Skip to main content
. Author manuscript; available in PMC: 2015 Sep 1.
Published in final edited form as: J Adolesc Health. 2014 Apr 19;55(3):380–387. doi: 10.1016/j.jadohealth.2014.02.023

Table 3.

Multinomial Regression Model Predicting Multiple Risk Behavior Trajectories from Significant PBT System Level Variables

Class Predictor (PBT System) OR B SE p
Increasing High-Risk Peer Pressure (PE) .92 −.09 .37 .81
Curfew (PE) .39* −.94 .36 .01
Self-Worth (P) .40 −.92 .49 .06
Violence Inevitability (P) 1.90 .64 .45 .15
Adolescent-Limited Peer Pressure (PE) 1.02 .02 .47 .97
Curfew (PE) .43 −.85 .46 .07
Self-Worth (P) .32 −1.13 .62 .07
Violence Inevitability (P) 2.54 .93 .59 .11
Early-Experimenters Peer Pressure (PE) 2.98* 1.01 .46 .02
Curfew (PE) .14* −2.00 .49 .00
Self-Worth (P) .10* −2.27 .68 .00
Violence Inevitability (P) 5.82* 1.76 .64 .01

Notes.

Regression analyses included adolescent gender as a control variable.

*

p < significant p –value derived from the false discovery rate control procedure (33).

B = regression coefficient. The “normative” group serves as the reference group. (PE)= Perceived Environment system. (P) = Personality system.