Is there a way to order factors with facet wrap?

Question

Am trying to do a graph but want to order the factor variable based on given values. It seems the plot does not mirror what I want. I would the languages be ordered based on the meanscore. Any ideas?

library(tidyverse)

set.seed(200) # reproducibility

df <- tibble(

  language = gl(4, 10, labels = c("Python", "R", "Javascipt", "Excel")),
  gender = factor(ifelse(sign(rnorm(40))==-1, 0, 1), labels = c("Male", "Female")),
  score = floor(runif(40, 25, 80))
)

df <- df %>% group_by(gender, language) %>%
  summarise(meanscore = mean(score))

df %>%
  mutate(language = fct_reorder(language, meanscore)) %>%
  ggplot(aes(language, meanscore, fill = gender)) +
  geom_col() +
  facet_wrap(~gender) +
  coord_flip()

Maybe this post will help https://stackoverflow.com/questions/18624394/ggplot-bar-plot-with-facet-dependent-order-of-categories — Ronak Shah, Sep 20 '21 at 12:47

score 1 · Accepted Answer · answered Sep 20 '21 at 12:46

I believe this is what you want? Utilizing the reorder_within from the package tidytext.

library(tidytext)

set.seed(200)

df <- tibble(
  
  language = gl(4, 10, labels = c("Python", "R", "Javascipt", "Excel")),
  gender = factor(ifelse(sign(rnorm(40))==-1, 0, 1), labels = c("Male", "Female")),
  score = floor(runif(40, 25, 80))
)

df <- df %>% group_by(gender, language) %>%
  summarise(meanscore = mean(score))

ggplot(df, aes(reorder_within(language, meanscore, gender), meanscore, fill = gender)) +
  geom_bar(stat = "identity") +
  coord_flip() +
  scale_x_reordered() +
  facet_wrap(gender ~., scales = "free")

Is there a way to order factors with facet wrap?

1 Answers1