In AMOS software, the PNFI index (Parsimonious Normed Fit Index) is an index to evaluate model fit in structural data analysis (SEM). It is used to measure the degree of fit between the model and the data.
PNFI is a shortened version of NFI (Normed Fit Index). The NFI index was used before, but it was somewhat wide-ranging and confusing. Therefore, the PNFI was created to provide a simpler and easier to understand index.
The PNFI value ranges from 0 to 1. A value close to 1 shows that the model is good at fitting the data. However, there is no specific threshold to evaluate the fit of a PNFI-based model without comparison to a reference model or the use of relative distributions. According to Hair et al. (1998), refer to the studies of Yao et al. (2007) and Monteiro et al. (2022), if the PNFI index is ≥ 0.6 the model fits well with the research data set.
Note that PNFI is not as popular and less used than other popular indices such as CFI (Comparative Fit Index) and RMSEA (Root Mean Square Error of Approximation). However, it can still be used as an additional means to evaluate model fit in SEM.
References
Hair, J., Andreson, R., Tatham, R., & Black, W. (1998). Multivariate data analysis. 5th (ed) Prentice-Hall Inc. United States of America.
Monteiro, A. P., Vale, J., Leite, E., Lis, M., & Kurowska-Pysz, J. (2022). The impact of information systems and non-financial information on company success. International Journal of Accounting Information Systems, 45, 100557.
Yao, S., Chen, H., Jiang, L., & TAM, W. C. (2007). Replication of factor structure of Wechsler Adult Intelligence Scale‐III Chinese version in Chinese mainland non‐clinical and schizophrenia samples. Psychiatry and Clinical Neurosciences, 61(4), 379–384.
