Data visualization with base R

R did not wait for ggplot2 to offer awesome data visualization features. It allows to build absolutely any type of chart, as described in this section and in the gallery.

However, it often requires additional quantity of code, but

some people will argue that it allows a greater flexibility.

This section is dedicated to tips and tricks applying to any base R chart. If you're interested in a specific chart type, browse the gallery as all chart section offer base R option.

Last but not least, this course can be a good starting point if you're new to data visualization with base R.

Shape parameters

Main shape features can be controled with a few parameters I always forget.

For lines, the lwd, type and lty parameters allow to control everything. For markers, cex, col and pch do the trick.

Here is a cheatsheet I often come back on to refresh my mind.

Parameters reminder

A cheatsheet to quickly reminder what option to use with what value to customize your chart.

Dealing with margins: mar and oma

Two types of margin are available on a base R chart, controled by the par and the oma parameters.

They are both called through the par() function as described in the dedicated post.

Base R margins

A post dedicated to margin management in base R.

Increase margin

Increase margin size with the mar argument of the par function

Other chart parameters: a cheatsheet

Here is a cheatsheet summarizing the most common base R option to customize the chart appearance.

Cheatsheet base R

A cheatsheet to remember a few chart parameters.

An overview of base R possibilities

Each section of the gallery provides several examples implemented with base R. Here is an overview of my favorite examples:

Mirrored histogram in base R

A mirrored histogram allows to compare the distribution of 2 variables. Here is how to build one in base R.

Vioplot package

The vioplot package allows to build violin charts. Learn how it works.

violin plot with Base R

Histogram with colored tails

Coloring tails sometimes allow to highlight specific areas of the distribution.

Tukey test

Tukey test compares the mean of all pairs of category. Here is how to perform it and represent its result on a boxplot.

Boxplot with jitter

Show individual observations on top of boxes, with jittering to avoid dot overlap.

Polynomial curve

Set a linear model with lm(), and plot it on top of your scatterplot with line().

Color scheme on the side

Add color beside the heatmap to compare actual structure with the expected one.

Image on the chart background

The rasterImage function allows to add an image on the background of the chart.

Number of obs

Add number of observation on top of barplot, and other customization

Reversed Y axis

Learn how to flip the Y axis upside down using the ylim argument.

Most basic

Most basic area chart you can build in base R using the polygon function.

Related chart types

Chord diagram

Network

Sankey

Arc diagram

Edge bundling