[R] Nagelkerke Pseudo R-squared

[David Winsemius]

On Jul 17, 2015, at 4:33 PM, varin sacha wrote:

> Dear R-Experts,
> 
> I have fitted an ordinal logistic regression with just 1 explanatory variable for the reproducible example here below.
> 
> Everything is working, now I try to calculate the Nagelkerke Pseudo R-squared. 
> I have found a package BaylorEdPsych providing many Pseudo R-squared, but the example shown in the package is for GLM (binary logistic regression) not for ordinal logistic regression.
> How can I calculate the Nagelkerke Pseudo R-squared for ordinal logistic regression ?

polr-objects have a deviance node. If this has statistical value (which I have some doubts regarding) then just apply the usual formula to compare nested models.

-- 
David.

> 
> Many thanks as usual for your precious help.
> 
> Reproducible example :
> 
> install.packages("MASS") 
> library(MASS) 
> a=factor(c("tres grand", "grand", "petit","petit","tres grand","grand","petit","petit","tres grand","grand")) 
> b=c("homme", "homme", "femme", "femme", "femme", "homme", "homme", "homme", "femme", "femme") 
> m <- polr(a ~ b, Hess=TRUE) 
> summary(m)
> 
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

David Winsemius
Alameda, CA, USA