Discriminant validity check

Questions about the implementation and application of the PLS-SEM method, that are not related to the usage of the SmartPLS software.
Post Reply
Catharsis
PLS Junior User
Posts: 7
Joined: Mon Mar 04, 2019 11:52 am
Real name and title: Human Kohzadi Student

Discriminant validity check

Post by Catharsis »

Hello,

I am currently investigating some reflective research models with SmartPLS and had a few questions about checking the measurement model, specifically discriminant validity.

I would like to know what to do if my Fornell-Larcker criterion or the cross-loading criterion is not met. If so, do I simply remove the corresponding items and recalculate the PLS algorithm? So far I haven't found much in the literature, because it is assumed that the criteria are fulfilled in principle. In isolated cases, however, they were not met by me.

I am grateful for any help :)

Best regards!
jmbecker
SmartPLS Developer
Posts: 1281
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Discriminant validity check

Post by jmbecker »

There is no standard procedure for such cases.
It means you constructs are not distinct. That can have several reasons. Are they conceptually different enough? Or are they conceptually similar?
If they are conceptually similar, may it make sense to include them in a higher-order construct? Do you need to separately? Does it make sense to merge them? Are specific items very unreliable (low loading) and cause the problem? Are specific indicators wrongly assigned (does switching constructs solve the problem?). Are measures established or new?
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
Post Reply