SPSS: Calculate HTMT ratios (Discriminant Validity) in SPSS

This will show you how to assess the discriminant validity of a construct through SPSS.
Reference:
Henseler, J., Ringle, C.M. & Sarstedt, M. A new criterion for assessing discriminant validity in variance-based structural equation modelling. J. of the Acad. Mark. Sci. 43, 115-135 (2015). doi.org/10.100...

[@dr.sandeepprabhu9014]

Excellent video. With respect: Never expected any good video from Karachi, Pakistan. May be as an Indian I had a bias. Big thanks.

[@asma7918]

Such a helpful video mam. Thank you for sharing with us.

[@educationalinsightskarachi4969]

Thanks for your kind words

[@bhootfm7912]

If the HTMT value rather than. 85 what can i do to change this value??? I wanna published

[@davidy6577]

Can you share how to conduct convergent validity as well? Particularly how do we report or what do we do when the correlations among items are less than 0.3?

[@educationalinsightskarachi4969]

You can assess the converget validity through AVE (Average Variance Extracted). You can watch the first half of the video on link below for AVE calculation. kzbin.info/www/bejne/Y6Sro5iYfryreLc

[@educationalinsightskarachi4969]

The AVE should be greater than 0.5. If it is lesser, the items with loading less than 0.7 should be dropped. (Hair et al., 2019)

[@davidy6577]

@@educationalinsightskarachi4969 Thank you for your response. It seems like HTMT ratio (this video) and the video that you just mentioned in this comment, seem to test both discrimant validity. Is this true? If so, is the only difference between HTMT vs. Fornall-Larcker condition - the fact that HTMT is used only to compare two factors and Fornall-Larcker condition can test multiple factors at the same time?

[@davidy6577]

@@educationalinsightskarachi4969 Thank you for your response! I'm not sure if I follow your guidance. I watched the video and I know how to calculate AVE but how do I test convergent validity once I find the AVE information?

[@educationalinsightskarachi4969]

Yes. Both a are methods of assessing discriminant validity.

[@sogenannt]

Do you need sig. correlations for that calculation or does this dont matter?

[@educationalinsightskarachi4969]

The correlation should be significant. However, in case of low correlation it might even be insignificant. Such cases should be reported in the interpretation.