures. New York, NY:Greenwood Press.

Bolen, R. M., & Scannapieco, M. (1999). Prevalence of child sexual abuse: Acorrective meta-analysis. Social Service Review, 73, 281–313.

Bornstein, B., Kaplan, D., & Perry, A. (2007). Child abuse in the eyes of the beholder:Lay perceptions of child sexual and physical abuse. Child Abuse & Neglect, 31,375–395.

Bottoms, B. L., & Goodman, G. S. (1994). Perceptions of children’s credibility insexual assault cases. Journal of Applied Social Psychology, 24, 702–732.

Briere, J., & Runtz, M. (1988). Symptomatology associated with childhood sexualvictimization in a nonclinical adult sample. Child Abuse & Neglect, 12, 51–59.

Broussard, S., & Wagner, W. (1988). Child sexual abuse: Who is to blame? ChildAbuse & Neglect, 12, 563–569.

Broussard, S., Wagner, W., & Kazelskis, R. (1991). Undergraduate students’perceptions of child sexual abuse: The impact of victim sex, perpetratorsex, respondent sex, and victim response. Journal of Family Violence, 6 ,267–278.

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

178 A. Sherrill et al.

Cullen, B. J., Smith, P. H., Funk, J. B., & Haaf, R. A. (2000). A matched cohortof a criminal justice system’s response to child sexual abuse: A profile ofperpetrators. Child Abuse & Neglect, 24, 569–577.

Daugherty, C., & Dambrot, F. (1986). Reliability of Attitudes toward Women Scale.Educational and Psychological Measurement, 46 , 449–453.

Davies, M., Pollard, P., & Archer, J. (2001). The influence of victim gender and sexualorientation on judgments of the victim in a depicted stranger rape. Violence andVictims, 16 , 607–619.

Davies, M., & Rogers, P. (2006). Perceptions of male victims in depicted sex-ual assaults: A review of the literature. Aggression and Violent Behavior, 11,367–377.

Davies, M., & Rogers, P. (2009). Perceptions of blame and credibility toward victimsof childhood sexual abuse: Differences across victim age, victim–perpetratorrelationship and respondent gender in a depicted case. Journal of Child SexualAbuse, 18, 78–92.

Davies, M., Rogers, P., & Hood, P. A. (2009). Perceptions of child sexual abusein a hypothetical cybersexploitation case: The importance of perpetrator hon-esty, outcome type, and respondent gender. Journal of Child Sexual Abuse, 18,422–441.

De Young, M. (1982). Innocent seducer or innocently seduced? The role of the childincest victim. Journal of Child Psychology, 11, 56–60.

Dhaliwal, G. K., Gauzas, L., Antonowicz, D. H., & Ross, R. R. (1996). Adult malesurvivors of childhood sexual abuse: Prevalence, sexual abuse characteristics,and long-term effects. Clinical Psychology Review, 16 , 619–639.

Federal Bureau of Investigation. (2002). Uniform crime report. Washington, DC:Author.

Fehrenbach, P. A., Smith, W., Monastersky, C., & Deisher, R.W. (1986). Adolescentsexual offenders: Offender and offense characteristics. American Journal ofOrthopsychiatry, 56 , 225–233.

Feiring, C., & Cleland, C. (2007). Childhood sexual abuse and abuse-specific attri-butions of blame over 6 years following discovery. Child Abuse & Neglect, 31,1169–1186.

Ferguson, D. M., & Mullen, P. E. (1999). Childhood sexual abuse: An evidence-basedperspective. London, England: Sage.

Finkelhor, D. (1979). Sexually victimized children. New York, NY: Free Press.Finkelhor, D. (1984). Sexual abuse as a social problem. In D. Finkelhor (Ed.), Child

sexual abuse: New theory and research (pp. 1–13). New York, NY: Free Press.Finkelhor, D., & Browne, A. (1986). Initial and long-tem effects: A review of

the research. In D. Finkelhor (Ed.), A sourcebook on child sexual abuse (pp.143–179). Beverly Hills, CA: Sage.

