TABLE 2.

Fit indices for the filial piety scale models.

	χ2(df)	CFI	TLI	SRMR	RMSEA [90% CI]	Model (M) comparison	Δχ2(df)	ΔCFI
Baseline model
Singapore	222.595 (101)***	0.913	0.897	0.086	0.073 [0.060, 0.087]	–	–	–
Australia	208.383 (101)***	0.892	0.868	0.089	0.077 [0.062, 0.092]	–	–	–
M1: configural	430.978 (202)***	0.905	0.887	0.086	0.053 [0.046, 0.060]	–	–	–
M2: metric	460.383 (216)***	0.898	0.887	0.088	0.053 [0.044, 0.052]	2 vs.1	29.405 (14)*	0.007
M3: scalar	690.069 (230)***	0.809	0.803	0.090	0.070 [0.064, 0.076]	3 vs. 2	229.686 (14)***	0.089
M3b: partial scalar	477.849 (224)***	0.894	0.887	0.090	0.053 [0.047, 0.060]	3b vs. 2	17.466 (8)*	0.004

In the Baseline Model, error terms of items 1 and 5, and items 4 and 16 were covaried as free parameters in each group. In Model 3b, intercepts of items 2, 3, 9, 10, 14, and 16 were freely estimated. *p < 0.05; ***p < 0.001.