Hi,
How do you convert the results of the model to a 0-100 scale.
Thanks
Leo
Converting to 0-100 scale
-
Diogenes
- PLS Super-Expert
- Posts: 899
- Joined: Sat Oct 15, 2005 5:13 pm
- Real name and title:
- Location: São Paulo - BRAZIL
- Contact:
Hi,
Using SmartPLS is easy:
- transform all indicators to 0-100 scale
- run PLS algorithm
- LV unstandardized scores will be computed in the same scale (0-100) - ready in the output
- LV Index value will be the mean value of the LV unstandardized scores
The standardization process is explained in TENENHAUS, Michel et al. PLS path modeling. Computational Science and Data Analysis. v.48, p.159-205, 2005. Specially page 167.
In SmartPLS we have in the output the "Measurement model" which could be understood as "relative weights", then LV scores are computed as usually is (LV score = b1 *X1 + b2 *X2 + b3 *X3...), but the sum of this weights is equal to one. This is done to have LV scores in the same unit of the indicators.
A kind of LV Index value is used in FORNELL, Claes et al. The American Customer Satisfaction Index: nature, purpose, and findings. Journal of Marketing. v.60, n.4, p.7-18, 1996.
ACSI = {E[LV] - Min[LV]} / {Max[LV] - Min[LV]} * 100
Where: Min[LV] is the LV score when all indicators of this LV are in their minimal values
Max[LV] is the LV score when all indicators of this LV are in their maximal values
Finally, remember that all outer weights must be positive.
Best Regards.
Bido
Using SmartPLS is easy:
- transform all indicators to 0-100 scale
- run PLS algorithm
- LV unstandardized scores will be computed in the same scale (0-100) - ready in the output
- LV Index value will be the mean value of the LV unstandardized scores
The standardization process is explained in TENENHAUS, Michel et al. PLS path modeling. Computational Science and Data Analysis. v.48, p.159-205, 2005. Specially page 167.
In SmartPLS we have in the output the "Measurement model" which could be understood as "relative weights", then LV scores are computed as usually is (LV score = b1 *X1 + b2 *X2 + b3 *X3...), but the sum of this weights is equal to one. This is done to have LV scores in the same unit of the indicators.
A kind of LV Index value is used in FORNELL, Claes et al. The American Customer Satisfaction Index: nature, purpose, and findings. Journal of Marketing. v.60, n.4, p.7-18, 1996.
ACSI = {E[LV] - Min[LV]} / {Max[LV] - Min[LV]} * 100
Where: Min[LV] is the LV score when all indicators of this LV are in their minimal values
Max[LV] is the LV score when all indicators of this LV are in their maximal values
Finally, remember that all outer weights must be positive.
Best Regards.
Bido