[R] problems with categorical variables

arinbasu@softhome.net arinbasu at softhome.net
Mon Sep 8 14:36:40 CEST 2003


Hi All: 

I am working on a dataset of a study on healthcare workers. One of the 
variables I am studying is a categorical variable (variable name:EDUC, 
indicates educational achievement, with 6 levels: "illiterate", "primary", 
"junior high school", "high school completed", "undergraduate", and 
"postgraduate"). 

I want to collapse the 6 levels to a 4-level categorical variable (let's 
call it "educrec", with the following levels: "postgraduate", 
"undergraduate", "some schooling" and "illiterate"). 

I initially tried the following code: 

x <- table(EDUC)
educrec <- c(x[1], x[2], (x[3]+x[4]+x[5]), x[6])) 

table(educrec) gives me the tally count just fine, but I cannot use educrec 
in any cross tabulation. 

Would greatly appreciate if you can advise on the correct way to recode 
categorical variables to categorical variables in R.
I use Windows XP (Home), R-version 1.7.1
I looked for solutions in the "introduction to R" and the archives, but 
wasn't successful. 

 -Arin Basu




More information about the R-help mailing list