I am using SmartPLS 3 for analysis. When I am running simple PLS the measurement model looks fine but when I run PLSc composite relaibility of interaction term falls to .23. Also, the loadings of indictors of interaction term decreases to mere .3 . There is no problem in CR and loadings of predictor and moderator are even in PLSc. The value of R2 is also increasing in PLSc.
Hi,
the difference between PLSc and PLS is that PLSc consistently estimates models where the constructs are modeled as common factors while PLS is only consistent for models where all constructs are modeled as composites. For a difference between common factors and composites see Rigdon (2012) and Bollen & Bauldry (2011)
Kind regards
Rigdon, Edward E. "Rethinking partial least squares path modeling: in praise of simple methods." Long Range Planning 45.5 (2012): 341-358.
Bollen, Kenneth A., and Shawn Bauldry. "Three Cs in measurement models: causal indicators, composite indicators, and covariates." Psychological methods 16.3 (2011): 265.
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running