Page 1 of 1

Question- Discriminant validity-PLEASE help!

Posted: Wed Sep 04, 2019 4:37 pm
by MaFa
Hello,
In my model, I have two independent variables, one moderator and one DV. The Fornell larcker test is ok, AVE is higher than correlations; however, I have high correlations between one IV and DV and also two DVs and one IV and moderator (higher than 0.7), also the items load on each other (higher than 0.6); but they load more on their own construct; can I justify that if there is only 0.1 difference?
Checking the HTMT shows that there are issues: the two IVs is 0.96 and the moderator and one IV is 0.95.
I am very confused! Is there any way to work on this model? The results of this model are very interesting and are paths can be justified well.
I really want to work on this model and finish the paper. Please let me know if you know any way I can fix the issue! My sample is 600.

Thanks a lot!

Re: Question- Discriminant validity-PLEASE help!

Posted: Tue Sep 10, 2019 9:02 am
by cringle
Can you merge the problematic constructs?

At the end of this article, you find some recommendations (flow chart): http://link.springer.com/article/10.100 ... 014-0403-8