Skip to main content
. 2016 Oct 1;17(8):1034–1043. doi: 10.1007/s11121-016-0712-x

Table 2.

Regression analyses predicting bullies’ intentions to change their behavior

Model 1: main effects Model 2: interactive effects
B SE β B SE β
Intercept 40.058*** 40.105***
Type of approach −0.093 0.098 −0.044 −0.112 0.097 −0.054
School level 0.255** 0.091 0.124 0.233* 0.091 0.114
Empathy arousal 0.392*** 0.068 0.320 0.354*** 0.069 0.289
Condemning behavior 0.226*** 0.045 0.275 0.227*** 0.045 0.275
Blaming child −0.039 0.036 −0.055 −0.041 0.036 −0.058
Self-efficacy 0.312*** 0.069 0.239 0.291*** 0.070 0.223
Empathy arousal × condemning −0.086* 0.040 −0.112
Empathy arousal × blaming child −0.007 0.037 −0.010

The coding for type of approach was as follows: confronting 0 and non-confronting 1. The coding for school level was as follows: primary school 0 and secondary school 1. Interactions between grade level and each of the four predictors of interest (empathy arousal, condemning behavior, blaming child, and self-efficacy) were tested and were not significant. The interaction between empathy arousal and condemning behavior remains significant when these interactions are included in the model

*p < .05, **p < .01, ***p < .001