Table 2.

Logistic Regression Predicting Extreme Consumption Gaming from Demographics, Typical Alcohol Consumption, and Frequency of Prepartying.

Variable	Model 1			Model 2
	B	SE B	OR	B	SE B	OR
Block 1
Age	0.20	0.21	1.22	0.15	0.23	1.02
Gender	−0.36	0.47	0.70	−0.27	0.49	0.76
Grades	−0.61**	0.18	0.54	−0.54**	0.19	0.58
Typical Alcohol Consumptiona	0.44***	0.12	1.55	0.31*	0.12	1.37
Block 2
Prepartying Frequencyb				0.38**	.014	1.46
−2 Log Likelihood	142.06***			133.11***
Nagelkerke R2	0.35			0.41

Note. N=180. B=unstandardized coefficient, OR=Odds ratio based on Exp(B);

^aIndexed by the AUDIT consumption subscale;

^bNumber of prepartying events in the past month. The model correctly classified 85.6% of the students (false positive rate=4.5%; false negative rate=10.9%).

^*p<.05,

^**p<.01,

^***p<.001.