. Author manuscript; available in PMC: 2019 Aug 1.

Published in final edited form as: J Abnorm Child Psychol. 2018 Aug;46(6):1217–1228. doi: 10.1007/s10802-017-0361-y

Table 4.

Fit statistics for cross-sectional and longitudinal multivariate models of CU, ADHD and ODD Behaviors

Age 2	Overall Fit of Model^a						Relative Fit of Model
	−2LL	df	χ²	df	p	AIC	Δχ²	df	p
	Overall Model (Saturated −2LL=1176.734, 1813)
Independent Pathway model	1184.466	1837	7.732	24	1.00	−40.268
Common Pathway Model	1185.096	1841	8.362	28	1.00	−47.638	.63^b	4	.96

Age 3	Overall Fit of Model^a						Relative Fit of Model

	−2LL	df	χ²	df	p	AIC	Δχ²	df	p

Overall Model (Saturated −2LL=1117.925, 1720)
Independent Pathway model	1152.588	1744	34.663	24	.074	−13.337
Common Pathway Model	1159.166	1748	41.241	28	.051	−14.759	6.578^b	4	.16

Longitudinal	Overall Fit of Model^a						Relative Fit of Model

	−2LL	df	χ²	df	p	AIC	Δχ²	df	p

Overall Model (Saturated −2LL=1871.364, 3461)
Bivariate Common Pathway Model	2105.077	3580	233.713	119	.000	−4.287
Reduced Bivariate Common Pathway Model	2108.295	3590	236.931	129	.000	−21.069	3.218^c	10	.98

Note. -2LL=Likelihood Statistic. χ²=Chi-square fit statistic. df=Degrees of freedom. AIC=Akaike’s Information Criterion. Best fitting model indicated in bold.

Overall fit of the model is determined by the difference in -2LL of the model and that of a saturated model.

χ² difference between the Independent Pathway model and Common Pathway model.

χ² difference between the full and reduced longitudinal models.