Are comparisons of mental disorders between Chinese and German students possible? An examination of measurement invariance for the PHQ-15, PHQ-9 and GAD-7

Table 4 Fit Indices from Comparative Factor Analysis (CFA) and Invariance Analyses between Groups for the PHQ-9

	χ2 (df)	CFI	RMSEA [90% CI]	SRMR	ΔCFI	Δχ2 (df)
model A. one-factor model
German Students	147.851 (27)	.900	.104 [.088, .120]	.051
Chinese Students	101.362 (27)	.951	.082 [.065, .099]	.038
model B. two-factor model
German Students	63.070 (26)	969	.059 [.040, .077]	.033		84.781
Chinese Students	91.453 (26)	.957	.078 [.061, .096]	.037		9.909, p > .001
Multiple groups CFA models
Configural invariance	154.523 (52)	.962	.069 [.057, .082]	.035
Metric invariance	229.833 (61)	.938	.082 [.071, .093]	.082	.024	75.310 (9)
λ₈ free	185.565 (60)	.954	.071 [.060, .083]	.064	.008	31.042 (8)
λ_8, λ₁ free	174.288 (59)	.958	.069 [.057, .081]	.068	.004	19.765 (7)
λ_8, λ_1, λ₃ free	169.246 (58)	.959	.068 [.056, .080]	.061	.002	14.723 (6), p > .001
Scale invariance	302.171 (65)	.913	.094 [.083, .105]	.085	.046	132.925 (7)
τ₈ free	216.478 (64)	.944	.076 [.065, .087]	.063	.015	47.232(6)
τ₈, τ₄ free	191.586 (63)	.953	.070 [.059, .082]	.061	.006	22.340 (5)
τ₈, τ₄, τ₁ free	179.934 (62)	.957	.068 [.056, .079]	.061	.002	10.688 (4), p > .001

Note. PHQ-9 = Patient Health Questionnaire-9 Depression Symptoms; Item 1 = ‘lack of interest’; item 3 = ‘sleep difficulties’; item 4 = ‘feeling tired or having little energy’; item 8 = ‘moving or speaking slowly, or fretful.’ All χ2 tests and Δχ2 were significant, p < .001

ISSN: 1471-244X