Continuous data can tell you exactly how a value is spread, but only if you choose the technique built for that job. In this module, you'll build practical judgment for selecting among histograms, density plots, strip plots, and box plots to match the distributional question you are actually trying to answer.

Discrete data comes with categories, and categories come with a question that looks obvious but rarely is: which technique actually fits the number of categories, the story you are telling, and the audience in front of you. In this module, you'll build judgment for selecting among bar graphs, dot plots, pie charts, and radar plots so you can match each technique to the discrete data challenge it was designed to solve.

Visualizing one distribution is a solved problem. Visualizing six, ten, or eighteen at once introduces challenges of opacity, color, layout, and scale that can turn a useful chart into an unreadable one in seconds. In this module, you'll build practical judgment for configuring histograms, density plots, box plots, violin plots, bar graphs, and circular charts to compare multiple distributions without sacrificing the analytical clarity that makes any visualization worth building.

Relationships in data are rarely announced by the numbers themselves: they have to be rendered visible through the right visualization choice. In this module, you'll build the techniques to reveal continuous and discrete relationships, from raw scatter plots and fitted trend lines through time-ordered line plots to the table and mosaic formats that expose patterns between two categorical variables.

Standard distributions and bivariate relationships capture much of what data has to say — but some analytical questions require a third, fourth, or even fifth dimension to answer fully. In this module, you'll examine three techniques purpose-built for multi-dimensional visualization: matrix scatter and trellis plots for mapping many bivariate relationships simultaneously across a set of variables or groups, bubble plots for encoding a third continuous variable and a categorical fourth directly into the scatter structure, and contour plots for revealing how a z variable changes continuously across a two-dimensional x-y surface.

This course has covered a large range of visualization techniques — from histograms and bar charts through scatter plots and lines of best fit to contour plots and multi-panel matrices. The conclusion brings those techniques together around the one question every visualization project starts with: given this data and this analytical goal, where do I begin? You'll also look beyond the course's technical content to the craft dimensions that separate functional charts from compelling ones, including the perceptual and psychological principles that determine what a viewer actually sees when they encounter a visualization.