Histogram bins and binwidth in ggplot2

Default histogram

By default, the underlying computation of geom_histogram through stat_bin uses 30 bins, which is not always a good default.

# install.packages("ggplot2")
library(ggplot2)

# Data
set.seed(05022021)
x <- rnorm(600)
df <- data.frame(x)

# Default histogram
ggplot(df, aes(x = x)) + 
  geom_histogram()

Default histogram bins in ggplot2

This is the reason why you get the following message every time you create a default histogram in ggplot2:

stat_bin() using bins = 30. Pick better value with binwidth.

Possible options to deal with this is setting the number of bins with bins argument or modifying the width of each bin with binwidth argument.

`bins` argument

The number of bins or bars of the histogram can be customized with the bins argument of the geom_histogram function. In this example 15 bins seem to be a good choice while 50 are too many.

Bin selection in ggplot histogram

15 bins

# install.packages("ggplot2")
library(ggplot2)

# Data
set.seed(05022021)
x <- rnorm(600)
df <- data.frame(x)

# Histogram bins
ggplot(df, aes(x = x)) + 
  geom_histogram(colour = 4, fill = "white", 
                 bins = 15)

Control the number of bins in ggplot2

50 bins

# install.packages("ggplot2")
library(ggplot2)

# Data
set.seed(05022021)
x <- rnorm(600)
df <- data.frame(x)

# Histogram bins
ggplot(df, aes(x = x)) + 
  geom_histogram(colour = 4, fill = "white", 
                 bins = 50)

`binwidth` argument

The other option is using the binwidth argument of the geom_histogram function. This argument controls the width of each bin along the X-axis. Note that this argument overrides the bin argument.

Binwidth of 0.5

# install.packages("ggplot2")
library(ggplot2)

# Data
set.seed(05022021)
x <- rnorm(600)
df <- data.frame(x)

# Histogram bin width
ggplot(df, aes(x = x)) + 
  geom_histogram(colour = 4, fill = "white", 
                 binwidth = 0.5)

Control the bin width of the ggplot2 histograms

Binwidth of 0.15

# install.packages("ggplot2")
library(ggplot2)

# Data
set.seed(05022021)
x <- rnorm(600)
df <- data.frame(x)

# Histogram bin width
ggplot(df, aes(x = x)) + 
  geom_histogram(colour = 4, fill = "white", 
                 binwidth = 0.15)

binwidth argument of geom_histogram