. Author manuscript; available in PMC: 2024 Jan 1.

Published in final edited form as: Crim Justice Behav. 2022 Mar 21;50(1):6–21. doi: 10.1177/00938548221082997

TABLE 2: Average Effect Due to Treatment (ATE) Across Multiple Models.

Statistical technique	Probation violation count		Probation violation any		Referral count		Referral any
Statistical technique	Effect	95% CI	Effect	95% CI	Effect	95% CI	Effect	95% CI
IPW	0.77	[−0.34, 1.88]	0.17	[−0.06, 0.39]	−0.61^*	[−0.96, −0.25]	−0.20	[−0.43, 0.03]
IPW with covariates	0.84	[−0.78, 2.48]	0.14	[−0.26, 0.54]	−0.59^*	[−1.09, −0.09]	−0.16	[−0.51, 0.19]
CEM	0.64	[−0.54, 1.84]	0.18	[−0.06, 0.42]	−0.82^*	[−1.37, −0.28]	−0.25^*	[−0.48, −0.01]
CEM with covariates	0.63	[−0.47, 1.72]	0.17	[−0.05, 0.41]	−0.82^*	[−1.37, −0.27]	−0.26^*	[−0.48, −0.03]

Note. CEM, n = 121; IPW, n = 290. ATE = average treatment effect; CI = confidence interval; IPW = inverse probability weight; CEM = coarsened exact matching.

p < .05.