What’s Wrong With This Pie Chart?

In this Boston Magazine graphic about the change in Boston’s ethnic composition, can you find what’s wrong? Let’s see:

Bostonracialcomposition

That’s right: a percentage that is supposedly greater than fifty percent is represented by a section that is… less than half of the area of the circle (the 2010 data). That should have been caught by someone. Math illiteracy ain’t cool (whereas poor grammar is).

For the record, it’s worth noting that Boston is now a majority minority city (the 53% for whites is wrong).

This entry was posted in Fucking Morons, Statistics. Bookmark the permalink.

5 Responses to What’s Wrong With This Pie Chart?

  1. Irit Rubin says:

    Well, it says right there: From 2010 U.S. Census. Participants in the census were permitted to select more than one identifying race, therefore the percentages add up to more than 100 percent. They add up to 113.1%. 53.9% of the people marked ‘white’ (some of them with one or more additional categories), but that’s less than 50% of the responses, if one counts multiple responses from the same person. IOW the mistake was to use a pie chart for these data in the first place. Or to have a pie chart with section for ‘more than one race’, and count only those identifying as members of a single racial category in the other sections.

    • Irit Rubin says:

      Oops, I should have written ‘They could solve the problem by having a pie chart with section for ‘more than one race’, and count only those identifying as members of a single racial category in the other sections.’

  2. Min says:

    In 1990, Boston’s pie was only 110%. In 2010 it was 113.1%. The pie has gotten bigger! 😉

  3. TheBrummell says:

    Pie in Boston… must be cream, yes?

  4. Mike Fortun says:

    Uh, aside from the math, there are White, Black, Asian, and Hispanic “races”?! Did Carleton Coon (http://en.wikipedia.org/wiki/Carleton_S._Coon) draw these graphs?

Comments are closed.