Table 4.
Statistical Summary of Goodness-of-Fit Indices for CFA
| Model Tested | 2 | p-value | AIC | NFI | RFI | NLI | BIC | aBIC | CFI | TLI/NNFI | SRMR | RMSEA | 90%CI |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Three-factor model | 338.49 (df=424) | <0.001* | 2847.1 | 0.921 | 0.937 | 0.957 | 4193.5 | 3983.9 | 0.943 | 0.96 | 0.026 | 0.048 | 0.037–0.059 |
| Two-factor model | 608.46 (df=424) | <0.001* | 3371.7 | 0.981 | 0.919 | 0.912 | 4344.1 | 4140.8 | 0.931 | 0.934 | 0.039 | 0.057 | 0.047–0.068 |
Notes: Goodness of fit criteria: excellent fit: RMSEA ≤0.06; CFI ≥0.95; TLI ≥0.95; SRMR ≤0.08. NFI >0.9; TLI >0.9; CFI >0.9; a lower value in BIC, aBIC or AIC indicates a better fit. *Statistical significance level at p<0.001
Abbreviations: AIC, Akaike information criterion; NFI, normed fit index; RFI, relative fit index; 2, chi-squared computed to indicate the association; BIC, Bayesian information criterion; aBIC, sample-size adjusted BIC; CFI, comparative fit index; TLI, Tucker–Lewis index or NNFI, non-normed fit index; SRMR, standardized root mean squared residual; NLI, nonlinear index; RMSEA, root mean square error of approximation; CI, confidence interval; df, degree of freedom.