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 4.

Results from multi-level modeling 11th grade: frequency drinking & driving/riding last 30 days

Fixed effect Coefficient se t Ratio
Model for Individuals (N = 3,073)
Commercial sources 0.056 0.006 8.81**
Friends >21 source 0.129 0.008 14.63**
Friends <21 source 0.100 0.013 7.40**
Parent source −0.082 0.016 −5.03**
Stole from home source −0.005 0.015 −0.34

Model for Communities (N = 93)
Commercial access rate (CAR) 0.078 0.024 3.19**
Minor in possession enforcement (MIP) 0.032 0.021 1.49

Cross-Level Effects
CAR → Commercial source −0.027 0.006 −4.06**
CAR → Friend > 21 source 0.044 0.008 5.09**
CAR → Friend < 21 source −0.084 0.013 −6.50**
CAR → Parent source 0.030 0.016 1.79+
CAR → Home source 0.057 0.016 3.47**
MIP → Commercial source −0.036 0.005 −6.22**
MIP → Friend > 21 source 0.056 0.008 6.68**
MIP → Friend <21 source −0.022 0.014 −1.52
MIP → Parent source −0.041 0.018 −2.25**
MIP → Home source −0.017 0.016 −1.10
Variance
Random effect Component se t-value
Community means, u0j 0.028 0.008 3.26**
Commercial source slopes, u1j 0.042 0.014 3.01**
Friends >21 source slopes, u2j 0.013 0.004 3.21**
Friends <21 source slopes, u3j 0.042 0.012 3.47**
Parent source slopes, u4j 0.028 0.013 2.13*
Stole from home source slopes, u5j 0.055 0.019 2.91**
Individual students, rij 0.523 0.014 36.21**
+

p < .10,

*

p < .05,

**

p < .01