Hi,
Can anybody tell me how to run a chi square difference test in SmartPLS?
Thanks
Andrea
Chi Square Difference test!
-
- PLS Junior User
- Posts: 4
- Joined: Thu Jun 15, 2006 7:00 pm
- Real name and title:
- ghozali
- PLS Expert User
- Posts: 39
- Joined: Sat Oct 15, 2005 1:18 am
- Real name and title: Prof. Imam Ghozali, Ph.D
- Location: Indonesia
Hi,
As long as I knew, PLS did not have Chi-square test because it use different methodology compared to covarince based SEM (Amos or Lisrel). PLS based on variance based. What PLS do is just to find the best relationship between variable based on regression (t statistics) and do not justified whether there is casual relationship between variables (soft modeling)
As long as I knew, PLS did not have Chi-square test because it use different methodology compared to covarince based SEM (Amos or Lisrel). PLS based on variance based. What PLS do is just to find the best relationship between variable based on regression (t statistics) and do not justified whether there is casual relationship between variables (soft modeling)
Faculty of Economics, Diponegoro University
Jl. Erlangga Tengah 17 Semarang, Indonesia
ghozali_imam@yahoo.com
Jl. Erlangga Tengah 17 Semarang, Indonesia
ghozali_imam@yahoo.com
- Diogenes
- PLS Super-Expert
- Posts: 899
- Joined: Sat Oct 15, 2005 5:13 pm
- Real name and title:
- Location: São Paulo - BRAZIL
- Contact:
Hi,
in covariance based SEM the main purpose is to minimize the discrepancy function "F" (difference between observed and reproduzed covariance matrix).
In CB-SEM we have ==> Chi-square = (N-1)*F
Others fit indexes are based in chi-square too.
These fit indexes are assessment of how well the model reproduces the observed covariance matrix.
PLS is "partial", then this fit indexes aren´t useful.
in covariance based SEM the main purpose is to minimize the discrepancy function "F" (difference between observed and reproduzed covariance matrix).
In CB-SEM we have ==> Chi-square = (N-1)*F
Others fit indexes are based in chi-square too.
These fit indexes are assessment of how well the model reproduces the observed covariance matrix.
PLS is "partial", then this fit indexes aren´t useful.
Prof. Dr. Diogenes de Souza Bido