Table 3.

Parameter Estimates (Standard Errors) for Multinomial Logistic Regressions Predicting Group Membership

Reference Group	Low sC–Low sAA			High sC–Low sAA
Comparison Group	vs. Low sC–High sAA	vs. High sC–Low sAA	vs. High sC–High sAA	vs. Low sC–Low sAA	vs. Low sC–High sAA	vs. High sC–High sAA
Intercept	−2.613	−0.500	−4.254	0.500	−2.113	−3.754
	(1.971)	(2.250)	(2.349)	(2.250)	(1.919)	(2.254)
Covariates
Child gender	−0.233	0.659	0.528	−0.659	−0.892	−0.131
	(0.482)	(0.551)	(0.603)	(0.551)	(0.495)	(0.607)
Child puberty	1.087	0.775	0.802	0.775	0.312	0.026
	(0.586)	(0.673)	(0.747)	(0.673)	(0.586)	(0.743)
Child medication	−0.086	−0.044	0.102	0.044	−0.042	0.147
	(0.195)	(0.224)	(0.206)	(0.224)	(0.218)	(0.225)
Predictor
Internalizing	0.054	0.049	0.089*	−0.049	0.005	0.040
Problems	(0.039)	(0.043)	(0.043)	(0.043)	(0.033)	(0.036)
Externalizing	0.018	−0.040	−0.008	0.040	0.058*	0.032
Problems	(0.026)	(0.033)	(0.033)	(0.033)	(0.028)	(0.035)
χ2 (df)			16.967 (15)
Nagelkerke’s R2			.127

Note. Parameter estimates reflect multinomial log-odds of group membership (reference vs. comparison) for each unit increase in covariate or predictor of interest.

^*p < .05.