Table 3.

Logistic regression model predicting driving after drinking

Predictor	Model 2			Model 3			Model 4
	B	SE	Odds ratio	B	SE	Odds ratio	B	SE	Odds ratio
Gendera	0.15	0.13	1.17	0.21	0.12	1.24	0.13	0.13	1.14
Ageb	0.81	0.12	2.25***	0.85	0.11	2.34***	0.82	0.12	2.27***
Racec	−0.15	0.14	0.86	−0.21	0.13	0.81	−0.15	0.14	0.86
Greek affiliationd	0.38	0.14	1.46**	0.30	0.13	1.35*	0.37	0.14	144**
Family history of alcohol abusee	0.51	0.12	1.67**	0.46	0.12	1.58***	0.51	0.12	1.57***
Total weekly drinks	0.07	0.01	1.07***	0.09	0.01	1.09***	0.07	0.01	1.07***
Self-approval	0.94	0.08	2.56***				0.99	0.09	2.97***
Typical student misperception				0.27	0.05	1.31***	−0.08	0.07	0.92
Model χ2	520.91***			378.85***			522.63***
Model df	7			7			8
Model Nagelkerke R2	29.0%			21.6%			29.0%

Note. Odds ratios are adjusted, controlling for all other predictors within the model. Outcome is driving after more than two drinks (0 = no, 1 = yes).

^aGender (0 = female, 1 = male).

^bAge (0 = younger than 21, 1 = 21 or older).

^cRace (0 = non-Caucasian, 1 = Caucasian).

^dGreek affiliation (0 = non-Greek, 1 = Greek).

^eFamily history of alcohol abuse (0 = negative, 1 = positive).

^*p < .05;

^**p < .01;

^***p < .001.