forum.smartpls.com

one follow-up question: does this "PLS is stronger in the measurement model whereas Covariance-Based is stronger in the structural model" also hold true for formative measures (i know, i cannot model a formative CB-SEM, but still). I was just re-reading Chin, 2010 and know I am under the i...

Regarding Myth 5: Sample Size: I think a possible answer can be found in Chin, 2010, p.661ff: PLS only performs a series of simple OLS regressions (i.e. correlations), thus, to determine the sample size you only have to consider the endogenous LV with the largest number of predictors, and not all LV...

i would be very interested in hearing an pls-expert commenting these "myths"

Dear Bido,

could you provide any literature that explaines the point that PLS is stronger in the measurement model whereas Covariance-Based is stronger in the structural model?

I couldn't find anything in the literature yet.

thank you in advance

thank you

hi cristian, thanks a lot, this indeed helps. i was thinking of something like that, but couldn't formulate it properly. But actually, i am not sure what your recommendation implies: you are not suggesting to report the higher R^2 before implementing the "mediation arrow", are you? you sim...

actually, I have the same problem, and I am not quite sure how to make sense of it. If I have 2 IVs and DV, i get a certain R^2 (i.e. explained variance) on the DV. However, when I also connect the two DVs (making on of them a mediator), the R^2 in the DV actually drops. This is in stark contrast to...

Dear Hengky, thanks for your reply but that doesn't actually answer my question. the question is whether it is to be expected that an identical model explains more variance in AMOS than in SmartPLS. In my naive understanding, I would guessed that SmartPLS can explain more (since it is "predicit...

Dear Hengky, you are right, I wasn't paying close attention to the fact that Vinzi et al. (2010) only talk about the relative GoF whereas you talk about the "absolute" GoF. This makes comparing the two even more difficult. If i want to compare a formative and a reflexive model, would you s...

Dear all, i've run the exact same model (both reflective) a) in AMOS b) in PLS I know that AMOS covariance-based and PLS is not, but i was still wondering whether it is to be expected that Amos outputs a higher proportion of variance explained on endogenous LVs Is this a coincidence or does this tel...

By the way, I used the communalities provided by SmartPLS because Prof. Ringle posted that they run fine (http://forum.smartpls.com/viewtopic.php?t=944&highlight=communality) however, they are computed as the "average squared outer loadings" not as the "average squared outer weigh...

Dear all, For the last couple of days I have been trying to construct Goodness of Fit indices for formative models based on Vinzi, Trinchera, Amato (2010), pp. 58-60, pp. 56-60. Here, I wanted to share my procedure in case anyone tries to do the same. For explanations WHY you need to do the followin...

Dear all, especially Prof. Ringle and the admin, i am following up on Andreas Pufall's question: the redundancies given from the PLS-Algorithm (not Blindfolding) in SmartPLS 2.0 M3 are still not computed as "R² * Com", correct? Therefore, should i still compute them "by hand"? al...