Learn to visualize your data like an expert with these practical how-tos for presentation, analysis, and understanding.
Also known as trellis charts, lattice chart, or whatever you want to call them, the technique lets you compare several categories in one view.
Beeswarm charts are useful to highlight individual categories or entities. Animating them can help highlight change over time.
Marimekko charts, or mosaic plots, allow you to compare categories over two quantitative variables.
Using geometric shapes as an encoding can provide another dimension to your charts.
Heatmaps quickly translate data tables into a visual form, making them a great tool to explore a new dataset.
Put multiple time series lines on the same plot, and you quickly end up with a mess. Here are practical ways to clean it up.
Show current evolution against expected historical variability and add one or more series that could account for the difference.
Using a spiral might not be the best way to encode data. But here's how to do it anyway. Just in case.
Also known as a bivariate area chart, the plot type focuses on the comparison between two time series.
Make them move to show a shift in distributions over time.
Quickly see what's below and above average through the noise and seasonal trends.
Using R, we look at how your decreased interaction with others can help slow the spread of infectious diseases.
Create better population pyramids that allow for improved comparisons between sexes and populations.
Network graphs are a good way to find structure and relationships within hierarchical data. Here are several ways to do it.
The chart type can be used to show patterns over time and relationships between variables. This is a comprehensive introduction to making them using two common libraries.
Layout multiple charts in a single view. Then adjust the scales appropriately for maximum comparability and a unified graphic.
Layering time series data or distributions with this method can change the feel and aesthetic versus a multi-line chart or small multiples. In some cases, frequency trails let you show more in less space.
Fill areas with varying line density to give more or less visual attention. With geographic maps, the technique is especially useful to adjust for population density.
Visualize rankings over time instead of absolute values to focus on order instead of the magnitude of change.
Easily compare multiple categories and spot differences between two or more series.
Almost all of my visualization projects that use data from the Census Bureau comes via IPUMS. In this guide, I provide five steps to getting the data you need using their tools.
Using the library command-line gets you more flexibility to highlight the important parts of the data.
With cyclical data, a circular format might be useful. Combine that with a smooth density to reduce noise, and you got yourself a plot.
By shifting the baseline to a reference point, you can focus a line chart on relative change, which can improve the visibility of smaller categories.