Hello,
I would like to use the HTMT criterion to assess discriminant validity. It is clear how to use it with lower constructs, but how could I use it on higher(2nd) order constructs?
Thanks in advance!
HTMT with higher order constructs?
-
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: HTMT with higher order constructs?
Hi
It works exacly the same way as for any other reflective construct.
The questions remain:
Why do you use higher order constructs in PLS-SEM when you have no limitations in the complexity of your structural model and can estimate all relationships from the lower constructs to the target constructs directly.
What is a higher oder construct?
Do you expect discriminant validity between the higher and the lower order constructs?
Do you expect discriminant validity between the lower order constructs?
For sure, you expect discriminant validity between the higher order construct and the other construct in the PLS path model.
It works exacly the same way as for any other reflective construct.
The questions remain:
Why do you use higher order constructs in PLS-SEM when you have no limitations in the complexity of your structural model and can estimate all relationships from the lower constructs to the target constructs directly.
What is a higher oder construct?
Do you expect discriminant validity between the higher and the lower order constructs?
Do you expect discriminant validity between the lower order constructs?
For sure, you expect discriminant validity between the higher order construct and the other construct in the PLS path model.
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
-
mdragt
- PLS Junior User
- Posts: 6
- Joined: Thu Oct 30, 2014 7:04 pm
- Real name and title: Marian Dragt MBA - PhD Student
Re: HTMT with higher order constructs?
Thanks for your answer Christian! I use a 2nd order construct as the construct has 4 dimensions (based on literature), but I will also check these 4 dimensions as lower order constructs, thanks for the suggestion. But, using the 2nd order construct (independent construct), I can ignore discriminant validity between the lower order constructs, and only focus on the 2nd order construct and the dependent construct?
Re: HTMT with higher order constructs?
Hi..
This issue is still confusing to me , as I have read a post of Dr.Ringle as under
Re: convergent and divergent validity
Postby cringle » Sun Nov 30, 2014 12:26 pm
HTMT is in SmartPLS 3, just check the discriminant validity report.
In 2nd order models, you don't care about discriminant validity problems between the lower order constructs and/or the higher order construct.
Best
CR
Prof. Dr. Christian M. Ringle
http://www.tuhh.de/hrmo/team/prof-dr-c-m-ringle.html
--> More info on PLS-SEM: https://www.smartpls.com/documentation/index
User avatar
cringle
SmartPLS Developer
SO I AM NOT SURE WHAT TO DO STILL " SHOULD I IGNORE ALL SECOND ORDER CONSTRUCTS ALONG WITH THEIR RESPECTIVE FIRST ORDER DIMENSIONS in HTMT..
PLEASE CLARITY FURTHER
This issue is still confusing to me , as I have read a post of Dr.Ringle as under
Re: convergent and divergent validity
Postby cringle » Sun Nov 30, 2014 12:26 pm
HTMT is in SmartPLS 3, just check the discriminant validity report.
In 2nd order models, you don't care about discriminant validity problems between the lower order constructs and/or the higher order construct.
Best
CR
Prof. Dr. Christian M. Ringle
http://www.tuhh.de/hrmo/team/prof-dr-c-m-ringle.html
--> More info on PLS-SEM: https://www.smartpls.com/documentation/index
User avatar
cringle
SmartPLS Developer
SO I AM NOT SURE WHAT TO DO STILL " SHOULD I IGNORE ALL SECOND ORDER CONSTRUCTS ALONG WITH THEIR RESPECTIVE FIRST ORDER DIMENSIONS in HTMT..
PLEASE CLARITY FURTHER