Cutoff values for each of the model fit indices

[123 456]

I am currently finalizing a manuscript. I think I need some advice in terms of the cut off values for each of the model fit indices. 

I  know Cho et al. (2020) described that when N > than 100, GFI needs greater or equal to 0.93; SRMR less than or equal to 0.08. 

What about about FIT,  AFIT, FITs ? Is there any suggested cut-off value?  

GSCA pro also provides a few new model fit indices OPE, OPEs, and OPEm, which Hwang and Takane 's (2014) book did not mention those? Is it necessary to report OPE, OPEs, and OPEm in the model fit? Or is it enough just to report GFI, SRMR, FIT, and AFIT as the indication of acceptable model fit. 

 Thank you so much for your help ! 

[Gyeongcheol Cho]

Hi,

Thanks for your interest in our software. 

Similar to the R squared in the regression model, FIT measures in GSCA cannot have their cutoff values. Instead, the values of FIT and FITs are interpretable; the FIT value informs how much proportion of variance of variables is explained by the GSCA model, whereas the FITs value indicates how much proportion of variance of components is explained by the structural model. AFIT in GSCA corresponds to the adjusted R squared in the regression model. It is for model comparison.

OPE is the cross-validation index for GSCA. As this index aims to compare competing models in terms of predictive generalizability, you should report its value if you conduct model comparison. Otherwise, you don't need to report it. If your competing models are different in their structural models, you should report the value of OPEs as well. For more detailed information, please refer to Cho et al., (2019)

Reference

Cho, G., Jung, K., & Hwang, H. (2019). Out-of-bag prediction error: A cross validation index for generalized structured component analysis. Multivariate Behavioral Research, 54(4), 505–513. https://doi.org/10.1080/00273171.2018.1540340

Best,

Gyeongcheol