Choosing a Chart Type#
At its core, data visualization highlights or draws attention to aspects of an underlying dataset.
Image source: A. Abela, Extreme Presentation Method
A few of the ways visualizations can highlight a dataset:
Comparison
Relationship
Distribution
Composition
Connection
Location
Some core questions to ask when building a visualization:
What aspect(s) of the data do you want to highlight?
What is the number of variables (and data types)?
Is this going to be a static visualization, or is there a dynamic component (i.e. showing change over time)?
Once you have a sense of what you want the visualization to accomplish, you can make strategic choices about visualization options.
The A. Abela graphic featured above is a good place to start.
Peter Aldhous’s Intro Data Viz course (Fall 2016) includes useful questions to consider.
Image source: Ferdio’s DataVizProject
Ferdio’s DataVisProject also includes examples of a wide range of visualization types. You can explore the library by visualization type, data structure, function, and shape.