Scatterplot Visualization Guide

When to use a scatterplot

A scatterplot is a classic statistical diagram that lets you visualize the relationship between numeric variables. For instance, if you have a table of data on cities, a scatterplot could let you see if there's a relationship between population and crime levels.

How our scatterplot works

The Many Eyes scatterplot (screenshot above) gives the conventional diagram an interactive twist.

Each row in the data table is represented by a dot. For example, in this scatterplot the rows represent data on American states. The x-coordinate of a dot corresponds to one of the columns in the table (in this case, the percentage of high school graduates). The y-coordinate corresponds to a different column (here, the median household income). Finally, the size of a dot can optionally reflect another column (population, above).

A user can select which columns are used for the x- and y-axes, and which for the dot size, with the three drop-down menus below the scatterplot.

If you want to talk about a particular outlier, use our selection feature: To highlight a particular point in the scatterplot, click on it. Use control-click to highlight more than one dot.

Data requirements

A scatterplot is one of the most flexible methods for visualizing numeric data. It will work with any data table that has at least two numeric columns (for the x- and y-axes) and a text column (to label points as the mouse moves over them).

An example data set suitable for a scatterplot is: