Below are some of my notes on the key cutoff values for confirmatory factor analysis (CFA). They are based on this PDF by Professor Lesa Hoffman .
-
Smaller chi square, and larger
$p$-value are better (we don’t want the p-value to be smaller than .05) -
SRMR (smaller is better; <.08 -> good fit)
-
RMSEA, smaller is better. <.05 or <.06 -> good; .05~.08 -> acceptable; .08~.10 -> mediocre; >.10 -> unacceptable
-
CFI (Comparative Fit Index): bigger is better. >.90 -> acceptable; >.95 good
-
TLI (Tucker-Lewis Index): bigger is better, >.95 -> good
Last modified on 2022-07-14