How to create a bar plot with one categorical variable in different years in ggplot2?

Question

I have a very large data frame where each row in the first column represents an id with numbers. The other rows have a categorical variable that can be of two types (in this example, A or B), each for a year. Here's a simplified data frame as an example:

id  var2017  var2018  var2019
1     A        B         A
2     B        A         A
3     B        A         B
4     A        A         A
5     A        B         B

I'd like to create a bar plot that contains the count of each type (A and B) for each year, with the bars being grouped by type. I am new with R language, so I've tried to create a plot for the years separately, which works fine, as follows:

graph <– ggplot(data = example) +
        geom_bar(aes(x = var2017))

The problem is I don't know how to put them all together. How can I create a plot with all the types for each year being in the x axis, and the count in the y axis? The id doesn't need to be in the output.

Answer 1

The way to plot multiple columns in ggplot is to first convert the data to long form, which can be done with tidyr::gather . Then you map the column it came from (now stored in the "year" column) to one aesthetic, and the count to another ( geom_bar does this for you by counting the number of rows).

library(tidyverse);  
ggplot(data = example %>%
         gather(year, type, -id)) +
  geom_bar(aes(x = year, fill = type), position = "dodge")

(Note, I changed the example to make the different years have different counts. Otherwise it's less clear to see if it's working.)

example <- read.table(
  header = T, 
  stringsAsFactors = F,
  text = "id  var2017  var2018  var2019
           1       A        B         A
           2       B        A         A
           3       B        A         B
           4       B        A         A     # var2017 A changed to B
           5       A        B         B")

Answer 2

Similar to the previous answer but using dplyr::count , geom_col and clearer syntax for the pipes:

library(ggplot2)
library(tidyr)
library(dplyr)

example %>% 
  gather(Var, Val, -id) %>% 
  count(Var, Val) %>% 
  ggplot(aes(Var, n)) + 
    geom_col(aes(fill = Val), 
             position = "dodge")