Robert Urban posted on Thursday, October 23, 2008 - 10:32 am
I am performing a CFA analysis on a scale consist of 68 items and 13 factors and 2 second ordered factors (N=240). I have found that CMIN/df(=2.43) is in the optimal range, RMSEA is acceptable more or less (0.75 [0.072-0.077]), however CFI (=0.77) and nNFI(=0.77) are not acceptable at all. I do not know how to interpret this type of discrepancy. What do you think about the second step. Should I perform an exploratory factor analysis or modify the model until I found better fit indices? Thank you very much in advance.