Table 2.

Empty models of variation in drinking motives across neighborhoods (Intermediate Data Zones) (95% credible intervals).

	Social	Enhancement	Coping	Conformity
Neighborhood-level variance	0.004 (0.000, 0.012)	0.005 (0.001, 0.014)	0.014 * (0.002, 0.028)	0.002 (0.000, 0.006)
Individual-level variance	0.254 (0.233, 0.277)	0.242 (0.221, 0.264)	0.246 (0.224, 0.270)	0.156 (0.144, 0.170)
Neighborhood% of variance accounted for	1.3%	1.9%	5.2%	1.2%
Improvement in Bayesian DIC with neighborhood inclusion	No	Yes	Yes	No
DIC-1 level model	1645.62	1586.56	1646.82	1098.83
DIC-2 level model	1646.59	1586.08	1633.47	1101.18

Burn-in 5000; chain 200,000; Bayesian Deviance Information Criteria is used to examine for model fit improvement in single level compared to multilevel models; * p $<0.05$ z-score test; motives are log-transformed.