Skip to main content
. Author manuscript; available in PMC: 2013 Jul 10.
Published in final edited form as: Prev Med. 2005 Mar;40(3):355–362. doi: 10.1016/j.ypmed.2004.06.014

Table 2.

Results from multi-level modeling for 11th grade: frequency binge drinking last 30 days

Fixed effect Coefficient se t Ratio
Model for Individuals (N=3,318)
Commercial sources 0.061 0.005 10.52**
Friends >21 source 0.187 0.006 27.53**
Friends <21 source 0.070 0.010 6.90**
Parent source −0.055 0.016 −3.42**
Stole from home source 0.010 0.015 0.69

Model for Communities (N=93)
Commercial access rate (CAR) 0.060 0.024 2.46**
Minor in possession enforcement (MIP) −0.035 0.020 −1.68+

Cross-Level Effects
CAR → commercial source 0.021 0.006 3.43**
CAR → Friend > 21 source −0.016 0.006 −2.59**
CAR → Friend < 21 source 0.030 0.010 2.97**
CAR → Parent source 0.088 0.017 5.21**
CAR → Home source −0.000 0.015 −0.00
MIP → Commercial source 0.001 0.006 0.18
MIP → Friend > 21 source −0.002 0.007 −0.37
MIP → Friend <21 source −0.033 0.011 −3.05**
MIP → Parent source 0.020 0.015 1.34
MIP → Home source 0.030 0.017 1.80+
Variance
Random effect Component se t-value
Community means, u0j 0.033 0.008 3.80**
Commercial source slopes, u1j 0.023 0.006 3.89**
Friends >21 source slopes, u2j 0.009 0.002 3.50**
Friends <21 source slopes, u3j 0.024 0.006 3.78**
Parent source slopes, u4j 0.026 0.011 2.40**
Stole from home source slopes, u5j 0.081 0.024 3.31**
Individual students, rij 0.392 0.010 38.06**
+

p < .10,

*

p < .05,

**

p < .01