Rape prevention: Methodological problems and new directions.

Sexual abuse prevention with high risk males: The roles of victim empathy and rape myths.

Primary prevention with high risk males: Testing two model-based rape prevention programs.

Rape: The role of outcome expectancies and hypermasculinity.

Preliminary evaluation of sexual problems in combat veterans with PTSD.

The credibility of child sexual abuse allegations: Perpetrator gender and subject occupational status.

Psychometrics of the Rape Conformity Assessment and other measures: Implications for rape prevention.

Predictive validity of a clinical evaluation protocol for adult male incest offenders.

Should men judge women's consent? A modification of the sexual experiences survey from a feminist perspective.

Schewe, P.A. & O'Donohue, W.T. (1993a). Rape prevention: Methodological problems and new directions. Clinical Psychology Review, 13, 667-682.

Interventions targeting males and aimed at the primary prevention of rape are reviewed. Because the causes of rape are not clearly known, primary prevention efforts are hampered. Moreover, etiological accounts have largely ignored situational variables, situation x person interactions, and the development of an appropriate taxonomy of rapists. Psychometrically adequate dependent measures of rape proclivity have yet to be developed. Moreover, most existing rape prevention research fails to distinguish between statistically significant and clinically significant results, and fails to assess the durability of any treatment effects. In addition, most programs are based on the problematic assumption that changes in attitudes will result in changes in the actual incidence of rape. Finally, issues such as the need to assess whether other prosocial changes occur (e.g., a decrease in potential to abuse children), the need to evaluate the specific mechanisms of change, as well as the need to assess possible iatrogenic effects of these efforts are discussed.

Schewe, P.A. & O'Donohue, W.T. (1993b). Sexual abuse prevention with high risk males: The roles of victim empathy and rape myths. Violence and Victims, 8(4), 339-351.

The outcome of two sexual abuse prevention programs, one emphasizing victim empathy and the other stressing modifying rape myths was evaluated with high risk males. Sixty-eight "high risk" males, as determined by self-reported likelihood of committing sexual abuse, were randomly assigned to an empathy treatment, facts treatment, or a no-treatment control group. Treatment effects were assessed using subjects' pre- and post-treatment scores on the Likelihood of Sexually Abusing scale, the Rape Empathy Scale, the Acceptance of Interpersonal Violence scale, the Adversarial Sexual Beliefs Scale, and on a test of self-reported sexual arousal to forced versus consenting sex. In addition, post-test scores on an Asch type conformity measure were obtained. Results of validity checks indicated that high risk subjects differed from low risk subjects on a number of rape-related variables, that the victim empathy condition increased subjects' empathy, and that subjects found both treatments to be credible and helpful. Comparisons between the empathy, facts, and no-treatment group contraindicated the practice of dispelling rape myths as a method of preventing rape among high risk males.

Schewe, P.A. & O'Donohue, W.T. (1996). Primary prevention with high risk males: Testing two model-based rape prevention programs. Archives of Sexual Behavior, 25(5), 455-471.

The purpose of this project was to develop and evaluate two model-based programs designed to reduce the amount of date rape attempted by male college students. One program, the Victim Empathy/Outcome Expectancies (VE/OE) intervention, was based on a proposed model which holds that in order to rape, men must lack empathy for their victims and believe that it is in their personal best interest to rape. The goal of this program was to help males generate or experience empathy for victims of rape and to convince them that it is in their own best interest to develop consenting sexual relationships rather than to coerce women into sexual activities. The second program, the Rape Supportive Cognitions (RSC) intervention, was based on a cognitive model which suggests that rape-supportive cognitions contribute to rape. The goal of this program was to help subjects replace rape-supportive cognitions with more pro-social cognitions. Both programs included a counter-attitudinal role-play in which subjects were instructed to convince a hypothetical male to stop raping women. Seventy-four subjects scoring in the upper third of the Attraction to Sexual Aggression scale (Malamuth, 1991) were randomly assigned to one of two treatment groups (VE/OE or RSC) or to a no-treatment control group. Treatment effects were assessed using subjects' scores pre- and post-treatment on the Attraction to Sexual Aggression Scale (Malamuth, 1989), the Rape Myth Acceptance, the Acceptance of Interpersonal Violence, and the Adversarial Sexual Beliefs Scales (Burt, 1980) as well as subjects' post-treatment scores on the Rape Conformity Scale (Schewe & O'Donohue, 1994). Manipulation checks indicated that subjects in the VE/OE condition retained and could use information regarding victim-empathy and the negative consequences of raping, while subjects in the RSC condition retained and could use information regarding rape-myths and sexual communication. Results indicated that both treatments were significantly more effective than no treatment. Importantly, the RSC group showed clinically significant changes on three of the five dependent measures, while the VE/OE group evidenced clinically significant changes on only one measure.

O'Donohue, W.T., McKay, J.S., & Schewe, P.A. (1996). Rape: The role of outcome expectancies and hypermasculinity. Sexual Abuse: A Journal of Research and Treatment, 8(2), 133-141.

