Novaco Anger Scale: reliability and validity within an adult criminal sample.

This study investigated the reliability and validity of the Novaco Anger Scale (NAS; Novaco, 1994) with two groups of correctional offenders, General Admissions and Violent Admissions. Predominantly White male offenders (N = 204), ranging in age from 18 to 69 years, participated in the study. One-month test-retest reliability for the General Admissions group ranged from .78 to .91 using both similar (paper-pencil) and dissimilar (computerized) retesting methods, with lower scores occurring on retest. Significantly lower scores were found for the Violent Admissions group as compared with the General Admissions group. Concurrent validity was examined in the Violent Admissions group using three anger/aggression measures and clinical ratings of eight anger dimensions. Stronger correlations with other similar anger measures than with negative affect indices revealed concurrent and discriminant validity. Implications for clinical use in an offender population are discussed.

Socially desirable responding and offence characteristics among rapists.

The relationship between socially desirable responding and offence characteristics is examined with 49 rapists. Socially desirable responding (SDR) was measured by the Balanced Inventory of Desirable Responding. Offence characteristics included: (a) perpetrator's relationship to the victim; (b) severity of victim injury; and (c) victim's age. Greater victim injury was associated with lower Impression Management (r = -.27, p < .03) and Denial scores (r = -.32, p < .01). No significant relationship occurred between offence characteristics and the self-deceptive scales of Denial of the Negative and Over Confident Rigidity. Alternative to the underreporting hypothesis, self-presentation and the acquisition of socially appropriate skills may explain the SDR/violence relationship.

Psychopathy and substance use in adolescent male offenders.

Previous research with incarcerated adult male offenders yielded correlations of .28 to .40 between measures of substance abuse and psychopathic characteristics. This comorbidity was replicated in 40 incarcerated adolescent male offenders. Drug and alcohol use was significantly correlated with the behavioral characteristics of psychopathy (r = .48 and .41, respectively).

The Criminal Sentiments Scale: predictive validity in a sample of violent and sex offenders.

This study investigated the predictive validity of the Criminal Sentiments Scale (CSS: Gendreau, Grant, Leipciger, & Collins, 1979) within a sample of violent and sex offenders using conviction and failure on conditional release as the criterion variables. The CSS was completed by 130 male offenders (65 sex offenders and 65 violent offenders) commencing a sentence of greater than two years in a Canadian federal institution. Average time at risk for the sample was 16 months. Arrest and conviction rates for violent offenders and sex offenders were 24.6% and 13.8% respectively, overall failure on release resulting in reincarceration was 41.5% and 18.5% respectively. The results showed no relationship between the CSS and recidivism or release failure. Implications for clinical use among this population of offenders are discussed.

Relationships among clinical and validity scales of the Basic Personality Inventory.

In interpreting the results of a self-report inventory it is important to evaluate the extent to which stylistic distortion may have been operative. This task is complicated because validity measures frequently are confounded with content measures. In order to evaluate the potential utility of the validity measures for the Basic Personality Inventory (BPI) the relationships between validity measures and content scales were evaluated in a sample of 71 inmates. While some validity indices had significant correlations with content scales, other validity indices were relatively independent of the content scales. Recommendations are provided for using the BPI validity scales.