Skip to main content
. Author manuscript; available in PMC: 2017 Sep 1.
Published in final edited form as: J Adolesc Health. 2016 Jun 16;59(3):269–275. doi: 10.1016/j.jadohealth.2016.05.008

Table 2.

Tests of variability in strength of relationships between predictors and two measures of marijuana use

Predictor Any marijuana use
F (7,262)		 Frequent marijuana use
F (7,262)
Low perceived harm 1.64     3.19**
Favorable attitudes to use 0.98 1.20
Easy access 1.94 1.20
Perceived community attitudes favorable to use   2.32* 0.97
Alcohol use 1.06 1.30
Cigarette use 1.18 1.22
*

p < .05;

**

p < .01.

Models included student gender and race/ethnicity as covariates and main effects for the given predictor and year dummy variables. Model estimates are adjusted for clustering of students within schools. The F statistic is for the Wald test of the null hypothesis that year-by-predictor interaction effects are zero.