Skip to main content
. Author manuscript; available in PMC: 2015 Sep 1.
Published in final edited form as: Addict Behav. 2014 May 5;39(9):1357–1360. doi: 10.1016/j.addbeh.2014.04.024

Table 2.

Multi-level Models Predicting Alcohol Use among American Indians

Current Drinking Usual Drinks Binge Drinking
B SE B SE B SE
Ever Married −0.432 0.287 −0.386 0.124 ** −0.124 0.117
Currently Married 0.460 0.401 0.245 0.120 * −0.216 0.163
Ever Cohabited 0.495 0.206 * −0.014 0.070 0.167 0.097
Currently Cohabiting −0.084 0.282 −0.117 0.130 −0.224 0.144
Ever Parents 0.228 0.211 0.281 0.118 * 0.087 0.104
Currently Parents −0.908 0.238 *** −0.357 0.099 *** −0.436 0.098 ***
Ever Attended College 0.056 0.158 −0.246 0.071 *** −0.297 0.074 ***
Currently Attending College 0.365 0.210 0.029 0.075 −0.003 0.079
Ever Worked Full-time 0.350 0.158 * −0.020 0.146 −0.068 0.093
Currently Working Full-time 0.644 0.139 *** 0.086 0.063 0.163 0.081 *
Constant −1.010 0.235 *** 0.396 0.225 0.749 0.103 ***
Random Effects
Level 1 Variancea 1.687 0.117
Level 2 Variance 4.299 0.964 1.464 0.090 0.629 0.127
Covariance Intercept and Slope −0.265 0.065 −0.109 0.009 −0.026 0.012
Slope (Age) Variance 0.044 0.012 0.009 0.001 0.005 0.002
Level 3 Variance 0.280 0.143 0.086 0.026 0.059 0.067

Source: National Study of Adolescent Health.

Models control for age, quadratic age, gender, family structure, parent education, and urbanicity.

a

Level 1 variance is only shown for the model using linear regression.

Note: Results are weighted to account for the complex survey design.

***

p <.001,

**

p <.01,

*

p <.05,

p <.1