. 2020 Mar 23;9(3):873. doi: 10.3390/jcm9030873

Table 3.

Summary of the Confirmatory Factor Analyses (CFA) for single, two, and bifactorial models of depression and anxiety.

	χ² (df)	χ²/df	p	RMSEA ^a	^b RMSEA 95% CI	^c SRMR	^d CFI	^e TLI	^f NFI	^g IFI
One-factorial M.	477.63 (104)	4.49	<0.001	0.042	0.038–0.045	0.048	0.995	0.994	0.994	0.995
Two-factorial M.	308.46 (103)	2.99	<0.001	0.031	0.027–0.035	0.039	0.997	0.997	0.996	0.997
Bifactorial M.	189.76 (88)	2.15	<0.001	0.024	0.019–0.028	0.032	0.999	0.998	0.998	0.999

^a RMSEA = root mean squared error of approximation; ^b RMSEA (95% CI) = 95% confidence interval of root mean squared error of approximation; ^c SRMR = standardized root mean square residual; ^d CFI = comparative fit index; ^e TLI = Tucker–Lewis index; ^f NFI = the Normed Fit Index, ^g IFI = incremental fit index. Notes: Traditionally, the standard cut-offs for RMSEA (<0.06) and CFI, TLI, NFI, and IFI (>0.95) indices were used to indicate acceptable model fit [58]. However, these cut-offs are not validated for WLSMV estimator and therefore we did not interpret them directly but used them for model comparisons.