. Author manuscript; available in PMC: 2013 Nov 3.

Published in final edited form as: Prev Sci. 2011 Dec;12(4):10.1007/s11121-011-0227-4. doi: 10.1007/s11121-011-0227-4

Table 4. Hierarchical linear model of multiply imputed data (imputations=20): the effects of alcohol specific communication on adolescent alcohol use.

	β (SE)	t-value
Sample Size	1511
Fixed Effects
Intercept	0.07 (0.07)	1.00
Age	0.04 (0.02)	1.71
Gender	−0.02 (0.05)	−0.33
Parent Education	−0.002 (0.02)	−0.12
Family Structure	−0.09 (0.06)	−1.58
Responsiveness	−0.003 (0.03)	−0.08
Demandingness	−0.02 (0.04)	−0.68
Parental Alcohol Use	0.007 (0.005)	1.30
Adolescent Baseline Alcohol Use	0.40 (0.03)^***	14.61
Black	−0.10 (0.06)	−1.68
Other Race/Ethnicity	−0.06 (0.10)	−0.62
Negative Alcohol Messages	−0.15 (0.08)	−1.80
Permissive Alcohol Messages	0.16 (0.10)	1.56
Negative Messages^* Adol Baseline Alcohol Use	0.06 (0.09)	0.63
Permissive Messages^* Adol Baseline Alcohol Use	0.23 (0.10)^*	2.34
Random Effects
Intercept	0.02 (0.01)	1.47
Residual	0.69 (0.03)^***	24.73
R² (compared to intercept only model)	.2113

Note:

p < .05

^***

p<.001