Dear dr Dhaval Maheta, I want to ask you about the research that I'm doing. I'm analyzing the user experience of a website using HEART Metrics ((with 5 variables : Happiness (x1), Engagement (x2), Adoption (x3), Retention (x4), Task Success (x5)). I want to do a validity test and a reliability test, then calculate it with IPA. How's the SmartPLS doing? I don't think I have the y variable, can it be done without the y variable?
Dear Dr... may I know who you used PLS Algorithm instead of Consistent PLS Algorithm? Given that all your constructs are reflective. Also, may I know if if 'outer loading' is the same as 'factor loading'?
You are right when we are sure that all our statements are reflective we sud go for consistent pls but in case we are dicey. Factor loadings are same as outer loadings
@@DhavalSaifaleeAaryash Thank you, Dr. But, is it absolutely unacceptable for us to use Pls algorithm (instead of Consistent Pls algorithm) for an all-reflective model?
@@DhavalSaifaleeAaryash Thank you for the confirmation. It's good to know that it's not wrong to still use the regular Pls-Algorithm for an all-reflective model. Because everywhere I read, it's always suggested that we use Consistent Pls for an all-reflective model. Thanks, Dr.
Dear Dr.. regarding HTMT that is above 0.85 ... you have mentioned that the only remedy is to refine the questions in the questionnaire. However, this is too late if we find out after data has already been collected. We cannot refine the questions in the questionnaire anymore. What is the implication of high HTMT on the analysis result if we proceed with the analysis (since there is nothing else we can do about it?) Are there any other remedies at this point? Thank you, Dr.
Dear dr Dhaval Maheta, I want to ask you about the research that I'm doing. I'm analyzing the user experience of a website using HEART Metrics ((with 5 variables : Happiness (x1), Engagement (x2), Adoption (x3), Retention (x4), Task Success (x5)).
I want to do a validity test and a reliability test, then calculate it with IPA. How's the SmartPLS doing?
I don't think I have the y variable, can it be done without the y variable?
Dear Dr... may I know who you used PLS Algorithm instead of Consistent PLS Algorithm? Given that all your constructs are reflective. Also, may I know if if 'outer loading' is the same as 'factor loading'?
You are right when we are sure that all our statements are reflective we sud go for consistent pls but in case we are dicey. Factor loadings are same as outer loadings
@@DhavalSaifaleeAaryash Thank you, Dr. But, is it absolutely unacceptable for us to use Pls algorithm (instead of Consistent Pls algorithm) for an all-reflective model?
@@joelkeats7192 not at all
@@DhavalSaifaleeAaryash Thank you for the confirmation. It's good to know that it's not wrong to still use the regular Pls-Algorithm for an all-reflective model. Because everywhere I read, it's always suggested that we use Consistent Pls for an all-reflective model. Thanks, Dr.
Dear Dr.. regarding HTMT that is above 0.85 ... you have mentioned that the only remedy is to refine the questions in the questionnaire. However, this is too late if we find out after data has already been collected. We cannot refine the questions in the questionnaire anymore. What is the implication of high HTMT on the analysis result if we proceed with the analysis (since there is nothing else we can do about it?) Are there any other remedies at this point? Thank you, Dr.
If we run with high HTMT, your VIF value will be high as divergent validity is lost and our model will not have correct estimates.
@@DhavalSaifaleeAaryash How bad is it if HTMT is over 0.9 but still below 1.0? Is it acceptable if F-L criterion is within limits? Thank you Dr.
@@joelkeats7192 it raises a big concern over authenticity of data