A marijuana screening inventory (experimental version): description and preliminary psychometric properties.

Objective: Marijuana use prevalence, culturally confusing messages about marijuana risks, assessment dilemmas, and current screening inadequacies justify developing a marijuana specific screening inventory for assessment purposes. This article describes the Marijuana Screening Inventory (MSI-X) and its preliminary psychometric reliability, factor analyses, and factor structure.

Method: The MSI-X was administered to a community sample of 420 Army reservists participating in substance abuse educational classes. Participants responded anonymously to the 39-item MSI-X. SPSS analyses were performed with 408 returned MSI-Xs from a sample of 49% males and 40% females.

Results: Analyses revealed 61% smoked marijuana during their lifetime. Reliability of the MSI-X was .89. Exploratory factor analyses of 31 scored items by principal components and varimax rotation supported a nine-factor structure, explaining 65.8% of the variance, with all items loading > or = .30. Within the sample, 7.84% scored > or = 7 suggesting "at risk" with marijuana; 6.12% scored 4 to 6 "suggestive of risk"; 20.83% scored 1 to 3 reflecting "normal or experimental" use; and 65.4% scored 0 suggesting "no problem."

Conclusions: The reliability, variance explained, factor-loading matrix of the nine-factor MSI-X structure and clinically predetermined scoring ranges appear useful for screening marijuana use patterns. Factor-based subscales were derived from the factor-loading matrix and described as a base for future confirmatory factor analysis. Although the MSI-X version needs psychometric strengthening, it shows potential as a marijuana-specific screening inventory for use in general mental health and primary care settings.