ggplot stacked bar graph combining 2 data columns

Question

I would like to produce a stacked barplot of my dataframe (df) using the melt function from the reshape package but cannot quite figure out how to transform my dataframe appropriately. Reproducible example below.

{time <- c("Day", "Day", "Night", "Night", "All", "All")
      a <- c(70, 60, 35, 40, 50, 30)
      b <- c(30, 40, 65, 60, 50, 70)
      df <- data.frame(time, a, b)}

Variables A and B represent time spent in two different behavioral states (all equal to 100%) while Time represents a general qualitative description of when data were collected (day, night, or all times). I would like to transform the data to combine variables A and B into a single column so I can produce a stacked barplot with three bars (day, night, all) of the percent time spent in each behavioral state.

Answer 1

You can use tidyr to get the data into the right format, then plot it with ggplot .

library(tidyr)
library(ggplot2)

df %>%
  tidyr::pivot_longer(a:b) %>%
  ggplot(aes(fill = name, y = value, x = time)) +
  geom_bar(position = "stack", stat = "identity")

Sample Data

df <- structure(list(
  time = c("Day", "Day", "Night", "Night", "All",
           "All"),
  a = c(70, 60, 35, 40, 50, 30),
  b = c(30, 40, 65, 60,
        50, 70)
),
class = "data.frame",
row.names = c(NA,-6L))