TABLE 5.
Model fit indices of the theoretical model (see Figure 1).
| Models | Threshold used values (see Kline, 2013; Abad et al., 2015) | Independence model | Theoretical model |
| χ2 | – | 5019.782 | 611.099 |
| p | – | <0.0001 | <0.0001 |
| Normed χ2 | – | 10.795 | 1.424 |
| RMSEA | <0.05 | 0.121 (0.118–0.124) | 0.025 (0.021–0.030) |
| AGFI | >0.9 | 0.414 | 0.936 |
| PRATIO | >0.9 | 1 | 0.923 |
| PCFI | >0.8 | ∼0 | 0.886 |
| CFI | >0.95 | ∼0 | 0.960 |
| TLI | >0.95 | ∼0 | 0.957 |
| IFI | >0.95 | ∼0 | 0.960 |
| BIC | – | 5221.369 | 1046.786 (4174.583**) MRR = 79.95% |
| AIC | – | 5081.782 | 745.099 (4336.683**) MRR = 85.33% |
| CAIC | – | 5252.369 | 1113.786 (4138.583**) MRR = 78.79% |
RMSEA, root mean square error of approximation; AGFI, adjusted goodness of fit index; PRATIO, parsimony ratio; PCFI, parsimony adjustment to the comparative fit index; CFI, comparative fit index; TLI, Tucker–Lewis coefficient; IFI, incremental fit index; BIC, Bayes information criterion; AIC, Akaike information criterion; CAIC, consistent Akaike information criterion; MRR, Mismatch Reduction Ratio estimated using equation [1]. **These values are the differences between independence model and theoretical model.