How can I change the bar names in ggplot geom_bar and the legend title?

Question

I have a data frame :

id = c("A","B","C","D","E")
C1 = c(T,F,T,F,T)
DAT = tibble(id,C1);DAT
ggplot(DAT,aes(C1,color="lightblue"))+ geom_bar(aes(color="black",fill = "red"))

I want to create a bar chart like this one :

But I want to change the true to "ok" and false to "not ok" and also this renaming to appear in the legend at the right with title not colour but "Check". How can I do it in R?

Optional

If I had a row with NA how I could change that ? For example:

id = c("A","B","C","D","E")
C1 = c(T,F,NA,F,T)
DAT = tibble(id,C1);DAT

Carl · Accepted Answer · 2022-06-14T10:52:54.520

2

Like this?

library(tidyverse)

id <- c("A", "B", "C", "D", "E", "F")
C1 <- c(T, F, T, F, T, NA)
DAT <- tibble(id, C1)
DAT
#> # A tibble: 6 × 2
#>   id    C1   
#>   <chr> <lgl>
#> 1 A     TRUE 
#> 2 B     FALSE
#> 3 C     TRUE 
#> 4 D     FALSE
#> 5 E     TRUE 
#> 6 F     NA

DAT |>
  mutate(C1 = case_when(
    C1 == TRUE ~ "ok",
    C1 == FALSE ~ "not ok",
    TRUE ~ "not declared"
  )) |>
  ggplot(aes(C1, fill = C1)) +
  geom_bar() +
  labs(fill = "Check")

^{Created on 2022-06-14 by the reprex package (v2.0.1)}

edited Jun 14 '22 at 10:52

answered Jun 14 '22 at 10:00

Carl

4,232
2
12
24

and if i had a row with NA ? – Homer Jay Simpson Jun 14 '22 at 10:30
If you don't want to see the NAs, you could `drop_na() |>` before the `mutate`. That would exclude the row if either or both columns are NA. `drop_na(id)` if you only wanted to exclude for the one column. – Carl Jun 14 '22 at 10:37
no i want to keep them but to rename the bar name NA that will appear to "not declared" – Homer Jay Simpson Jun 14 '22 at 10:47
Then you can do the above. `case_when` is useful if you have more than two conditions. `TRUE` covers everything else not covered by the first two conditions. – Carl Jun 14 '22 at 10:53
I tried `DAT %>% mutate(C1 = case_when( C1 == TRUE ~ "ok", C1 == FALSE ~ "not ok", NA ~ "not declared" ))%>% ggplot(aes(C1, fill = C1)) + geom_bar() + labs(fill = "Check")` but the NA as a bar name remains. – Homer Jay Simpson Jun 14 '22 at 10:57
`is.na(NA) ~ "not declared"` – Carl Jun 14 '22 at 11:01
awesome.thank you for your clear answers.i appreciate it.you are very helpful – Homer Jay Simpson Jun 14 '22 at 11:04

score 1 · Answer 2 · answered Jun 14 '22 at 10:15

1

In place of if_else you can use case_when which allows to vectorise multiple if_else() statements like

library(tidyverse)

id = c("A","B","C","D","E")
C1 = c(T,F,T,F,T)
DAT = tibble(id,C1)

DAT  %>% 
  mutate(C2 = case_when(C1 == TRUE ~ "OK",
            C1 == FALSE ~ "Not OK")) %>% 
  ggplot(aes(C2, fill = C2)) + 
  geom_bar() +
  labs(fill = "Check")

answered Jun 14 '22 at 10:15

UseR10085

7,120
3
24
54

and if i had a row with NA ? – Homer Jay Simpson Jun 14 '22 at 10:30

score 0 · Answer 3 · answered Jun 14 '22 at 09:58

See this answer:

library(tidyverse)
id = c("A","B","C","D","E")
C1 = c(T,F,T,F,T)
DAT = tibble(id,C1)

ggplot(
  DAT,
  aes(C1,color="lightblue")) + 
  geom_bar(aes(color="black", fill = "red")) +
  scale_x_discrete(
    breaks = c(FALSE, TRUE),
    labels = c("not ok", "ok"))

score 0 · Answer 4 · answered Jun 14 '22 at 10:07

To have variables appear in your legends, avoid fixing the colour, e.g. fill = "lightblue", instead use the variable you want, e.g. fill = C1

id = c("A","B","C","D","E")
C1 = c(T,F,T,F,T)
DAT = tibble(id,C1);DAT
    
ggplot(DAT,aes(C1)) + 
  geom_bar(aes(fill = C1)) +
  scale_x_discrete(labels = c("Not OK", "OK")) +
  scale_fill_manual(labels = c("Not Ok", "OK"), values = c("darkred", "darkblue")) +
  labs(fill = "Anything here")

How can I change the bar names in ggplot geom_bar and the legend title?

4 Answers4