GGplot2 facet_wrap() histograms given a value of other column

I have a large dataframe data that looks like this

     state    values
1      0       0.25
2      1       0.73
3      1       0.21
4      2       0.95
5      3       0.37
6      0       0.72
7      2       0.81
8      1       0.13
9      1       0.05
10     3       0.46

I would like to use facet_wrap() to create four histograms: each one of them would correspond to values for a given value of state, so i.e. value histogram only for the observations that come from the state == 1. At the same time I want my histograms to have the same x values: min and max of the whole dataset, so that I can compare them easily. The number of observations for each class may vary.

I tried to do that in this fashion, but it's obviously wrong.

ggplot(gather(data), aes(value)) + 
  geom_histogram(bins = 6) + 
  facet_wrap(~ values + state, scales = "free_x")

Solution 1:

Something like this?

df %>% 
  mutate(state=factor(state)) %>% 
  ggplot(aes(x=values)) + 
  geom_histogram(bins = 6)+
  facet_wrap(.~state)