Table 3.
Model (Subsamples) | Nine-Item DrVac-COVID19S | 12-Item DrVac-COVID19S | ||
---|---|---|---|---|
Fit Indices | One-Factor | Four-Factor | One-Factor a | Four-Factor a |
Configural (Taiwan vs. China) | ||||
χ2 (df)/p-value | 482.57 (50)/<0.001 | 207.81(40)/<0.001 | 1171.60 (98)/<0.001 | 771.18 (86)/<0.001 |
CFI | 0.993 | 0.997 | 0.984 | 0.990 |
RMSEA | 0.067 | 0.046 | 0.075 | 0.064 |
SRMR | 0.026 | 0.021 | 0.042 | 0.039 |
Loading constrained (Taiwan vs. China) | ||||
Δχ2 (df)/p-value | 153.75 (8)/<0.001 | 32.55 (5)/<0.001 | 207.93 (13)/<0.001 | 182.24 (15)/<0.001 |
ΔCFI | −0.002 | 0.000 | −0.003 | −0.002 |
ΔRMSEA | 0.004 | 0.001 | 0.002 | 0.002 |
ΔSRMR | 0.008 | 0.003 | 0.006 | 0.004 |
Loadings and intercepts constrained (Taiwan vs. China) | ||||
Δχ2 (df)/p-value | 198.23 (8)/<0.001 | 500.43 (5)/<0.001 | 252.63 (10)/<0.001 | 143.47 (7)/<0.001 |
ΔCFI | −0.004 | −0.030 | −0.002 | −0.002 |
ΔRMSEA | 0.006 | 0.037 | 0.003 | 0.002 |
ΔSRMR | 0.009 | −0.008 | 0.011 | 0.043 |
Configural (male vs. female) | ||||
χ2 (df)/p-value | 466.72(50)/<0.001 | 201.80 (40)/<0.001 | 1152.51(98)/<0.001 | 806.23 (86)/<0.001 |
CFI | 0.994 | 0.997 | 0.986 | 0.991 |
RMSEA | 0.065 | 0.046 | 0.074 | 0.066 |
SRMR | 0.029 | 0.018 | 0.047 | 0.041 |
Loading constrained (male vs. female) | ||||
Δχ2 (df)/p-value | 55.59 (8)/<0.001 | 23.07 (5)/<0.001 | 117.06 (13)/<0.001 | 109.59 (15)/<0.001 |
ΔCFI | −0.001 | 0.000 | −0.001 | −0.001 |
ΔRMSEA | −0.001 | −0.001 | −0.001 | −0.002 |
ΔSRMR | 0.007 | 0.003 | 0.026 | 0.027 |
Loadings and intercepts constrained (male vs. female) | ||||
Δχ2 (df)/p-value | 236.36 (8)/<0.001 | 186.10 (5)/<0.001 | 143.41 (10)/<0.001 | 63.20 (7)/<0.001 |
ΔCFI | −0.018 | −0.010 | −0.003 | −0.002 |
ΔRMSEA | 0.009 | 0.016 | 0.001 | 0.000 |
ΔSRMR | −0.023 | −0.015 | −0.010 | −0.021 |
Configural (health vs. non-health) | ||||
χ2 (df)/p-value | 476.22 (50)/<0.001 | 186.48 (40)/<0.001 | 1189.99 (98)/<0.001 | 803.93 (86)/<0.001 |
CFI | 0.994 | 0.998 | 0.985 | 0.990 |
RMSEA | 0.066 | 0.043 | 0.076 | 0.065 |
SRMR | 0.028 | 0.020 | 0.047 | 0.043 |
Loading constrained (health vs. non-health) | ||||
Δχ2 (df)/p-value | 60.75 (8)/<0.001 | 21.24 (5)/<0.001 | 97.28 (13)/<0.001 | 116.93 (15)/<0.001 |
ΔCFI | −0.001 | 0.000 | −0.001 | −0.001 |
ΔRMSEA | −0.001 | 0.000 | −0.002 | −0.001 |
ΔSRMR | 0.001 | 0.001 | 0.001 | 0.001 |
Loadings and intercepts constrained (health vs. non-health) | ||||
Δχ2 (df)/p-value | 105.16 (8)/<0.001 | 388.86 (5)/<0.001 | 165.84 (10)/<0.001 | 72.71 (7)/<0.001 |
ΔCFI | −0.002 | −0.026 | −0.003 | −0.001 |
ΔRMSEA | 0.002 | 0.032 | 0.001 | 0.001 |
ΔSRMR | 0.003 | −0.013 | 0.005 | 0.025 |
CFI = comparative fit index; TLI = Tucker–Lewis index; RMSEA = root mean square error of approximation; SRMR = standardized root mean square residual. Excellent fit values are in bold; i.e., CFI and TLI > 0.95; RMSEA and SRMR < 0.08. Supported measurement invariance values are in bold; i.e., ΔCFI > −0.01; ΔRMSEA < 0.015; ΔSRMR < 0.03 (for factor loading) or < 0.01 (for item intercept). a Using correlated trait correlated method minus one model to control wording effects.