Table 2.
Model Fit |
Model Comparison |
||||||||
---|---|---|---|---|---|---|---|---|---|
Models(M) | SBχ2 | df | RMSEA (90% CI) | CFI | TLI | SRMR | ΔM | ∆SBχ2(∆df) | ΔCFI |
Measurement models | |||||||||
Model in-person | 585.206 | 168 | 0.058 (0.053–0.063) | 0.932 | 0.915 | 0.046 | − | − | |
Model flexible | 323.936 | 168 | 0.061 (0.051–0.071) | 0.915 | 0.894 | 0.061 | − | − | |
M1: Configural | 913.868 | 336 | 0.059 (0.054–0.063) | 0.928 | 0.910 | 0.050 | − | − | |
M2: Metric | 926.215 | 350 | 0.057 (0.053–0.062) | 0.928 | 0.914 | 0.053 | M1-M2 | 180.347(16)ns | 0 |
Structural models | |||||||||
S4: Unconstrained structural effects across groups with metric invariance | 954.667 | 361 | 0.057 (0.053–0.062) | 0.926 | 0.914 | 0.058 | − | − | |
S5: Constrained structural effects across groups with metric invariance | 926.217 | 350 | 0.057 (0.053–0.062) | 0.928 | 0.914 | 0.053 | S4-S5 | 280.487(11)ns | −0.002 |
Note. At each step in the sequence of invariance tests, all earlier constraints remained in place. df = degrees of freedom; RMSEA = root-mean-square error of approximation; CFI = comparative fit index; TLI = Tucker–Lewis index; SRMR = standardized root-mean-squared residual.