What data is being visualized?

Infovis designers make choices about what data to represent and how to define their variables. So a good first step in understanding an information visualization is questioning the data itself:

  • where did it come from?
  • who collected it?
  • how it was categorized?
  • what was left out?
  • how might the data have been selected and organized differently?

Here composition teachers can draw on their expertise with research strategies: asking students to question and compare sources of information, to consider the credibility of a source, to analyze the limitations of a source, and so on.

for example

In the following three graphics from Michael Newman's Election Maps for the 2008 U.S. presidential elections, we can see a visual overview of voting in America that differs in crucial ways depending on the data that's being represented.

This map represents votes at the state level.

This map represents votes at the county level.

This map uses shades of purple to indicate percentages of votes at the county level.

This cartogram, a map in which the sizes of states are rescaled according to their population, offers perhaps the most accurate view of all, by introducing population as a key variable.

This map from Wikipedia factors in time as one of the variables, representing the relative blue-ness and red-ness of states over the last four presidential elections.