Needle R, Su S, Lavee Y
Department of Family Social Science, University of Minnesota, St. Paul 55108.
Addict Behav. 1989;14(4):429-41. doi: 10.1016/0306-4603(89)90030-0.
Researchers have relied on a number of measurement techniques to construct a summated index of drug involvement to reflect both the qualitative and quantitative dimensions of adolescent drug-using behavior. The purpose of this study was to examine the adequacy of three strategies for constructing a composite measure--stage-of-drug-use, unweighted sum of frequencies of use of different substances, and a weighted composite index of substance use. Data for this study were derived from two independent samples of adolescents. The three drug-use measures were assessed both as predictors of consequences of drug use and as outcome measures explained by familial, intrapersonal and interpersonal factors. The weighted composite index of substance use, though conceptually and methodologically superior to the simple (unweighted) sum of frequency, did not perform any better as a predictor or as an outcome variable than the unweighted measure. The weighted composite index of drug involvement is somewhat complicated to calculate and requires substantial resources. The decision whether to use a weighted composite index or a simple sum of frequencies measure of overall drug involvement should be based on both scientific and practical considerations.
