Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Joseph C. Witt


This study used 57 adolescent substance abusers receiving treatment and 510 adolescent controls in a validation of the Teen Substance Abuse Rating Scale (TSARS; Hemstreet, 1991). The TSARS was originally designed as a relatively brief, easy to read self-report questionnaire to discriminate between adolescent substance abusers and nonabusers. The 92 items of the TSARS represent specific behaviors that are answered in a Yes/No format. The TSARS was examined to determine if it would meet validity criteria presented by Winters (1990) and Swadi (1990), as well as those presented in the Standards for Educational and Psychological Testing (APA, 1985). A principal components factor analysis revealed 2 stable TSARS factors including a total of 53 of the original TSARS items. Further analyses were completed on the revised 53-item TSARS. Factor 1, Substance Usage, appeared to relate primarily to adolescent substance involvement, while Factor 2, Interpersonal Skills, was broadly construed as representative of specific social skills. Internal consistency was found to be adequate for the 53 TSARS items (Alpha = 0.94). Significant correlations (p $<$.01) were found between the TSARS and two measures of adolescent substance abuse, the Personal Experience Inventory (PEI; Winters & Henley, 1989) and the Substance Abuse Subtle Screening Inventory (SASSI; Miller, 1985). A discriminant function analysis showed the TSARS discriminated significantly (p $<$.0000) between adolescent substance abusers and controls. The TSARS classified 92.65% of the subjects by group correctly. A Lie Scale included as part of the TSARS was also examined. It was reduced from 12 to 9 items that met an endorsement rate of $>$.90 or $<$.10. The 9-item Lie Scale showed promise as a possible measure of atypical response patterns, but further evidence is needed of its reliability. Though several limitations of this study were presented, the TSARS showed promise of becoming a widely used and researched self-report measure of adolescent substance abuse.