Psychometric evaluation of the traditional Chinese version of PedsQL™ 3.0 cardiac module scale in adolescents with congenital heart disease: reliability, validity, measurement invariance, and adolescent-parent agreement

Table 4 Measurement invariance across self-reports and parent proxy-reports of the traditional Chinese PedsQL-CM

Model	\({\upchi }^{2}\)	df	CFI	RMSEA	90% CI	SRMR	Comparison (Δ)	ΔCFI	ΔRMSEA	ΔSRMR
M1	485.6***	408	0.972	0.034	(0.020, 0.046)	0.073	–	–	–	–
M2	505.6***	425	0.971	0.034	(0.020, 0.045)	0.079	M1 vs M2	0.001	0.000	0.006
M3	511.5***	429	0.970	0.035	(0.021, 0.046)	0.086	M2 vs M3	0.001	0.001	0.007
M4	544.9***	446	0.964	0.037	(0.025, 0.048)	0.087	M3 vs M4	0.006	0.002	0.001
M5	554.9***	450	0.962	0.038	(0.026, 0.048)	0.089	M4 vs M5	0.002	0.001	0.002

M1: Configural invariance
M2: Metric invariance of the first-order factors
M3: Metric invariance of the first- and second-order factors
M4: Scalar invariance of the first-order factors
M5: Scalar invariance of the first- and second-order factors
^***: The chi-square test was significant with p < 0.001
The cut-off criteria were ≤ .10 for ΔCFI, ≤ .15 for ΔRMSEA, ≤ .03 for ΔSRMR testing metric invariance, and ≤ .015 for ΔSRMR testing scalar invariance

ISSN: 1477-7525