How to customize a horizontal stacked-bar chart with ggplot2 - r

Question

I'm just starting with R visuals, and I'm stuck on this one. In the code below you will find the table I'm working with and its rendering.
What I'm trying to do is to obtain a horizontal bar chart where for each Pool value I have two variables (D_MTD and I_MTD).
To achieve that (and I don't know if this is the right way of doing it) I added two geom_bar layers and what I get is pretty much what it should look like (it's okay to have stacked columns).
MyTable <- data.frame(
Pool = c(" ", "(CA) Domestic", "(CA) International", "(FR) International", "(US) Domestic", "(US) International" ),
ADomestic = c(0, 36000, 3000, 1200, 54000, 6000),
AInternational = c(0, 0, 600, 600, 0, 1200),
D_MTD = c(23.4, 0, 0, 0, 8.38, 0),
I_MTD = c(6.63, 0, 0, 0, 0, 0)
)
MyTable
MyVisual <- MyTable %>%
ggplot(
Data = MyTable,
aes(y = Pool)
) +
geom_bar(
aes(
weight = D_MTD,
colour = "Domestic"
),
alpha = 0.5
) +
geom_bar(
aes(
weight = I_MTD,
colour = "International"
),
alpha = 0.5
) +
labs(
title = "Month-To-Date Consumption by Pool",
subtitle = "(in minutes)",
y = "Pools",
x = "MTD Used Minutes",
colour = "Pool context"
)
MyVisual
However, the visual has two outline colours that I don't want (I would like the bars to be filled with colors instead). I tried to use the fill argument to assign a fixed color to the bars (ex: red for Domestic and blue for International) but that generates another legend that I don't want to show.
Example:
MyVisual <- MyTable %>%
ggplot(
Data = MyTable,
aes(y = Pool)
) +
geom_bar(
aes(
weight = D_MTD,
colour = "Domestic",
fill = "red"
),
alpha = 0.5
) +
geom_bar(
aes(
weight = I_MTD,
colour = "International",
fill = "blue"
),
alpha = 0.5
) +
labs(
title = "Month-To-Date Consumption by Pool",
subtitle = "(in minutes)",
y = "Pools",
x = "MTD Used Minutes",
colour = "Pool context"
)
MyVisual
Also, how do I apply labels to each bar with the corresponding values?
Thanks in advance!

Answer 1

This solution should work for you. It contains elements, and transformation that for a starter may be confusing, but you will have to face these at some point. I recommend you to read the book R for Data Science, at least the chapter reated to data visualization with ggplot2.
Steps:
1- Reshape the data usin the pivot_longer() function
2- Creating factors and labels for the plot
library(tidyr)
library(dplyr)
library(ggplot2)
MyTable <- data.frame(
Pool = c(" ", "(CA) Domestic", "(CA) International", "(FR) International", "(US) Domestic", "(US) International" ),
ADomestic = c(0, 36000, 3000, 1200, 54000, 6000),
AInternational = c(0, 0, 600, 600, 0, 1200),
D_MTD = c(23.4, 0, 0, 0, 8.38, 0),
I_MTD = c(6.63, 0, 0, 0, 0, 0)
)
to_plot <- MyTable %>%
select(Pool, D_MTD, I_MTD) %>%
# Data to long format
pivot_longer(
cols = contains('MTD'),
names_to = 'pool_context',
values_to = 'used_minutes'
) %>%
mutate(
pool_context = factor(
pool_context, c('D_MTD', 'I_MTD'), c('Domestic', 'International')
),
# Labels only for non 0 values
label = ifelse(used_minutes == 0, NA, used_minutes)
)
to_plot %>%
ggplot(aes(y = Pool, x = used_minutes)) +
# Fill aesthetic only for bars, to avoid labels to get color too
geom_col(aes(fill = pool_context)) +
theme(legend.position = 'bottom') +
geom_label(aes(label = label), hjust = 1.1) +
labs(fill = NULL, x = 'Used minutes (MTD)', y = 'Pool')
Created on 2022-02-17 by the reprex package (v2.0.1)