Finkelhor, D., & Redfield, D. (1984). How the public defines sexual abuse. In D.Finkelhor (Ed.), Child sexual abuse: New theory and research (pp. 107–133).New York, NY: Free Press.

Frazier, P. A., & Schauben, L. (1994). Causal attributions and recovery from rape andother stressful life events. Journal of Social and Clinical Psychology, 13, 1–14.

Friedrich, W. N., Urquiza, A. J., & Beilke, R. (1986). Behavioral problems in sexuallyabused children. Journal of Pediatric Psychology, 11, 47–57.

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

Undergraduate Students’ Attributions 179

Graham, L., Rogers, P., & Davies, M. (2007). Attributions in a hypothetical childsexual abuse case: Roles of abuse type, family response and respondent gender.Journal of Family Violence, 22, 733–745.

Hendrick, S. S., & Hendrick, C. (1987). Multidimensionality of sexual attitudes. TheJournal of Sex Research, 23, 502–526.

Hendrick, C., Hendrick, S., & Reich, D. A. (2006). The Brief Sexual Attitudes Scale.The Journal of Sex Research, 43, 76–86.

Jackson, T., & Ferguson, W. (1983). Attribution of blame in incest. American Journalof Community Psychology, 11, 313–322.

Jones, E. D., & McCurdy, K. (1992). The links between types of maltreatment anddemographic characteristics of children. Child Abuse & Neglect, 16 , 201–251.

Kite, M. E., & Whitley, B. E. (1996). Sex differences in attitudes toward homosexualpersons, behaviors, and civil rights: A meta-analysis. Personality and SocialPsychology Bulletin, 22, 336–353.

Leitenberg, H., & Saltzman, H. (2003). College women who had sexual intercoursewhen they were underage minors (13–15): Age of their male partners, relationto current adjustment, and statutory rape implications. Sexual Abuse: A Journalof Research and Treatment, 15, 135–147.

Lippert, T., Cross, T. P., Jones, L., & Walsh, W. (2009). Telling interviewers aboutsexual abuse: Predictors of child disclosure at forensic interviews. ChildMaltreatment, 14, 100–113.

Maynard, C., & Wiederman, M. (1997). Undergraduate students’ perceptions of childsexual abuse: Effects of age, sex and gender-role attitudes. Child Abuse &Neglect, 21, 833–844.

McCauley, M., & Parker, J. F. (2001). When will a child be believed? The impact ofvictim’s age and juror’s gender on children’s credibility and verdict in a childabuse case. Child Abuse & Neglect, 25, 523–539.

Mezey, G., & King, M. (1989). The effects of sexual assault on men: A survey of 22victims. Psychological Medicine, 19, 205–209.

Nielsen, T. (1983). Sexual abuse of boys: Current perspective. Journal of Counselingand Development, 63, 139–142.

O’Leary, P., Coohey, C., & Easton, S. D. (2010). The effect of severe child sex-ual abuse and disclosure on mental health during adulthood. Journal of ChildSexual Abuse, 19, 275–289.

Pollard, P. (1992). Judgments about victims and attackers in depicted rape: A review.British Journal of Social Psychology, 31, 307–326.

Redlich, A. D. (2001). Community notification: Perceptions of its effectivenesspreventing child sexual abuse. Journal of Child Sexual Abuse, 10, 91–116.

Renk, K., Liljequest, L., Steinberg, A., Bosco, G., & Phares, V. (2002). Preventionof child sexual abuse: Are we doing enough? Trauma, Violence, & Abuse, 3,68–84.

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

Rogers, C. M., & Terry, T. (1984). Clinical intervention with boy victims of sex-ual abuse. In I. R. Stuart & J. G. Greer (Eds.), Victims of sexual aggression:

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

180 A. Sherrill et al.

Treatment of children, women, and men (pp. 91–104). New York, NY: VanNostrand Reinhold.

