Skip to main content
. Author manuscript; available in PMC: 2018 Jun 1.
Published in final edited form as: Psychol Methods. 2017 Jun;22(2):361–381. doi: 10.1037/met0000145

Table 3.

The Results of Frequentist MFIs, LRT, and Bayesian MCC in the Empirical Study.

RMSEA SRMR CFI TLI Chisq DF Di3 Pv
Set 1
 M1 0.097 0.075 0.815 0.772 771.17 74
 M2 0.084 0.063 0.865 0.826 580.07 71
Set 2
 M1 0.089 0.074 0.831 0.792 667.81 74
 M2 0.080 0.064 0.868 0.831 533.53 71
 M3 0.048 0.031 0.959 0.939 255.27 61 278.26 < 10−6
BF BIC DIC pD LOO
Set 2
 M1 37405.78 37228.92 44.78 37189.57
 M2 > 1016 37292.43 37101.87 47.13 37060.17
 M3 > 108 37033.59 36805.07 57.53 36756.32

Note. The Bayes factor shown are the ratio between the posterior probabilities of the model for the previous line and that for the current line. DF – degree of freedom of the chi-square statistic; Diff – the difference between the chi-square statistic in the current row and the previous row; Pv – the p-value of the chidiff test; pD – effective number of parameters. The LRT is between M2 and M3.