Recently, Weed, Butcher, McKenna, and Ben-Porath (1992) introduced two new scales for assessing alcohol and drug abuse with the Minnesota Multiphasic Personality Inventory-2 (MMPI-2; Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989): the Addiction Potential Scale (APS), an empirically derived measure similar to the MacAndrew Alcoholism Scale (MAC; MacAndrew, 1965), and the Addiction Acknowledgement Scale. Like the MAC, the APS was designed to identify the personality dimensions associated with substance abuse. Although there has been evidence to support the discriminative validity of the APS among samples of abusers, nonabusers, and psychiatric inpatients (Greene, Weed, Butcher, Arrendondo, & Davis, 1992; Weed et al., 1992), there has been no attempt to determine the personality dimensions tapped by the APS. Thus our study investigated the internal complexities of the APS. Using MMPI-2 protocols from a sample of alcoholic and psychiatric inpatients, we subjected the APS to a principal components analysis with subsequent varimax rotation. Both qualitative and quantitative methods were used to name the components that were identified. Five components consistently emerged across settings: Satisfaction/Dissatisfaction with Self, Powerlessness/Lack of Self-Efficacy, Antisocial Acting-Out, Surgency, and Risk-Taking/Recklessness, Clinical advantage gained from a knowledge of the internal structure of the APS was illustrated using 6 patients from the alcoholic sample. A rational approach to validating the empirical structure of the APS was discussed as a potential avenue for future research.
