Abstract
Reliability, measured by Cronbach's coefficient alpha, and concurrent validity, measured by Pearson's r and polychoric correlation coefficients, were evaluated in this study. A sample of 371 sexual offenders referred to the Behavioral Medicine Institute of Atlanta for evaluation of sexual interests and behaviors by the courts were assessed using the Sexual Interest Cardsort Questionnaire (SI), a self-report measure of deviant and nondeviant sexual interest, as well as indicator variables obtained from classifications assigned by clinicians as a result of 2 hour-long, semistructured clinical interviews. Internal consistency of 75 items from the SI ranged from 0. 71 to 0.96, across 15 categories of sexual interest and behavior. Additionally, the SI was shortened utilizing Cronbach's alphas to maintain a high level of internal consistency. The resulting questionnaire, the shortened SI (SIS), had 45 items and 15 categories. Cronbach's alpha ranged from 0.78 to 0.97. Utilizing Pearson's r and polychoric correlation coefficients, significant correlations were found for the 11 sexually deviant categories of the SI and indicator variables, and the 10 sexually deviant categories of the SIS and indicator variables. The SI and SIS showed a high level of reliability and concurrent validity. Clinical and research issues pertaining to the clinical assessment of male sexual offenders utilizing self-report and clinical interview data, both obtained as the result of comprehensive evaluations, are discussed.
Original language | English (US) |
---|---|
Pages (from-to) | 107-122 |
Number of pages | 16 |
Journal | Sexual Abuse: A Journal of Research and Treatment |
Volume | 12 |
Issue number | 2 |
DOIs | |
State | Published - Apr 2000 |
Externally published | Yes |
Keywords
- clinical assessment
- deviant sexual interest
- male sex offenders
- self-report
ASJC Scopus subject areas
- General Psychology
- Psychiatry and Mental health