Model Fit Indices

Measures of goodness of fit typically summarize the discrepancy between observed values and the values expected under the model in question.

Model fit indices. The goodness of fit of a statistical model describes how well it fits a set of observations. Fit indices for structural equation modeling author. Such measures can be used in statistical hypothesis testing. Cfi comparative fit index a revised form of nfi.

An nfi of 95 indicates the model of interest improves the fit by 95 relative to the null model. Note that if the model is saturated or just identified then most but not all fit indices cannot be computed because the model is able to reproduce the data. Absolute fit indices compare the tested model to the saturated model. The more complex the model the lower the fit index.

Compares the fit of a target model. Relative fit indices also called the incremental fit includes a factor that represents deviations from a null model. The adjustments are to penalize models that are less parsimonious so that simpler theoretical processes are favored over more complex ones. These include the goodness of fit index gfi.

This null model almost always contains a model in which all of the variables are uncorrelated and as a result has a very large chi square. Is the model good with this indices. If the model is acceptable researchers then establish whether specific paths are significant. Parsimony corrected fit indices are relative fit indices that are adjustments to most of the fit indices mentioned above.

Sometimes the nnfi is called the tucker lewis index tli nfi 0 95 nnfi 0 95. Relative fit indices also called incremental fit indices and comparative fit indices compare the chi square for the hypothesized model to one from a null or baseline model. Jöreskog sörbom 1986 its adjusted form agfi and the aforementioned. The major reason for computing a fit index is that the chi square is statistically significant but the reseacher still wants to claim that the model is a good fitting model.

Acceptable fit indices do not imply the relationships are strong. For the approximate fit indices such as srmr and nfi you may directly look at the outcomes of a pls sem or plsc sem model estimation i e the results report and these criteria s values with a certain threshold e g srmr 0 08 and nfi 0 90. Nnfi is preferable for smaller samples. Not very sensitive to sample size.