R help - Dec 2012 - how to control the naming of factors in regression

Dear R listers,

I am regressing an ordered variable over an ordered independent variable
using polr.

The output lists every level of the ordered independent variable repeating
its name and adding a suffix. In my case my variable is called
"o.particip"
and in my regression table I get the following:

o.particip.L

o.particip.Q

o.particip.C

o.particip^4

o.particip^5

o.particip^6


Is there any way to control the way R name the levels? I'd like each level
to be consistently identified (either numbers or letters) and preferably
I'd prefer to see the original naming of the level (the value label - in
Stata dictionary). Moreover, the little "hat" in
"o.particip^4" drives
latex nuts. Any ideas on how to fix these things?
Thank you very much for your kind help,
f.

-- 
Francesco Sarracino, Ph.D.
https://sites.google.com/site/fsarracino/

	[[alternative HTML version deleted]]

The '.L', ..., '^6' are the column names from the output
contr.poly(),
the default contrasts function for ordered factors.  You can define
your own contrasts function that calls contr.poly() and renames the
columns to something you like, then set options("contrasts") so your
contrasts function is used for ordered factors.
> contr.myPoly <- function(...) { tmp <- contr.poly(...) ;
colnames(tmp) <- paste0("Degree",seq_len(ncol(tmp))) ; tmp }
> options(contrasts=c("contr.treatment", "contr.myPoly"))
> d <- data.frame(y=1:20, x=ordered(rep(letters[1:6],length=20)))
> lm(y ~ x, data=d)
Call:
lm(formula = y ~ x, data = d)

Coefficients:
(Intercept)     xDegree1     xDegree2     xDegree3     xDegree4     xDegree5  
    10.5000       1.3148       1.3093       0.4472      -1.1339       0.7559  

Bill Dunlap
Spotfire, TIBCO Software
wdunlap tibco.com

> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at
r-project.org] On Behalf
> Of Francesco Sarracino
> Sent: Friday, December 21, 2012 5:13 PM
> To: r-help at r-project.org
> Subject: [R] how to control the naming of factors in regression
> 
> Dear R listers,
> 
> I am regressing an ordered variable over an ordered independent variable
> using polr.
> 
> The output lists every level of the ordered independent variable repeating
> its name and adding a suffix. In my case my variable is called
"o.particip"
> and in my regression table I get the following:
> 
> o.particip.L
> 
> o.particip.Q
> 
> o.particip.C
> 
> o.particip^4
> 
> o.particip^5
> 
> o.particip^6
> 
> 
> Is there any way to control the way R name the levels? I'd like each
level
> to be consistently identified (either numbers or letters) and preferably
> I'd prefer to see the original naming of the level (the value label -
in
> Stata dictionary). Moreover, the little "hat" in
"o.particip^4" drives
> latex nuts. Any ideas on how to fix these things?
> Thank you very much for your kind help,
> f.
> 
> --
> Francesco Sarracino, Ph.D.
> https://sites.google.com/site/fsarracino/
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> 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.