Table 5.

Intent-to-Treat (ITT) and Treatment-on-Treated (TOT) Effect Estimates for Cumulative Disciplinary Events through Sixth Grade (RCT Analytic Sample)

	ITT						TOT
	Treatment Group Meana	Control Group Meana	Pooled SDb	Coefficient for T-C Differencec	Effect Sized	p - valuee	Coefficient for T-C Differencee	Effect Sized
Observed Values
School Rules	.231	.185	.396	.047*	−.119	.004	.088*	−.222
Major Offenses	.137	.109	.331	.028*	−.083	.043	.052*	−.157
All Offenses	.273	.234	.429	.039*	−.090	.025	.073*	−.170
Weighted Observed Values
School Rules	.249	.194	.409	.055*	−.135	.001	.103*	−.253
Major Offense	.139	.117	.339	.022	−.066	.121	.042	−.123
All Offenses	.287	.250	.440	.037*	−.084	.041	.070*	−.159
	N = 1618–1626	N = 974–980		N = 2592–2606

^*p < .05 for coefficients

Note. School rules: violations of school rules or other administrative issues; major offenses: fighting, bullying, weapon in school, and the like; all offenses: total across school rule and major offenses categories. These are coded for whether there is any infraction recorded in school records cumulatively from K through the sixth-grade year (1 = yes, 0 = no).

^aCovariate-adjusted means generated by the multilevel analysis models.

^bPooled treatment and control group standard deviations. There were minor variations between the pooled SDs for the ITT and TOT; the mean is presented here but effect sizes were computed on the exact values.

^cCoefficients for the treatment-control differences from multilevel models with children nested in R-Lists and R-Lists nested in districts and the standard set of covariates (see text). The multiplier for ITT coefficients that estimates TOT coefficients is 1.8790 for school rule violations, 1.8811 for major offenses, and 1.8847 for all offenses.

^dEffect size: coefficient for the treatment-control difference divided by the pooled standard deviation.

^eThe p-values for statistical significance that are the same for the ITT and TOT coefficients.