HTMT discriminant validity

Questions about the implementation and application of the PLS-SEM method, that are not related to the usage of the SmartPLS software.
StevenGregory
PLS Junior User
Posts: 6
Joined: Sun Jan 13, 2019 12:49 pm
Real name and title: Mr Steven Gregory Bamba

Re: HTMT discriminant validity

Post by StevenGregory »

ohh okey so Sir is it alright if i only using the cross loading test for the discriminant validity test?
because my cross loading test values are ok/valid as required
Thank you in advance Sir
jmbecker
SmartPLS Developer
Posts: 1281
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: HTMT discriminant validity

Post by jmbecker »

No it is not ok. See also my response here: viewtopic.php?f=5&t=1479.
If one of your tests shows discriminant validity problems you should address them and not just find another test that does not show problems (especially if that test is only a weak test of the problem).
You should try to fix your data and model problems. That would likely also resolve the HTMT value problems. Without fixing the problems the results from your analysis are not reliable and findings cannot be trusted.
Dr. Jan-Michael Becker, BI Norwegian Business School, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de
lebo01
PLS Junior User
Posts: 1
Joined: Tue Mar 26, 2019 4:03 pm
Real name and title: Leon Bongers

Re: HTMT discriminant validity

Post by lebo01 »

Dear Professor Ringle,

I have the same problem. The HTMT value for two constructs was found to be 0.945. As you suggested in another comment I bootstrapped HTMT (2-sided test with 10%). The values for the upper and lower bound are 0.897 and 0.991 (which indicates discriminant validity?), but I do not understand why I have to use a 10% probability of error with the 2-sided test instead of 5%. Do I only need to change it for bootstrapping HTMT or also for path coefficients etc.?
Thank you very much.

Kind regards
Leon
cdrsam
PLS Junior User
Posts: 1
Joined: Mon May 13, 2019 6:26 am
Real name and title: shahid aslam mirza

Re: HTMT discriminant validity

Post by cdrsam »

majortiy of my HTMT values are 0.000
what does that indicate
what is the possible interpretetion
jmbecker
SmartPLS Developer
Posts: 1281
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: HTMT discriminant validity

Post by jmbecker »

Interpretation is that constructs are not correlated, so the cross correlations between indicators of the different constructs are small (zero).
Dr. Jan-Michael Becker, BI Norwegian Business School, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de
mklacmer
PLS Junior User
Posts: 3
Joined: Fri Aug 30, 2019 2:01 pm
Real name and title: Mario Klačmer

Re: HTMT discriminant validity

Post by mklacmer »

Dear coleagues,
my HTMT ratio between two variables is 0,906.
Everythig else is OK.
Could I say that discriminant validity is OK or not?
Many thanks!
MaFa
PLS User
Posts: 12
Joined: Fri Aug 23, 2019 11:32 pm
Real name and title: Ana Foster, PhD student

Re: HTMT discriminant validity

Post by MaFa »

Hello,
Based on the discussion here and the paper introduced, in my data, two constructs have HTMT values equal to 0.94; but when I ran the complete bootstrap, the confidence intervals at 0.95 are lower than 1 (greatest 0.98); can I consider discriminant validity conditions have met?

Thanks a lot for the help.
Post Reply