How to Evaluate Model Fit in AMOS: Understanding Fit Indices
- Evaluating model fit in AMOS ensures that your structural equation model (SEM) accurately represents the observed data. Fit indices serve as statistical benchmarks for assessing the quality of your model. Here is a breakdown of key fit indices in AMOS:
- Chi-Square (χ²): It tests the model’s goodness of fit by comparing the observed covariance matrix to the model’s predicted covariance matrix. A non-significant chi-square (p > 0.05) indicates a good fit but is sensitive to sample size.
- Comparative Fit Index (CFI): CFI measures how your model fits the data better than a null model (no relationships). A CFI value near or exceeding 0.95 suggests a good fit.
How to Evaluate Model Fit in AMOS: Understanding Fit Indices - Root Mean Square Error of Approximation (RMSEA): RMSEA quantifies the discrepancy between your model and the population covariance matrix. A smaller RMSEA (e.g., < 0.08) indicates a better fit, with a 90% confidence interval providing additional insights.
- Standardized Root Mean Square Residual (SRMR): SRMR assesses the model’s discrepancy based on residuals. A lower SRMR (e.g., < 0.08) indicates a better fit.
- Normed Fit Index (NFI) and Tucker-Lewis Index (TLI): NFI and TLI measure the proportionate improvement in fit over a null model. Values near or above 0.90 indicate a good fit.
- Goodness of Fit Index (GFI) and Adjusted Goodness of Fit Index (AGFI): These indices estimate the explained variance in the sample covariance matrix. Values closer to 1 suggest a better fit (e.g., > 0.90).
Interpret these fit indices collectively, considering theoretical context and model complexity. If they indicate poor fit, modify the model and iterate until acceptable fit indices are achieved. Document your findings in your research report, emphasizing the fit indices to communicate the model’s quality to your audience.
