Psychometric testing of the adolescent version of the Cook-Medley hostility scale.

Hostility, a phenomenon that has been linked to cardiovascular disease and all-cause mortality, is often measured by the Cook-Medley hostility (Ho) scale. Although there is an adolescent version of the Ho scale, it has had little testing with only Anglo-American samples. This pilot study tested the reliability and validity of the adolescent version of the Ho scale with a multiethnic sample of adolescents. Sixty adolescents participated. Reliability was measured using Chronbach's alpha. The Ho scale was correlated with the Speilberger State-Trait Anger Expression Inventory (1988) scales to determine concurrent validity. A factor analysis assessed construct validity; content validity was assessed by analyzing tape-recorded descriptions of a circumstance that provoked anger, as remembered by each adolescent. Internal consistency reliability was .75. The hostility measure was most highly correlated with anger expression (r = .62, p = .000) and trait anger (r = .50, p = .000). The factor analysis generated three factors (suspicious alienation, cynical aggression and justified mistrust), that accounted for 34.5% of the variance. The content analysis resulted in five anger-provoking themes: aggression, unfulfilled personal expectations, mistrust/lying, criticism of effort, and rejection. The majority of items in the scale (74%) were related to one of the themes.