Urgent~PLS Consistent and Bootstrapping Issue
Posted: Sat Jun 20, 2020 7:07 am
Dear all,
I am a new beginner in doing research work using SMART PLS and currently I am facing issue with the interpretation of results.
My model is Technology Acceptance Model 3 consisting of 9 latent variables as shown in figure below.
Since, it's reflective model so I applied PLSc to evaluate the model.
I have carried out two times of data collection in which first was for training session and second was for implementation session. So i have two sets of datasets and I ran the PLSc for both the datasets separately for the evaluation of measurement model and then I ran bootstrapping for the evaluation of structural model to know the significance.
However, I am now dealing with issues that are:
(a) In measurement model,
- some of the cross loadings are higher than the others loadings constructs when going down or across the column and row to determine the discriminant validity. So how could I justify based on the values obtained? Even they have established good reliability prior to that.
(b) In structural model,
- the results obtained for T-statistics and p value are strange because some were missing and some were too high (as attached below)
- what should I do to make it right or how should I justify for that?
Thank you in advance!!!
I am a new beginner in doing research work using SMART PLS and currently I am facing issue with the interpretation of results.
My model is Technology Acceptance Model 3 consisting of 9 latent variables as shown in figure below.
Since, it's reflective model so I applied PLSc to evaluate the model.
I have carried out two times of data collection in which first was for training session and second was for implementation session. So i have two sets of datasets and I ran the PLSc for both the datasets separately for the evaluation of measurement model and then I ran bootstrapping for the evaluation of structural model to know the significance.
However, I am now dealing with issues that are:
(a) In measurement model,
- some of the cross loadings are higher than the others loadings constructs when going down or across the column and row to determine the discriminant validity. So how could I justify based on the values obtained? Even they have established good reliability prior to that.
(b) In structural model,
- the results obtained for T-statistics and p value are strange because some were missing and some were too high (as attached below)
- what should I do to make it right or how should I justify for that?
Thank you in advance!!!