Data, code, and images for a posting summarizing three studies about pie charts
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.

Summary Stats of Pie Chart Studies

This is a re-processing of the data we collected in three studies on pie charts for a talk and a blog posting. The idea is to present the data in a way that's readable to most people. In particular, use signed and absolute error instead of log error, and 95% confidence intervals instead of p-value testing and violin plots.

This repository contains all the data (copied from the original sources, see below), an R script to create the images, and a collection of images in three different formats.

I've also added a dataset that had not been published before for a blog posting about a reanalysis of data from a 2010 paper.

What's Where?

  • pie-summaries.R: the R script that does all the magic
  • data: the raw datasets
  • raw-pdf: the PDFs exported from R, as well as the iDraw/Graphic files I used to create all the other images
  • pdf: nice PDFs for each chart, with small images for the legend
  • svg: SVGs of all of the charts with small images for the legend
  • png: PNGs of all of the charts with small images for the legend

Data Set Origin

The datasets in the data directory are copies of the files in the following repositories:


All the images and data are free to use. Please cite these two papers if you use the pie charts study materials, though:

For the simplevis data, please cite: Robert Kosara, Caroline Ziemkiewicz, Do Mechanical Turks Dream of Square Pie Charts?, Proceedings BEyond time and errors: novel evaLuation methods for Information Visualization (BELIV), pp. 373–382, ACM Press, 2010