Model Fit

Frequently asked questions about PLS path modeling.
Post Reply
Rama
PLS Junior User
Posts: 3
Joined: Mon Feb 01, 2016 1:35 pm
Real name and title: Mr. Ramakrishna Permal Naidoo

Model Fit

Post by Rama » Sun May 15, 2016 6:34 am

For my research I am using SmartPLS version 3.2.4 for data analysis. I have obtained the following model fit values:
Saturated Model Estimated Model
SRMR 0.068 0.068
d_ULS 4.552 4.552
d_G 5.969 5.969
Chi-Square 2,378.253 2,378.253
NFI 0.659 0.659
rms Theta 0.157
I have some difficulties to interpret them, in particular the geodesic discrepancy dG, the unweighted least squares discrepancy dULS and the RMStheta. I would highly appreciate to have an expert’s view on these data and some advice on their interpretation.
Ramakrishna Permal Naidoo

User avatar
Hengkov
PLS Super-Expert
Posts: 1619
Joined: Sun Apr 24, 2011 10:13 am
Real name and title: Hengky Latan
Location: AMQ, Indonesia
Contact:

Re: Model Fit

Post by Hengkov » Thu May 26, 2016 6:55 am

Necessary caution in interpreting the chi-square value, NFI, RMS Theta etc. However GoF index of CB-SEM is different to be applied in PLS-SEM.

Rama
PLS Junior User
Posts: 3
Joined: Mon Feb 01, 2016 1:35 pm
Real name and title: Mr. Ramakrishna Permal Naidoo

Re: Model Fit

Post by Rama » Thu May 26, 2016 2:04 pm

I agree caution should be taken in interpreting the chi-square value, NFI, RMS Theta etc in SmartPLS. However, SRMR, seems to be a reasonable index to evaluate model fit.

jmbecker
SmartPLS Developer
Posts: 743
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Model Fit

Post by jmbecker » Mon Aug 08, 2016 12:44 pm

You will find some information that we have assembled here: https://www.smartpls.com/documentation/fit
Dr. Jan-Michael Becker, University of Cologne, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Ja ... v=hdr_xprf
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de

agalvez
PLS Expert User
Posts: 32
Joined: Mon Jul 04, 2016 10:17 pm
Real name and title: Alex

Re: Model Fit

Post by agalvez » Fri Apr 14, 2017 11:56 am

Dear Rama

I see in your model that the saturated and estimated models values are equal. Do you know what this means?

Thank you

jmbecker
SmartPLS Developer
Posts: 743
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Model Fit

Post by jmbecker » Fri Apr 21, 2017 3:01 pm

If you model is fully saturated the estimated and saturated model will necessarily be equivalent.
Dr. Jan-Michael Becker, University of Cologne, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Ja ... v=hdr_xprf
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de

Post Reply