Model Fit Indices in CFA
In evaluating model fit, I focused on two commonly used indices: CFI (Comparative Fit Index) and RMSEA (Root Mean Square Error of Approximation).
...Model Fit Indices in CFA
In evaluating model fit, I focused on two commonly used indices: CFI (Comparative Fit Index) and RMSEA (Root Mean Square Error of Approximation).
For CFI, I adopted the threshold of ≥ 0.90 as an acceptable fit and ≥ 0.95 as a good fit, based on Hu and Bentler (1999). For RMSEA, I used ≤ 0.08 as an acceptable fit and ≤ 0.06 as a good fit, also supported by Hu and Bentler (1999), as well as Kline (2016).
In my model, the CFI value was 0.92 and the RMSEA value was 0.07. Based on these thresholds, the model can be considered to have an acceptable fit, although it does not reach the level of a “good” fit.
Personally, I believe that an “almost acceptable” model fit can still be considered valid, especially in social science research where data are often complex. However, researchers should justify their decisions and consider model modification carefully rather than relying solely on fit indices.