Thursday, January 21, 2021
Using indicator correlation fit indices in PLS-SEM: Selecting the algorithm with the best fit
The article below explains how one can use indicator correlation fit indices for selecting the analysis algorithm with the best fit in the context of structural equation modeling via partial least squares (PLS-SEM).
Kock, N. (2020). Using indicator correlation fit indices in PLS-SEM: Selecting the algorithm with the best fit. Data Analysis Perspectives Journal, 1(4), 1-4.
A link to a PDF file is available ().
Abstract:
Upon completion of a PLS-SEM analysis, one can obtain the model-implied indicator correlation matrix and compare it with the actual indicator correlation matrix. The latter is obtained directly from the data being analyzed. Indicator correlation fit indices are quantifications of the differences among these two matrices. Our focus in this paper is on the use of indicator correlation fit indices in PLS-SEM for selecting the analysis algorithm with the best fit.
Subscribe to:
Post Comments (Atom)
2 comments:
Hi, Ned.
Acceptable model fit and quality indices discussed in the article are as follows: 1) Values of SRMR and SMAR are less than 0.1; 2) P value associated with SChS is .05 or lower; and 3) Values of STDCR and STDSR are .70 or lower. For the sake of discussion, let's assume that after the SEM analysis we found out that such model fit and quality indices are not acceptable. This means that there is a large difference between the model-implied indicator correlation matrix and the actual indicator correlation matrix. In simple terms, the sample correlation matrix does not fit the model correlation matrix. One possible reason for the misfit might be the misspecification of the research model. For the future version of WarpPLS, do you have plans to add some features to the software? For example, providing additional outputs on "modification indices" which can be used by the researcher/analyst to modify or improve his/her model.
Modification indices are reported in many CB-SEM software (e.g., Amos). I think modification indices are also appropriate in your WarpPLS since the model fit and fit indices (i.e., SRMR, SMAR, SChS, STDCR and STDSR) that you discussed are based on the model-implied indicator correlation matrix and the actual indicator correlation matrix.
Thanks for the suggestion.
Post a Comment