CODEWITHSUNDEEP
rggplot2
Keelin
Keelin

Reputation: 397

dynamic scale_y_continuous for dynamic axis

having scanned everything I can find on scale_y_continuous is there are way to achieve the following:

Dynamic y axis for survey data where limits change for % and mean result. P is a % value and the axis needs to be set at 0 - 100 M is a mean value and needs to be set at 1 - 6

This would be really easy if all values were in the same plot; however I have two graphs together in a ggarrange and the second graph has values that are less than the first and the bars are not aligned.

library(ggplot2)

df <- data.frame(dose=c("D0.5", "D1", "D2"),
                 P = c(42, 98, 77.5),
                 M = c(2,3.4,5.5)
                 )
ggplot(data=df, aes(x=dose, y=P)) +
  geom_bar(stat="identity")

Is there a way to set scale_y_continuous for P to c(0,100), and for M c(0.6).

Upvotes: 0

Views: 731

Answers (1)

Ronak Shah
Ronak Shah

Reputation: 388817

To get one plot based on user input choice we can do :

library(ggplot2)
choice = 'P'

if(choice == 'P') {
  p1 <- ggplot(data=df, aes(x=dose, y=P)) + geom_col() + 
              scale_y_continuous(limits = c(0, 100))
} else {
  p1 <- ggplot(data=df, aes(x=dose, y=M)) + geom_col() + 
          scale_y_continuous(limits = c(0, 6))
}
p1

To get both the plots together :

You can set limits for individual plots using scale_y_continuous before combining them in ggarrange.

p1 <- ggplot(data=df, aes(x=dose, y=P)) + geom_col() + 
           scale_y_continuous(limits = c(0, 100))
p2 <- ggplot(data=df, aes(x=dose, y=M)) + geom_col() + 
           scale_y_continuous(limits = c(0, 6))
ggpubr::ggarrange(p1, p2)

enter image description here

However, you can also consder using facets with different y-axis :

df %>%
  tidyr::pivot_longer(cols = c(P, M)) %>%
  ggplot() + aes(dose, value) + geom_col()  +
  facet_wrap(.~name, scales = "free_y")

enter image description here

Upvotes: 2

Related Questions