Table 3.

The Association Between Exposure to Violence, Neighborhood Characteristics, and the Variety of Past Month Tobacco, Alcohol, and Marijuana (TAM) Use and the Variety of Past Year Violence, and Cross-Level Interactions

	Variety of Past Month TAM (N=1661)		Variety of Past Year (N=1718)
	b	SE	b	SE
Intercept	−2.18**	.08	−1.17**	.05
Exposure to violence	.15**	.02	.26**	.02
Age	.28**	.03	−.01	.02
Male	−.04	.12	.15	.08
Hispanica	−.14	.15	−.06	.16
African Americana	−.40**	.13	.35*	.15
Other race/ethnicitya	−.28	.25	−.37	.26
Socioeconomic status	.02	.06	−.07	.04
Low self-control	−.07	.05	.08	.05
Routine activities	.33**	.06	.21**	.04
Curfew	−.08	.06	−.08	.07
Family social support	−.04	.04	−.08*	.04
Peer social support	.12	.07	.01	.04
Peer drug use	.05	.05	--	--
Perceptions of drug harmfulness	−.28**	.04	--	--
Any prior TAM use	.74**	.14	--	--
Peer delinquency	--	--	.04	.04
Any prior violence	--	--	.57**	.10
χ2	108.13		113.52
Direct Neighborhood Effects
Level-1 intercept	−2.18**	.01	−1.17**	.01
Collective efficacy	.13**	.05	.10	.07
Concentrated disadvantage	−.01	.01	.01	.02
R2	.18		.03
Cross-Level Interactions
Exposure to violence	.15**	.01	.26**	.001
x Collective efficacy	−.07**	.03	.001	.01
R2	.09		.00

^**p ≤ .01

^*p ≤ .05

Notes: Individual-level results are based on overdispersed Poisson models; italicized coefficients indicate that the effect of exposure to violence was allowed to vary randomly across the79 neighborhood clusters; all other variables were fixed across neighborhoods. The level-two and cross-level interaction results are based on overdispersed Poisson models using Empirical Bayes estimates; Cross-level interactions control for the direct effects of the neighborhood characteristics and all individual-level variables.

^aCaucasian is the reference group