Page 1 of 1

Path Analysis in SmartPLS 3

Posted: Fri Oct 13, 2017 3:54 am
by esece1
I ran a path analysis (consistent bootstrapping) in SmartPLS 3 with a total of five latent and one endogenous variables. I ran this with a sample of 90 cases. While reviewing the report, I noticed the path coefficients result is absent the T-stats and P Values (see below). Why did the T-stat and P-value not appear?

Original Sample (O) Sample Mean (M) Standard Deviation (STDEV) T Statistics (|O/STDEV|) P Values
CP -> CIS 0.007 n/a n/a
ISA -> CIS -0.167 n/a n/a
MP -> CIS 0.377 n/a n/a
NP -> CIS -0.175 n/a n/a
TMS -> CIS 0.459 n/a n/a

Re: Path Analysis in SmartPLS 3

Posted: Sun Oct 22, 2017 8:48 pm
by jmbecker
This is an inherent problem of the consistent PLS algorithm. It sometimes produces inadmissable solutions similar to Heywood cases in CB-SEM.

Please read these related discussions:
viewtopic.php?f=5&t=3531
viewtopic.php?f=3&t=3766