After deletion of 3 item from a 24 item scale on a sample of 545, the model fit indices are GFI- .900, CFI- .925, RMSEA- .07, SRMR- .04. on further deletion of items the cfi and gfi doesn't increase. i read an article that cfi and gfi should be more than .95. does it means that my model isn't a good one? many people also report that the threshold is .90. Can somebody guide?
Brown (2006) Confirmatory factor analysis for applied research, p.87:
"CFI and TLI values below .90 should lead the researcher to strongly suspect (reject) the solution, CFI and TLI values in the range of .90–.95 may be indicative of acceptable model fit".
You may compare your results with the following Cut off Values for good fit
Model Chi-Square p-value > 0.05
GFI ≥ 0.95 AGFI ≥ 0.90
NFI ≥ 0.95 NNFI ≥ 0.95
CFI ≥ 0.90
RMSEA < 0.08 (closer to zero)
SRMR < 0.08 (closer to zero)
AVE > 0.5
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling