HTMT for second order reflective construct

Small talk on SmartPLS that does not correspond to the other forums!
Post Reply
PLS Junior User
Posts: 1
Joined: Sun Dec 16, 2018 2:13 pm
Real name and title: MS Tuan Sarifah Aini Syed Ahmad

HTMT for second order reflective construct

Post by tsyaini » Tue Dec 18, 2018 12:43 pm

I am a novice user of PLS SEM.
The following is some of the HTMT results for my model.
2nd order constructs are A, B, C & D. M is the DV.
1st order constructs are A = A1, A2, A3, A4, A5, B = B1, B2, B3 & B4, C=C1&C2, D=D1, D2, D3, D4 &D5
I have problem with my HTMT as the values for A1 and D1 and D2 exceeds 0.85.
However, when I delete A1, D1 and D2, all HTMT values are below 0.85.
My questions are:
1. Can I delete A1, D1 and D2 from my model?
2. Can I continue the bootstrapping analysis after that and write my findings using the new model that is without A1, D1 and D2?

Thank you.

SmartPLS Developer
Posts: 908
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: HTMT for second order reflective construct

Post by jmbecker » Fri Dec 21, 2018 11:08 am

The question is: With which construct do they have these high HTMT values? And what type of second-order constructs do you have?
For reflective-reflective higher-order constructs it is not unusual (and conceptually no a big problem) that the lower-order constructs within the higher-order model are not discriminant. They belong to the same overall concept.
However, one may reconsider whether joining them to one construct would be better.
Dr. Jan-Michael Becker, University of Cologne, SmartPLS Developer
Researchgate: ... v=hdr_xprf
GoogleScholar: ... AAAJ&hl=de

Post Reply