Citation

Waddell JT, Corbin WR, Meier MH, Morean ME, Metrik J. Psychol. Assess. 2020; ePub(ePub): ePub.

Copyright

(Copyright © 2020, American Psychological Association)

DOI

10.1037/pas0000881

PMID

33151731

Abstract

Prior research suggests that cannabis expectancies are related to cannabis misuse and problems. Although there are established measures of cannabis expectancies, existing measures have psychometric limitations and/or are lengthy. Existing measures typically have a two-factor structure of positive and negative expectancies, but recent conceptualizations of alcohol expectancies support a valence- (positive vs. negative) and arousal-based (high vs. low arousal) structure. Thus, the present study sought to test a similar structure for cannabis. Cannabis expectancy items underwent 2 preliminary studies, assessing item valance/arousal (n = 233) and relevance to cannabis (n = 124). A final pool of 76 items underwent exploratory factor analysis (n = 303), and remaining items underwent confirmatory factor analysis in a separate sample (n = 469). Lastly, an additional sample (n = 435) examined validity.

RESULTS suggested a 3-factor structure (general positive, high arousal negative, low arousal negative) for the 17-item Anticipated Effects of Cannabis Scale (AECS), which was invariant across cannabis use frequency, sex, and race/ethnicity. Positive expectancies were strongly associated with cannabis use, whereas low arousal negative expectancies were protective against cannabis frequency; high arousal negative expectancies were strongly associated with more negative consequences and dependence symptoms. In addition, the proposed interpretation of AECS test scores showed evidence of incremental validity relative to another abbreviated measure. The current study provides initial support for the AECS, a brief, psychometrically sound cannabis expectancies measure.

The AECS captures the full range of cannabis effects and may be suited to test discrepancies between cannabis expectancies and subjective response. Additional research is needed to validate its structure and predictive utility. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Language: en