From Data to Viz by Conor Healy, Yan Holtz
The ‘data-to-viz’ project aims to guide anyone to the most appropriate graphic representation for their given dataset.
To do so, major chart types have been classified based on input data format. This classification has been translated in a visually appealing decision tree leading to a set of potentially appropriate visualizations.
The decision tree comes in the form of a printed poster and is also embedded within a website that extends its usefulness (data-to-viz.com).
The website includes:
- an interactive decision tree, based on input data format
- an extensive list of graph types with their description
- reproducible R code for every chart
- several data analysis examples based on real data
- a gallery of common dataviz pitfalls: more than 40 caveats are described and possible workarounds are discussed.
While many websites providing examples and descriptions of graph types already exist, this project is different in that it attempts to maximise user-usefulness. Besides providing a decision tool (the tree), it gives an immediately usable creation tool (associated R code), thus hopefully contributing to users engaging more easily and widely with data visualisation.
Data to Viz is a project built by a data analyst (Yan Holtz) and a designer (Conor Healy). It is fully open source, with code available on Github.