I used smartpls and spss to find out the correlations. The smartpls correlations are totally different than the spss. The spss result is more logical. For example, my model is like IV--> different mediators --> DV and one of the mediators is negative emotions which is negatively correlated to IV and DV and other mediators in Spss. However, that is not the case in smartpls. Negative emotions is positively related to most variables. The following is the correlation results of negative emotion with other variables in the model from spss and smartpls:
Spss result
Neg Emotion
-.161*
-.137*
-.303**
1
-.245**
-.220**
0.062
-.224**
-.308**
Smartpls result
NEG emotion
0.89
.382**
.372**
.414**
-.276**
-0.033
0.01
0.007
-0.067
0.016
.348**
0.001
Does anyone knows why there is a big different in spss and smartpls correlations? The change is huge from sig. negative relationship to sig. positive ones.
spss and smartpls huge correlation differences
-
- PLS Junior User
- Posts: 5
- Joined: Thu Sep 05, 2013 8:53 pm
- Real name and title:
- cringle
- SmartPLS Developer
- Posts: 818
- Joined: Tue Sep 20, 2005 9:13 am
- Real name and title: Prof. Dr. Christian M. Ringle
- Location: Hamburg (Germany)
- Contact:
Hi
I check this issue many times and I never had differences in the correlations between SPSS and SmartPLS.
-> Estimate your PLS path model; use the settings in the PLS algorithm to obtain standardized results
-> SmartPLS generates the standardized LV scores using SmartPLS
-> SmartPLS also returns the correlations of these LV scores
-> If you copy and paste the LV scores to SPSS and compute the correlation of the LV scores, they exactly match the SmartPLS outcomes.
Reasons for differences are various but if you follow this procedure and do everything correctly, no differences should occur.
Please also check viewtopic.php?t=2940
Best regards
Christian Ringle
I check this issue many times and I never had differences in the correlations between SPSS and SmartPLS.
-> Estimate your PLS path model; use the settings in the PLS algorithm to obtain standardized results
-> SmartPLS generates the standardized LV scores using SmartPLS
-> SmartPLS also returns the correlations of these LV scores
-> If you copy and paste the LV scores to SPSS and compute the correlation of the LV scores, they exactly match the SmartPLS outcomes.
Reasons for differences are various but if you follow this procedure and do everything correctly, no differences should occur.
Please also check viewtopic.php?t=2940
Best regards
Christian Ringle
Prof. Dr. Christian M. Ringle, Hamburg University of Technology (TUHH), SmartPLS
- Literature on PLS-SEM: https://www.smartpls.com/documentation
- Google Scholar: https://scholar.google.de/citations?use ... AAAJ&hl=de
- Literature on PLS-SEM: https://www.smartpls.com/documentation
- Google Scholar: https://scholar.google.de/citations?use ... AAAJ&hl=de
-
- PLS Junior User
- Posts: 5
- Joined: Thu Sep 05, 2013 8:53 pm
- Real name and title: