Plotting multiple binary variables on the same plot in ggplot

Question

I am hoping to use ggplot to construct a barplot of frequencies (or just % 1s) of a bunch of binary variables, and am having trouble getting them all together on one plot.

The variables all stem from the same question in a survey, so ideally it'd be nice to have data that is tidy with one column for this variable, but respondents could select more than one option and I'm hoping to retain that instead of having a "more than one selected" option. Here is a slice of the data:

structure(list(gender = structure(c("Male", "Male", "Female", 
"Female", "Female", "Female", "Male", "Male", "Male", "Male"), label = "Q4", format.stata = "%24s"), 
    var1 = structure(c("0", "0", "1", "1", "0", "0", "0", "0", 
    "0", "0"), format.stata = "%9s"), var2 = structure(c("0", 
    "98", "1", "0", "0", "0", "0", "0", "0", "0"), format.stata = "%9s"), 
    var3 = structure(c("0", "0", "0", "0", "0", "0", "0", "0", 
    "0", "0"), format.stata = "%9s"), var4 = structure(c("1", 
    "0", "1", "0", "0", "0", "1", "1", "0", "0"), format.stata = "%9s"), 
    var5 = structure(c("1", "0", "0", "0", "0", "1", "0", "0", 
    "0", "0"), format.stata = "%9s")), row.names = c(NA, -10L
), class = c("tbl_df", "tbl", "data.frame"))

Can you clarify what you are looking to construct based on the data provided? It's not clear to me how you want to organize your intended chart, given the example data. Also.. is there supposed to be a 98 in there or is it supposed to be all 1's and 0's? — chemdork123
– chemdork123, Commented Apr 6, 2021 at 1:36
@chemdork123 Sorry, the 98 represents a missing value. And the ideal chart would have var1, var2, var3, etc. along the x axis and with a frequency or percentage of 1s along the y for each respective var. — user14212134
– user14212134, Commented Apr 6, 2021 at 2:17

Ronak Shah · Accepted Answer · 2021-04-06 02:28:22Z

1

Get the data in long format so that it is easier to plot.

library(tidyverse)

df %>%
  pivot_longer(cols = starts_with('var')) %>%
  group_by(name) %>%
  summarise(frequency_of_1 = sum(value == 1)) %>%
  #If you need percentage use mean instead of sum
  #summarise(frequency_of_1 = mean(value == 1)) %>%
  ggplot() + aes(name, frequency_of_1) + geom_col()

In base R you can do this with colSums and barplot.

barplot(colSums(df[-1] == 1))
#For percentage
#barplot(colMeans(df[-1] == 1))

edited Apr 6, 2021 at 2:28

answered Apr 6, 2021 at 2:22

Ronak Shah

391k20 gold badges173 silver badges237 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

user14212134 Over a year ago

Thank you very much, this is perfect. Is there an easy way to incorporate each value broken down (by something like gender)? It doesn't seem like it will work the same way as it would with a non-pivot_longer table would (which would just be ggplot(d, aes(x, fill = gender) or whatever)...

Ronak Shah Over a year ago

You need to include gender in group_by. Try :

df %>% pivot_longer(cols = starts_with('var')) %>%group_by(gender, name) %>%summarise(frequency_of_1 = sum(value == 1)) %>% ggplot() + aes(name, frequency_of_1, fill = gender) + geom_col(position = 'dodge')

Collectives™ on Stack Overflow

Plotting multiple binary variables on the same plot in ggplot

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related