Rogers, P. (1998). Call for research into male rape. Mental Health Practice, 1, 34.Rogers, P., & Davies, M. (2007). Perceptions of victims and perpetrators in a depicted

child sexual abuse case: Gender and age factors. Journal of InterpersonalViolence, 22, 566–584.

Rogers, P., Titterington, L., & Davies, M. (2009). Attributions of blame and credibilityin a hypothetical child sexual abuse case: Roles of victim disability, victim resis-tance and respondent gender. International Journal of Disability, Developmentand Education, 56 , 205–228.

Russell, D. E. H. (1986). The secret trauma: Incest in the lives of girls and women.New York, NY: Basic Books.

Sahlstrom, K. J., & Jeglic, E. L. (2008). Factors affecting attitudes toward juvenile sexoffenders. Journal of Child Sexual Abuse, 17 , 180–196.

Shaver, K., & Drown, D. (1986). On causality, responsibility, and self-blame: Atheoretical note. Journal of Personality and Social Psychology, 50, 697–702.

Smith, H. D., Fromuth, M. E., & Morris, C. C. (1997). Effects of gender on perceptionsof child sexual abuse. Journal of Child Sexual Abuse, 6 , 51–63.

Spence, J., & Hahn, E., (1997). The Attitudes Toward Women Scale and attitudechange in college students. Psychology of Women Quarterly, 21, 17–34.

Spence, J., & Helmreich, R. (1978). Masculinity and femininity: Their psychologicaldimensions, correlates, and antecedents. Austin, TX: University of Texas Press.

Spence, J., Helmreich, R., & Stapp, J. (1973). A short version of the Attitudes TowardWomen Scale (AWS). Bulletin of Psychonomic Society, 2, 219–220.

Summit, R. C. (1983). The child abuse accommodation syndrome. Child Abuse &Neglect, 4, 265–273.

Szymanski, L. (2009). Megan’s law: Juvenile sex offender registration (2009 update).National Center for Juvenile Justice Snapshot, 14. Pittsburgh, PA: National Centerfor Juvenile Justice.

U.S. Department of Health and Human Services. (2006). Child maltreatment 2004.Washington, DC: US Government Printing Office. Retrieved August 28, 2008,from http://www.acf.hhs.gov/programs/cb/pubs/cm04/

Waterman, C., & Foss-Goodman, D. (1984). Child molesting: Variables relating toattribution of fault to victims, offenders, and 112 nonparticipating parents. TheJournal of Sex Research, 20, 329–349.

Weiner, B. (1995). Inferences of responsibility and social motivation. In M. P. Zanna(Ed.), Advances in experimental social psychology (vol. 27, pp.1–47). Hillsdale,NJ: Academic Press.

Williams, L., & Farrell, R. (1990). Legal response to child sexual abuse in day care.Criminal Justice and Behavior, 17 , 284–302.

Yoder, J., Rice, R., Adams, J., Priest. R., & Prince, H. (1982). Reliability of the AttitudeToward Women Scale (AWS) and the Personal Attributes Questionnaire (PAQ).Sex Roles, 8, 651–657.

Zolondek, S. C., Abel, G. G., Northey, W. F., & Jordan, A. D. (2001). The self-reportedbehaviors of juvenile sexual offenders. Journal of Interpersonal Violence, 16 ,73–85.

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011

Undergraduate Students’ Attributions 181

AUTHOR NOTE

Andrew Sherrill completed his Bachelor’s Degree at the University of CentralFlorida and is now a doctoral student at Northern Illinois University.

Kimberly Renk is an Associate Professor of Psychology at the University ofCentral Florida with a specialization in child clinical psychology and parent-child interactions.

Valerie K. Sims is an Associate Professor of Psychology at the Universityof Central Florida with a specialization in cognitive psychology and theinteraction of humans with technology and animals.

Anne Culp is Department Chair and Professor in the Department of Child,Family, and Community Sciences with a specialization in developmental andchild psychology.

Downloaded By: [Sherrill, Andrew] At: 18:41 24 March 2011