Chapter 16. Visualizing Uncertainty

One of the most challenging aspects of data visualization is the visualization of uncertainty. When we see a data point drawn in a specific location, we tend to interpret it as a precise representation of the true data value. It is difficult to conceive that a data point could actually lie somewhere it hasn’t been drawn. Yet this scenario is ubiquitous in data visualization. Nearly every dataset we work with has some uncertainty, and whether and how we choose to represent this uncertainty can make a major difference in how accurately our audience perceives the meaning of the data.

Two commonly used approaches to indicate uncertainty are error bars and confidence bands. These approaches were developed in the context of scientific publications, and they require some amount of expert knowledge to be interpreted correctly, yet they are precise and space-efficient. By using error bars, for example, we can show the uncertainties of many different parameter estimates in a single graph. For a lay audience, however, visualization strategies that create a strong intuitive impression of the uncertainty will be preferable, even if they come at the cost of either reduced visualization accuracy or less data-dense displays. Options here include frequency framing, where we explicitly draw different possible scenarios in approximate proportions, or animations that cycle through different possible scenarios.

Framing Probabilities as Frequencies

Before we can ...

Get Fundamentals of Data Visualization now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.