Change levels with dplyr

[Earl Brown]

I'm wondering if there is a way to use dplyr to change the levels of a factor before piping the data frame to ggplot(), in order to create more human-reader-friendly levels in a ggplot. Here's a toy example:

toy.df <- data.frame(originalVar = c("A", "B", "C", "D", "E"))

I've tried the following three ideas, but to no avail:

toy.df %>%

  do(

    levels(.$originalVar) <- c("V", "W", "X", "Y", "Z")

  )

toy.df %>%

  mutate(newVar = factor(.$originalVar, levels = c("V", "W", "X", "Y", "Z")))

toy.df %>%

  mutate(

    newVar = do(

                factor(.$originalVar, levels = c("V", "W", "X", "Y", "Z"))

              )

  )

You can simply change the levels of the factor with levels() before calling ggplot(), or use ggplot(transform(...)), as was offered on SO here:

http://stackoverflow.com/questions/3472980/ggplot-how-to-change-facet-labels

But, it feels like this could be accomplished in the dplyr pipeline, but I can't figure it out.

Once again, the end goal is to change the level labels in a ggplot boxplot or a faceted scatterplot to something more human-friendly, like can be accomplished with base::boxplot(..., names = c("V", "W", "X", "Y", "Z")).

Thanks for any ideas. Earl

[Hadley Wickham]

I'd think you'd want

toy.df %>%
mutate(newVar = factor(originalVar, levels = c("V", "W", "X", "Y", "Z")))

Hadley

> --
> You received this message because you are subscribed to the Google Groups
> "manipulatr" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to manipulatr+...@googlegroups.com.
> To post to this group, send email to manip...@googlegroups.com.
> Visit this group at http://groups.google.com/group/manipulatr.
> For more options, visit https://groups.google.com/d/optout.



--
http://had.co.nz/

[Earl Brown]

I think I had tried that before, but didn't mention it. It results in "<NA>":

> toy.df <- data.frame(originalVar = c("A", "B", "C", "D", "E"))

> 

> toy.df %>%

+   mutate(newVar = factor(originalVar, levels = c("V", "W", "X", "Y", "Z")))

  originalVar newVar

1           A   <NA>

2           B   <NA>

3           C   <NA>

4           D   <NA>

5           E   <NA>

[Hadley Wickham]

x <- factor(c("A", "B", "C", "D", "E"))
factor(x, levels = c("V", "W", "X", "Y", "Z"))
factor(x, labels = c("V", "W", "X", "Y", "Z"))

Hadley

[Earl Brown]

Simple fix. Thank you very much. That is what I was looking for. 

Here's my (slightly larger) toy example, for those who would like to change the labels in ggplots:

library("ggplot2")

library("dplyr")

toy.df <- data.frame(

  originalVar = rep(c("A", "B", "C", "D", "E"), 10),

  value = runif(50)

)

toy.df %>%

  mutate(newVar = factor(originalVar, labels = c("V", "W", "X", "Y", "Z"))) %>% # mind the 'labels', not the 'levels'

  ggplot(aes(newVar, value)) +

  geom_boxplot()