Validity of the Substance Abuse Scales of the MMPI-2 in a College Student Sample

Abstract

This study assessed the validity of three Minnesota Multiphasic Personality Inventory-2-based (MMPI-2; Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989) substance abuse scales. The scales were the MacAndrew Alcoholism Scale-Revised (MAC-R; Butcher et al., 1989) and the recently developed Addiction Acknowledgment Scale and Addiction Potential Scale (AAS & APS; Weed, Butcher, McKenna, & Ben-Porath, 1992). Study participants were 308 male and female college students who completed the MMPI-2 and the substance use disorder modules of the Diagnostic Interview Schedule. The MAC-R and the APS had a nonsignificant-to-weak relationship with substance dependence as defined by the Diagnostic and Statistical Manual of Mental Disorders (3rd ed., rev. [DSM-III-R]; American Psychiatric Association, 1980). The AAS had a moderate ability to identify the 33 study participants who met DSM-III-R criteria for a substance use disorder, primarily alcoholism. Broadening the definition of substance abuse to a continuum of alcohol/drug problems did not alter the character of findings. These results highlight the relative superiority of direct (AAS) versus subtle (APS, MAC-R) scales to detect substance dependence and support studies that question the overall utility of traditional MMPI scales to identify substance abuse.