Weighting Scheme

[cringle]

I want to measure a pls path model for strategic success factors with empirical data. From a practical point of view, does it matter if I use the centroid, factor, or path weighting scheme?

[swende]

From a practical point of view, the choice of a weighting scheme does not matter. The results will be (almost) identical.

[joerghenseler]

If you want to maximize the R-square, you should rather use the path weighting scheme.

[rablt]

Why is the path weighting scheme recommended to maximize the R-square? Why is it maximizing R-square?

I remember that in one of your PLS training courses the factor weighting scheme was recommended to achieve "well-balanced models" because the estimation treats all constructs equally. (see Chin, 1998: "It may be possible to obtain stable estimates for weights and loadings of each component independent of the final estimates for the structural model.”). Is this argumentation correct?

[bwilson]

Dear all,

Developmental work by:
Wilson, B. and Henseler, J. (2007), Modeling Reflective Higher-Order Constructs using Three Approaches with PLS Path Modeling: A Monte Carlo Comparison.Thyne, M.; Deans, K. R., and Gnoth, J. in Australian and New Zealand Marketing Academy Conference. Department of Marketing, School of Business. University of Otago. University of Otago., New Zealand.; 791-800.

is showing that for a tested population model weighting choice is important.

happy reading.

best regards,