How to model using PLS

[venugopalc]

This is my problem:

I have a set of 21 critical success factors (CSFs)for the success of a software implementation. I have also a set of 20 factors that measure the users feedback on "success" (MES)of the implementation. I am also asking an overall question " are you satisfied with the implementation? "

I want to use PLS to identify the relationship between the predictive CSFs and the resultant measures of success(MES). I want to be able to build a model that would predict success based on the CSFs and test the validity of the model.

Could someone help by answering the following :

1) Can PLS be used for this?

2) How can this model be built?

3) How would the results of a PLS model help me devlop a predictive model for Software implementaion success?

[viswadatta]

Yes you can use PLS

The indicators are to be decided first along with the constructs based on theory.

Collect the data in metric scales only.

YOu can use any PLS software to test the model. Decide the sample size using g-power 3.0.6 software (it is a free download).

[venugopalc]

Could you give me some more details of this software 3.0.6 ?

[viswadatta]

Check out the following link
http://www.psycho.uni-duesseldorf.de/ab ... d-register

[venugopalc]

Thanks.