Outcome expectancies regarding coercive sexual behavior of male undergraduates were investigated to understand their possible association with self-reported past and future coercive sexual behavior and hypermasculine personality styles. Subjects with higher self-reported future likelihood of rape and subjects who reported a greater past history of coercive sexual behavior had lower negative outcome expectancies regarding rape, rated these negative consequences as less important, and were more likely to fit a hypermasculine personality pattern. Implications for future research and primary prevention are discussed.

Letourneau, E. J., Schewe, P. A. Schewe, & Frueh, B. C. (1997). Preliminary evaluation of sexual problems in combat veterans with PTSD. Journal of Traumatic Stress, 1, 125-132.

This study evaluated the potential relationship between posttraumatic stress disorder (PTSD) and sexual problems. The Golombok Rust Inventory of Sexual Satisfaction was mailed to combat veterans currently in treatment at an outpatient PTSD clinic. Completed questionnaires were received from 90 patients. Results indicated that over 80% of subjects were experiencing clinically relevant sexual difficulties. Impotence and premature ejaculation were the most frequently reported problems that have corresponding psychological diagnoses. The rates of sexual problems for this sample of veterans with PTSD was similar to those reported in other studies and exceeded rates of similar problems found in samples from community samples. These data suggest that PTSD may be a risk factor for sexual problems.

Schewe, P.A. & O'Donohue, W.T. (in press). Psychometrics of the Rape Conformity Assessment and other measures: Implications for rape prevention. Sexual Abuse: A Journal of Research and Treatment.

Investigated the test-retest reliability and internal consistency of Malamuth's (1989) Attraction to Sexual Aggression scale and Burt's (1980) Rape Myth Acceptance, Acceptance of Interpersonal Violence, and Adversarial Sexual Beliefs scales. In addition, psychometric properties and aspects of the construct validity of the Rape Conformity Assessment (RCA; Schewe & O'Donohue, 1993b) were investigated. Forty-two undergraduate males served as subjects. Findings suggest that the RCA may be adequate for use as a disguised measure that addresses the social desirability problem in rape prevention research. The psychometric properties of most of the scales were adequate. The conformity manipulation revealed strong effects which suggests that male rape-potential should be viewed not only as a trait, but also as having situational components.

Saunders, B. & Schewe, P.A. Predictive validity of a clinical evaluation protocol for adult male incest offenders. Manuscript under editorial review.

A standardized protocol for assessing incest offenders was developed and administered to 101 incest offenders (and their families) who met the screening criteria for the study and entered outpatient treatment for their sexually abusive behavior. Post-treatment, 124 therapists were identified who participated in the treatment of these offenders. Therapist rated outcome questionnaires, including the TORS which was specifically developed for this study, were sent to each of the therapists. Of the 53 responses received, 43 identified themselves as the primary therapist for the specified offender, and it is these reports that were used to indicate treatment outcome. A progressive, hierarchical analysis approach for predicting treatment outcome was taken. Factor scores from the initial evaluation results were used as predictors of TORS score. A series of within person and instrument stepwise regression analyses were conducted to test for relationships with the TORS. Significant factors from each of these analyses were then combined to form the final prediction models. Two models were developed. The Offender-Partner prediction model used data from the 26 cases where complete offender and partner data were available to predict the TORS. The Offender-only model used data from the 41 perpetrators to predict the TORS. Results of the simple regression analysis for these two final predictor variable sets were statistically significant and, most important, each model was able to explain over half the variance in treatment outcome scores.

O'Donohue, W.T., Smith, V., & Schewe, P.A. (in press). The credibility of child sexual abuse allegations: Perpetrator gender and subject occupational status. Sexual Abuse: A Journal of Research and Treatment.

Investigated the extent to which the alleged perpetrator's gender, and the subject's mandated reporter status and acquintanceship with sexual abuse affect judgments of the credibility of children's reports that they have been sexually abused. Eighty-one educators and 104 undergraduates read a short vignette in which a child alleged that she was sexually abused and the accused denied the abuse. Results revealed that males and females found the child equally credible; that allegations involving male perpetrators were believed more than allegations involving females; and that educators believed the allegations less than undergraduates. There was a significant interaction between gender of subject and gender of perpetrator that was not consistent with a hypothesized "gender sympathy effect". No relationship between acquintanceship with sexual abuse and credibility ratings was found.

Schewe, P.A. & Leverett, J.P. Should men judge women's consent? A modification of the sexual experiences survey from a feminist perspective. Manuscript in preparation.

This study investigated the effects of rewording the original Sexual Experiences Survey (SES) in an attempt to improve the validity of the instrument when administered to men. Evidence suggests that a population of sexually coercive men may not be identified by the SES (Koss, Gidycz & Wisniewski, 1987). Removal of reference to female consent from the SES was hypothesized to increase the reporting of sexually coercive behavior by men. The rationale for this hypothesis was that reference to the victims' consent may facilitate rationalizations or misperceptions of events on the part of men responding to the items (e.g., "yes, I used force, but not without her consent--she really wanted me to", "she said no, but really meant yes", etc.). Results of this study indicate that removing reference to female consent does not alter the construct validity nor the reliability of the instrument, and only modestly increases reporting. The overall findings suggest that for philosophical as well as empirical reasons, researchers using the SES should omit references to female consent when administering the scale to men.