CODEWITHSUNDEEP
rggplot2
Petra
Petra

Reputation: 43

ggplot two side by side graphs with the same scale

I'm trying to create two side by side graphs to compare the values (one absolute values and one proportions). I managed to create some simple graphs, but I cannot figure out if I have to wrap them or use a grid? I just keep getting errors.

My data looks something like this:

recent_quarter <- c(12, 15, 2, 3)
all_data <- c(218, 323, 34, 12)
recent_perc <- c(38,47,6,9)
all_perc <- c(37,55,4,5)
gender <- factor(c("M", "F", "Unknown", "Other"),
              levels = c("M", "F", "Unknown", "Other"))

df <- data.frame(gender, all_data, recent_quarter, all_perc, 
                 recent_perc, all_data)

Then I created a simple plot

ggplot(df, aes(x = gender, y = recent_perc)) +
  geom_col(fill = "gray70") +
  theme_minimal()

For this one, I'd like to add a second plot with the all_perc as the y axis. I'm stumped on how to do this.

Upvotes: 0

Views: 1558

Answers (2)

Ben Bolker
Ben Bolker

Reputation: 226047

You could:

g1 <- ggplot(df, aes(x = gender, y = recent_perc)) +
  geom_col(fill = "gray70") +
  theme_minimal() 
g2 <- g1 + aes(y=all_perc)
cowplot::plot_grid(g1,g2)

gridExtra (as referenced in @Josh's answer) and patchwork are two other ways to do the grid assembly.

Or:

library(tidyverse)
df <- data.frame(gender, all_data, recent_quarter, all_perc, all_data, recent_perc)
df_long <- df %>% 
    select(gender, ends_with("perc")) %>% 
    pivot_longer(-gender) ## creates 'name', 'value' columns
ggplot(df_long, aes(gender, value)) + geom_col() +
   facet_wrap(~name)

Upvotes: 1

Josh
Josh

Reputation: 321

install the package gridExtra and use:

grid.arrange(
ggplot(df, aes(x = gender, y = recent_perc)) +
  geom_col(fill = "gray70") +
  theme_minimal(), 
ggplot(df, aes(x = gender, y = all_perc)) +
  geom_col(fill = "gray70") +
  theme_minimal(), 
ncol = 2)

Upvotes: 0

Related Questions