Switch branches/tags
Nothing to show
Clone or download
Maëlle Salmon
Maëlle Salmon update website
Latest commit 7c39a04 Jul 9, 2018
Failed to load latest commit information.
R update documents with merge May 7, 2017
data-raw update plots to reset par settings May 7, 2017
inst update documents with merge May 7, 2017
tests even MORE examples! 😎 May 7, 2017
.Rbuildignore close #21 by adding a codemeta.json May 2, 2018
.travis.yml Coveralls May 7, 2017
CONDUCT.md Initial commit May 1, 2017
CONTRIBUTING.md update links and README Jul 9, 2018
DESCRIPTION update website Jul 9, 2018
LICENSE Add minimal test May 1, 2017
NEWS.md Updated comments May 8, 2017
README.Rmd update links and README Jul 9, 2018
README.md update links and README Jul 9, 2018
_pkgdown.yml update website Jul 9, 2018
codecov.yml Initial commit May 1, 2017
cran-comments.md Updated comments May 8, 2017



CRAN version Downloads Build Status Project Status: Active – The project has reached a stable, usable state and is being actively developed.

This package wraps the awesome Datasaurus Dozen datasets. The Datasaurus Dozen show us why visualisation is important – summary statistics can be the same but distributions can be very different. In short, this package gives a fun alternative to Anscombe’s Quartet, available in R as anscombe.

The original Datasaurus was created by Alberto Cairo in this great blog post.

The other Dozen were generated using simulated annealing and the process is described in the paper Same Stats, Different Graphs: Generating Datasets with Varied Appearance and Identical Statistics through Simulated Annealing by Justin Matejka and George Fitzmaurice.

In the paper, Justin and George simulate a variety of datasets that the same summary statistics to the Datasaurus but have very different distributions.


The latest stable version (0.1.2) is available on CRAN


You can get the latest development version from GitHub, so use devtools to install the package



You can use the package to produce Anscombe plots and more.

ggplot(datasaurus_dozen, aes(x=x, y=y, colour=dataset))+
  theme(legend.position = "none")+
  facet_wrap(~dataset, ncol=3)

Contributing to the package

Code of Conduct

Anyone getting involved in this package agrees to our Code of Conduct. If someone is breaking the Will Wheaton rule aka Don’t be a dick, or breaking the Code of Conduct, please let me know at steph@itsalocke.com

Bug reports

When you file a bug report, please spend some time making it easy for us to follow and reproduce. The more time you spend on making the bug report coherent, the more time we can dedicate to investigate the bug as opposed to the bug report.


Got an idea for how we can improve the package? Awesome stuff!

Please raise it with some succinct information on expected behaviour of the enhancement and why you think it’ll improve the package.

Package development

We really want people to contribute to the package. A great way to start doing this is to look at the help wanted issues and/or contribute an example.

Examples for this package are done in base R or with ggplot2 as an optional example, using the structure:

#ggplot2 code here

As this is a data package, most of the documentation is sitting in one file (R/Datasaurus-package.R) so we keep the examples in a separate directory (inst/examples).

  • If there isn’t a file for the dataset you want to write an example for, you can make one by just calling it datasetname.R. To reference an example file, add the line @example inst/datasetname.R in the relevant documentation section of R/Datasaurus-package.R.


We’re relatively loose on coding conventions.

  • Datasets are lower-case with underscores between words
  • R code should be formatted with the “Reformat code” option in RStudio
  • There are no standards for base R plots
  • My preferred ggplot2 themes are theme_minimal where axes labels matter and theme_void when they do not but I’m OK with the default ggplot2 theming if you want to avoid writing longer ggplot2 code