Plot multiple variables on one barplot? - r

Question

I'm working with this dataset: https://archive.ics.uci.edu/ml/datasets/Wholesale+customers
I want to plot a histogram with all the variables that have the amount spent (everything but region and channel). I'd like them to be plotted by channel, which has 2 levels. I got the following code from an example on a website, but put in my variables:
category=c(rep("Fresh",2),rep("Grocery",2),rep("Milk",2),rep("Frozen",2),
rep("Detergents_Paper",2),rep("Delicassen",2))
condition=rep(c("Food Service", "Retail"))
value=abs(rnorm(12 , 0 , 15))
data=data.frame(category,condition,value)
ggplot(data, aes(fill=condition, y=value, x=category)) +
geom_bar(position="dodge", stat="identity")
This produces what I want, but it doesn't use my data. Here's the graph I get, but the values don't mean anything since they're basically random.
How do I get my data to plot like this?

Answer 1

By loading the tidyr package, the data can be reshaped to support the expected output.
library(ggplot2)
library(tidyr)
After reading in the data with the correct classes for the columns (factors for Channel and Region while the remaining six fields are numeric), check the data for correctness.
df <- read.csv(file = url('https://archive.ics.uci.edu/ml/machine-learning-databases/00292/Wholesale%20customers%20data.csv'), colClasses = c('factor','factor','numeric','numeric','numeric','numeric','numeric','numeric'))
str(df)
'data.frame': 440 obs. of 8 variables:
$ Channel : Factor w/ 2 levels "1","2": 2 2 2 1 2 2 2 2 1 2 ...
$ Region : Factor w/ 3 levels "1","2","3": 3 3 3 3 3 3 3 3 3 3 ...
$ Fresh : num 12669 7057 6353 13265 22615 ...
$ Milk : num 9656 9810 8808 1196 5410 ...
$ Grocery : num 7561 9568 7684 4221 7198 ...
$ Frozen : num 214 1762 2405 6404 3915 ...
$ Detergents_Paper: num 2674 3293 3516 507 1777 ...
$ Delicassen : num 1338 1776 7844 1788 5185 ...
head(df)
Channel Region Fresh Milk Grocery Frozen Detergents_Paper Delicassen
1 2 3 12669 9656 7561 214 2674 1338
2 2 3 7057 9810 9568 1762 3293 1776
3 2 3 6353 8808 7684 2405 3516 7844
4 1 3 13265 1196 4221 6404 507 1788
5 2 3 22615 5410 7198 3915 1777 5185
6 2 3 9413 8259 5126 666 1795 1451
The data appear to have imported correctly.
Next we use a combination of tidyr::gather and ggplot2::ggplot to produce the desired bar plot (not histogram).
df %>%
tidyr::gather(Type, Amount, -c(Channel, Region)) %>%
ggplot(aes(x=Type, y=Amount, fill=Channel, group=Channel)) +
geom_col(position = position_dodge())
tidyr::gather(Type, Amount, -c(Channel, Region)) will reshape the data set from this:
Channel Region Fresh Milk Grocery Frozen Detergents_Paper Delicassen
1 2 3 12669 9656 7561 214 2674 1338
2 2 3 7057 9810 9568 1762 3293 1776
3 2 3 6353 8808 7684 2405 3516 7844
4 1 3 13265 1196 4221 6404 507 1788
5 2 3 22615 5410 7198 3915 1777 5185
6 2 3 9413 8259 5126 666 1795 1451
To a "longer" data set which now has the product type as rows:
Channel Region Type Amount
1 2 3 Fresh 12669
2 2 3 Fresh 7057
3 2 3 Fresh 6353
4 1 3 Fresh 13265
5 2 3 Fresh 22615
6 2 3 Fresh 9413
This prepares the data to be plotted using ggplot2::ggplot where the x input can be mapped to the new Type variable and the y variable to Amount.
Be sure to use Group=Channel and position=position_dodge() so that ggplot will know that you want the bars side-by-side.