weighting scheme in PLS

Questions about the implementation and application of the PLS-SEM method, that are not related to the usage of the SmartPLS software.
Post Reply
ruchi
PLS Expert User
Posts: 75
Joined: Wed Dec 22, 2010 3:09 pm
Real name and title:

weighting scheme in PLS

Post by ruchi »

Hello

I have done survey with 148 total usable dataset. I have 4 reflective construct (measured by 3 items) and one formative construct (measured by 2 items).

I want to know what "weighing scheme " should i select under PLS algorithm ?
There are 3 options- path weighting, factor weighting and centroid weighting.
what does these 3 indicate and when should i use each?

Thanks
Ruchi
User avatar
Diogenes
PLS Super-Expert
Posts: 899
Joined: Sat Oct 15, 2005 5:13 pm
Real name and title:
Location: São Paulo - BRAZIL
Contact:

Post by Diogenes »

Hi,

Factor = the estimation of LV will be estimated using correlations (inner estimation)
Path = the estimation of LV will be estimated with regression and correlations (depends on the arrows)
Centroid = used when the other does not run (it is an older procedure and uses the sum of the other LV as the LV score of the focal LV in the inner estimation)

See:
Henseler, J., Ringle, C. M., & Sinkovics, R. R. (2009). The use of partial least squares path modeling in international marketing. Advances in International Marketing, 20, 277-319. doi: 10.1108/S1474-7979(2009)0000020014. http://php.portals.mbs.ac.uk/Portals/49 ... cs-PLS.pdf

Best regards,

Bido
ruchi
PLS Expert User
Posts: 75
Joined: Wed Dec 22, 2010 3:09 pm
Real name and title:

formative and reflective mixed weightimg schme

Post by ruchi »

If the model contains reflective indicators for few constructs and formative indicators for other few constructs, then which weighting scheme is appropriate ?
User avatar
Hengkov
PLS Super-Expert
Posts: 1599
Joined: Sun Apr 24, 2011 10:13 am
Real name and title: Hengky Latan
Location: AMQ, Indonesia
Contact:

Post by Hengkov »

Hi,
Used factor or Path weighting schame.
ruchi
PLS Expert User
Posts: 75
Joined: Wed Dec 22, 2010 3:09 pm
Real name and title:

Post by ruchi »

But why so ? In some papers i have seen researchers using centroid scheme when formative and reflective indicators also
User avatar
Hengkov
PLS Super-Expert
Posts: 1599
Joined: Sun Apr 24, 2011 10:13 am
Real name and title: Hengky Latan
Location: AMQ, Indonesia
Contact:

Post by Hengkov »

Hi,

Noonan and Wold (1982) Quote:
"Although each weighting scheme follows a particular logic, it has been noted that its choice tends to have little influence on the result: 0.005 or less for structural path and 0.05 or less for measurement path"

Some researcher later agree with this statement (See also Chin and Newsted 1999, p. 317; Gefen et al. 2011, p. viii; Latan 2013, p. 7) and some researcher also recommendation for used "PATH WEIGHTING SCHEME" (See Esposito Vinzi et al. 2010, p. 53; Garson 2012, p. 21; Hair et al. 2012, p. 430; Henseler 2010, p.118; Latan and Ghozali 2012, p.52 Tenenhaus et al. 2005, p. 171). I think it's clear for all user PLS choice weighting scheme. Similar Henseler (2010, p. 117) Quote:
"Therefore, the choice of the inner weighting scheme deserves more attention from PLS researchers and users than it has received up to now in literature"

References:
Chin, W.W. and Newsted, P.R. 1999. "Structural equation modeling analysis with small samples using partial least squares," In R. Hoyle, (Ed.), Statistical Strategies for Small Sample Research, Sage Publications, Thousand Oaks,CA, pp. 307-341.

Esposito Vinzi, V., Trinchera, L., and Amato, S. 2010. “PLS Path Modeling: From Foundations to Recent Developments and Open Issues for Model Assessment and Improvement,” in Handbook of Partial Least Squares: Concepts, Methods and Applications in Marketing and Related Fields, Vincenzo Esposito Vinzi, Wynne W. Chin, Jörg Henseler, and Huiwen Wang, eds., Berlin: Springer, pp. 47-82.

Garson, G. D. 2012. Partial Least Squares. “Blue Book”, Statistical Associates Publishing.

Gefen, D., Rigdon, E. E., and Straub, D. 2011. “An Update and Extension to SEM Guidelines for Administrative and Social Science Research,” MIS Quarterly (35:2), pp. iii-xiv.

Hair, J. F., Sarstedt, M., Ringle, C.M., and Mena, J.A. 2012. “An Assessment of the Use oof Partial Least Squares Structural Equation Modeling in Marketing Research,” Journal of the Academy of Marketing Science (40:1), pp. 414-433.

Henseler, J. 2010. “On the convergence of the partial least squares path modeling algorithm,” Computational Statistics (25:1), pp. 107-120.

Latan, H., and Ghozali, I. 2012. Partial Least Squares: Concept, Technique and Application SmartPLS 2.0 M3, Badan Penerbit Universitas Diponegoro, Semarang.

Latan, H. 2013. "Reporting Result of Partial Least Squares Structural Equation Modeling Analyses (In Press).

Noonan, R., and Wold, H. 1982. “PLS path modeling with indirectly observed variables: A comparison of alternative estimate for latent variables,” In: Jöreskog K. G, Wold, H (Eds) Systems under indirect observation. Causality, structure, prediction, vol II. North-Holland, Amsterdam, pp 75-94.

Tenenhaus, M., Esposito Vinzi,V., Chatelin,Y.-M., and Lauro, C. 2005. “PLS path modeling,” Computational Statistics and Data Analysis, (48:1), pp. 159-205.

Best Regards,
Hengky
bostjanzvanut
PLS Junior User
Posts: 1
Joined: Tue Oct 25, 2022 8:28 am
Real name and title: dr. Boštjan Žvanut, associate prof.

Re: weighting scheme in PLS

Post by bostjanzvanut »

Please, help me with these two questions:

1) In the new version of SMART PLS 4 there is no option "Connect all LVs for Initial calculation" - what happened with it? Is it now the default for Factor?
2) In the factor analysis some item loadings are slightly above 1. Is it OK?
Post Reply