. Author manuscript; available in PMC: 2019 Oct 1.

Published in final edited form as: J Youth Adolesc. 2018 Jun 15;47(10):2206–2219. doi: 10.1007/s10964-018-0874-x

Table 4.

Linear growth models predicting intercept and slope for adolescents’ alcohol use in the past year, centered at age 16

Predictors	Intercept (age 16)			Linear Slope (age 16)
Predictors	γ	SE	t-ratio	γ	SE	t-ratio
Model 3: Interactions
Sensation seeking	0.30	0.28	1.08	0.06	0.07	0.78
Parental control	−0.62	0.33	−1.90⁺	0.13	0.09	1.55
Parent-SUD	0.68	0.32	2.14^*	−0.04	0.09	−0.49
SenSk × ParCont	−1.03	0.32	−3.23^***	0.10	0.08	0.23
SenSk × Parent SUD	0.33	0.42	0.77	−0.05	0.11	0.63
ParCont × Parent SUD	0.06	0.61	0.09	0.07	0.17	0.66

Model 3 represents two-way interactions among predictors. Presented are gamma coefficients, standard errors, and t-ratios for multilevel growth models (N = 345)

Note. Analyses controlled for parent education, adolescents’ sex and minority status. SenSk = Sensation seeking; ParCont = Parental Control

⁺

p < .10;

p < 05;

^***

p < .001