Exporting levels within a data.frame to csv

Question

#creating dataframe    
df1=data.frame(c("Mary","Sarah","Linda","Mark","Shaun","Jo"),c(1,2,3,4,5,6),c(2,2,2,1,1,2))
names(df1)=c("a","b","c")
#setting levels and labels for questions b and c
blevels=c(1,2,3,4,5,6)
blabels=c("bike","walk","car","bus","train","subway")
clevels=c(1,2)
clabels=c("male","female")
#creating labelled data.frame
df2=df1
#applying labels
df2$b<-factor(df1$b,blevels,blabels)
df2$c<-factor(df1$c,clevels,clabels)

I am using R to create frequency tables for survey responses and use value labels to create more useful frequency tabels.

I also edit the data in R before the frequency tables are created to remove invalid values and use write.csv to export the data.frame for reference.

I want the exported data.frame to show the coding for each question rather than it's label. Currently to do this I create a new labelled data.frame for the frequencies and export the original data.frame.

This leads to the danger of differences between the reference and analysis data.frames.

Is there a way of exporting the levels rather than the labels to a csv file?

Answer 1

Create a function to strip the levels from a single vector:

f <- function(x) {
   if(is.factor(x)) {
      return(as.numeric(x))
   } else {
      return(x)
   }
}

Apply that function to the data.frame:

df <- sapply(f,df)