Page 1 of 1

Handling of ordinal variables

Posted: Fri Sep 16, 2016 10:22 am
by janschreier
Hi smartPLS-developers!

Are there any plans or a time line for implementing the proposed algorithm as proposed here:
Schuberth, F., Henseler, J., & Dijkstra, T. K. (2016). Partial least squares path modeling using ordinal categorical indicators. Quality & Quantity. http://doi.org/10.1007/s11135-016-0401-7
or here:
Cantaluppi, G. (2012). A Partial Least Squares Algorithm Handling Ordinal Variables also in Presence of a Small Number of Categories. arXiv:1212.5049 [Stat]. Retrieved from http://arxiv.org/abs/1212.5049

I believe I'm not the only one working with survey data and as far as my understanding goes, this would improve result-quality.

Thanks for a quick info!
Jan

PS: If you happen to have an R package tip which uses either of these algorithms I'd be glad :)

Re: Handling of ordinal variables

Posted: Sun Sep 18, 2016 11:09 am
by jmbecker
It is on our list for future releases, but not for the next release. We have some other high priorities and thoroughly testing new features also takes a lot of time.

Re: Handling of ordinal variables

Posted: Sun Sep 18, 2016 1:32 pm
by janschreier
Hi Jan,

Thanks for the Info. Can understand this and appreciate the testing efforts.

BR, jan

Re: Handling of ordinal variables

Posted: Sun Mar 19, 2017 7:36 am
by Hengkov
More detailed explanation about ordinal variable can be viewed later on PLS Handbook, edited (2018). I think, this algorithm will be added.