Hello everyone,
so I have developed a model in which all but two variables are reflective. For the reflective constructs I am aware of the tests of reliability and validity as well on how to report results. My question is how to ensure validity and reliability of formative constructs, and how to report these.
My predictor construct is a formative construct with 12 items in which weights are not significant towards the variable itself. I have seen that this is not a problem, and unlike reflective constructs I shouldn't proceed in deleting items that are non-significant.
I also have a second order construct (formative-formative), how do I also determine validity and reliability of this variable and its sub-variables?
Finally, how do I present results in a uniform manner. I know that when all constructs are reflective a matrix is developed with sqrt of AVE in the diagonal and construct correlations below the diagonal. Also CR is used. What do I do in the case of formative measures for which CR and AVE are 0?
Thanks in advance :)
Validity and reliability of formative construct
-
- PLS Expert User
- Posts: 51
- Joined: Tue Nov 04, 2014 5:27 am
- Real name and title: Dr. Bradley Wilson. Senior Lecturer in Advertising. School of Media and Communication. RMIT University.
Re: Validity and reliability of formative construct
Agreed. CR, Alpha and AVE are a nonsense for formative constructs.
The overall quality and predictive ability of the model is best test.
brad
The overall quality and predictive ability of the model is best test.
brad
Bradley Wilson. Ph.D. Senior Lecturer in Advertising. School of Media and Communication. RMIT University. Melbourne, Australia.
SEE: http://www.researchgate.net/profile/Bradley_Wilson2 http://scholar.google.com.au/citations? ... AAAJ&hl=en
SEE: http://www.researchgate.net/profile/Bradley_Wilson2 http://scholar.google.com.au/citations? ... AAAJ&hl=en
Re: Validity and reliability of formative construct
Mr. Brad, I was searching for "Is measuring AVE valid for formative constructs" and came across with your post.bradley7680 wrote:Agreed. CR, Alpha and AVE are a nonsense for formative constructs.
The overall quality and predictive ability of the model is best test.
brad
I would appreciate if you could share a reference paper for your above comment.
Regards,
Merve
- 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:
Re: Validity and reliability of formative construct
No, it's not.
Best
CR
Best
CR
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