The problems involved in measuring psychotherapeutic effectiveness must be addressed first by developing specific criteria for measuring behavior change. These criteria can be established when a clear-cut distinction between acceptable and unacceptable behavior is well defined. To illustrate the potential use of regression methodology in measuring therapeutic effectiveness, two groups of high school students, i.e., drug users (N=40) and non-drug users (N=40), completed a data questionnaire and the Sixteen Personality Factor Questionnaire (16PF), Form A, to determine whether users and non-users differed on basic personality characteristics. Analyses of results revealed that nine personality factors differentiated the two groups. A multiple regression technique applied to these statistical differences established a prediction equation accounting for 60% of the criterion variance while identifying potential drug users. The results indicate that regression techniques can establish a data base for measuring therapeutic effectiveness when clear-cut distinctions between acceptable and unacceptable behavior are defined. (Author/MCF)