Table 2.

Summary of goodness-of-fit statistics for the CCFA, CT-C(M-1), and LD models.

Model	χ2Wlsmv	df	p	CFI	TLI	RMSEA	pRmsea	Δχ2	Δdf	p
(1) CCFA models
(1a) PIQ-S	78.44	51	0.008	0.988	0.985	0.040	0.809
(1b) PIQ-T	85.75	51	0.002	0.997	0.997	0.046	>0.999
(2) CT-C(M–1) models
(2a) Baseline model	304.23	228	<0.001	0.992	0.991	0.032	>0.999
(2b) Model with NWF	299.33	226	<0.001	0.993	0.991	0.031	>0.999	9.34	2	0.009
(3) LD models
(3a) Configural invariance	304.23	228	<0.001	0.992	0.991	0.032	>0.999
(3b) Weak invariance	314.22	237	<0.001	0.992	0.991	0.031	>0.999	16.44	9	0.058
(3c) Strong invariance	339.94	258	<0.001	0.992	0.991	0.031	>0.999	39.32	21	0.009
(3d) Model with explanatory variables	384.82	314	0.004	0.993	0.992	0.026	>0.999

CT-C(M–1), correlated trait-correlated method minus one; LD, latent difference; CFI, comparative fit index; TLI, Tucker–Lewis index; RMSEA, root mean square error of approximation. Δχ², Chi-square difference tests (with reference to the previous model).