Reconciling unidimensional and multidimensional models of patterns of drug use.

Abstract
Guttman-scale analyses and factor analyses seem to produce contradictory results in analyzing patterns of multiple drug use. Guttman-scale analyses usually show that drug use fits a unidimensional and cumulative model of involvement; factor analyses reject the unidimensional model. The two methods, however, can be reconciled. Guttman scaling is appropriate for discerning patterns of drug involvement, factor analysis for operationalizing degree of involvement within a specific pattern of use.

This publication has 1 reference indexed in Scilit: