Chapter 19. Beautiful Political Data

Andrew Gelman

Jonathan P. Kastellec

Yair Ghitza

SOME OF THE EARLIEST HISTORICAL EXAMPLES OF DATA ANALYSIS INVOLVE POLITICS AND GOVERNMENT; even the word "statistics" reveals the connection of data collection for and about the state. Statistical pioneers, including Playfair, Laplace, and Galton, devoted much of their effort to designing and analyzing public data, and, in the 20th century, statistics was associated with Gallup polls, economic and military organization (Five Year Plans and all that), and even Svengali-like political consultants (as in The 480, a novel from 1964 by the coauthor of The Ugly American, Fail-Safe, and other Cold War–era bestsellers). More recently, TV viewers have become accustomed to colored maps and charts of the latest polls and election results broken down by locality and demographic slices. And at the next level of sophistication are USA Today, the New York Times, and blogs such as FiveThirtyEight.com.

This chapter gives some examples where data visualization has increased our understanding of politics, along with a discussion of the factors involved in making each choice. Here we are focusing on the uses of graphics for research as well as presentation.

We try to apply the following template:

  • "Figure X shows…"

  • "Each point (or line) in the graph represents…"

  • "The separate graphs indicate…"

  • "Before making this graph, we did…which didn't work, because…"

  • "A natural extension would be…"

We do not have a full theory of statistical ...

Get Beautiful Data now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.