Untitled

Abstract

Cognitive distortions are commonly viewed as an important factor in the assessment and treatment of sexual offenders. However, consensus about the operational definition of cognitive distortions and the best instrument to measure such distortions is lacking. This paper evaluates the Bumby MOLEST and RAPE scales as measures of cognitive distortions with patients civilly committed under California's Sexually Violent Predator law. Rapists and child molesters in the current sample endorsed markedly fewer cognitive distortions than the sexual offenders in Bumby's original study (K. M. Bumby, 1996). It is suggested that current self-report measures such as the MOLEST and RAPE scales are too susceptible to a socially desirable response set to provide useful data with sexual offenders who are involuntarily committed for treatment.

A psychometric examination of treatment change in a multisite sample of treated Canadian federal sexual offenders.

“…This sample was not appreciably different overall from published findings in other sex offender samples in their profiles on measures of social desirability (Mathie & Wakeling, 2011), cognitive distortions (Arkowitz & Vess, 2003), loneliness (Marshall et al, 1997), and social intimacy (Nunes et al, 2011); on the latter two measures of loneliness and intimacy, the sample's scores were not particularly pathological and were close to the normative means of the nonoffender development samples. The sample scored, on average, below the nonoffender normative mean on This document is copyrighted by the American Psychological Association or one of its allied publishers.…”

“…Subsequent research has found other sex offender samples to score lower to varying degrees on both scales. Rape scale means of 50.0 ( SD = 17.8) and 65.7 ( SD = 19.7) were reported in two rapist samples, Molest scale means from 66.0 ( SD = 24.03) to 80.1 ( SD = 21.2) in three child molester samples, and inmate and community controls have been found to score in the 50’s on both measures (Arkowitz & Vess, 2003; Blumenthal, Gudjonsson, & Burns, 1999; Marshall, Marshall, Sachdev, & Kruger, 2003). Bumby (1996) found the Rape and Molest scales to have high internal consistency (α = .97 for both), test–retest reliability ( r = .86 and .84, respectively), and convergent validity with established measures of sex offender cognitions.…”

“…Moreover, on the Rape and Molest subscales, the sample as a whole scored considerably lower than that of Bumby’s (1996) normative sample. Rapists and child molesters each scored one or more standard deviations lower on the pretreatment Rape ( M = 56.71, SD = 16.39 and M = 61.25, SD = 16.04, respectively) and Molest ( M = 55.34, SD = 15.74 and M = 65.92, SD = 17.57, respectively) scales compared with the Bumby (1996) sample, although these scores were only within a few points of child molester and rapist samples published elsewhere (cf., Arkowitz & Vess, 2003; Blumenthal et al, 1999; Marshall et al, 2003). Interestingly, the child/incest group of offenders in the current sample scored significantly higher than the rapist/mixed group on both the Rape, t (249) = 2.19, p = .03, d = .28, 95% CIs [.03, .53] and Molest, t (246) = 4.83, p < .001, d = .62, 95% CIs [.36, .87] scales.…”

“…The Rape Scale consists of 36 items (possible scores 36 to 144) and the Molest Scale, 38 items (possible scores 38 to 152). Bumby (1996) (Arkowitz & Vess, 2003;Blumenthal, Gudjonsson, & Burns, 1999;Marshall, Marshall, Sachdev, & Kruger, 2003). Bumby (1996) (Russell, Peplau, & Cutrona, 1980) is a 20-item selfreport measure in which the individual rates their subjective feelings of loneliness on a 1-to 4-point Likert-type scale.…”

Community Males Show Multiple-Perpetrator Rape Proclivity

“…Participants indicate their response on a 4-point Likert-type scale ranging from strongly agree to strongly disagree . When compared with similar scales, such as the Abel-Becker Cognition Scale (Abel, Becker, & Cunningham-Rathner, 1984) or the Burt Rape-Myth Acceptance Scale (Burt, 1980), Bumby’s (1996) Rape Questionnaire holds superior psychometric properties (α = .97; see Arkowitz & Vess, 2003; Bumby, 1996; Grady, Brodersen, & Abramson, 2011).…”