Table A2.

Results of Multilevel Models Estimating Effects of Entity Theory, and Individual and School‐Average Victimization on Depressive Symptoms Controlling for Victimization in the Previous School and School Quality Rating

	Final Model
	b (95% CI)
Level 1 (Individual) (N = 6,237)
Entity theory	0.07*** (0.06; 0.08)
Victimization	0.19*** (0.17; 0.21)
Gender	0.12*** (0.10; 0.13)
Previous school victimization	0.06*** (0.04; 0.07)
Entity Theory × Victimization	0.00 (−0.02; 0.02)
Level 2 (School) (k = 25)
School‐average victimization	0.46*** (0.22; 0.71)
School achievement	0.00 (0.00; 0.00)
School quality	0.00 (0.00; 0.01
Cross‐level
Entity Theory × School‐average victimization	−0.01 (−0.12; 0.07)
Victimization × School‐average victimization	0.07 (−0.13; 0.28)
Entity Theory × School‐average victimization × Victimization	0.26 * (0.05; 0.47)
Reduction in variance compared to null model
Reduction Level 1 residual variance	0.011 a
Reduction Level 2 intercept variance	0.001 b
Level 2 variability
Entity theory	0.001
Victimization	0.000
Entity Theory × Victimization	0.001*

The bold value underscores the three‐way interaction that tests the main hypothesis.

Unstandardized effects are shown. We controlled for clustering at the school level, and covariates were gender, previous school victimization, and school‐average achievement.

^aNull model residual variance = 0.079; final model residual variance = 0.068.

^bNull model residual variance = 0.002; final model residual variance = 0.001.

^* p < .05;

^** p < .01;

^*** p < .001.