Table 1.

Fit statistics for all models and model comparisons.

		#pa	(Δ)χ2	p	CFI	TLI	SRMR	AIC	BIC
Model 1	No coupling between SOC and psychopathological symptoms	98	117.95	.083	.992	.992	.022	108,868	108,979
Model 2	Unidirectional coupling from psychopathological symptoms to SOC	97	115.18	.100	.992	.993	.025	108,861	108,978
Model 1 vs. Model 2	df = 1		1.99	.159
Model 3	Unidirectional coupling from SOC to psychopathological symptoms	97	112.39	.136	.992	.993	.023	108,860	108,977
Model 1 vs. Model 3	df = 1		78.70	< .001
Model 4	Bidirectional coupling between SOC and psychopathological symptoms	96	110.07	.155	.992	.993	.024	108,855	108,976
Model 3 vs. Model 4	df = 1		1.76	.185
Model 5	Multigroup model (Unidirectional coupling from SOC to psychopathological symptoms)	194	282.21	< .001	.984	.985	.034	104,899	105,130
Model 3 vs. Model 5	df = 97		194.54	< .001

Note. AIC: Akaike Information Criterion, BIC: Bayesian information criterion, CFI: Comparative Fit Index, #pa: number of parameters, df: degrees of freedom, SRMR: Standardized Root-Mean-Square Residual, TLI: Tucker-Lewis Index.