The following book review appeared in the New Yorker recently:


I don’t know much about Brooklyn and I haven’t seen the book in question but I think the reviewer is making a classic statistical error in the last sentence above. Various people bought houses in these Brooklyn neighborhoods: some of these people were white, some were not, some were liberal, some were not, maybe some styled themselves as romantics but surely many did not, some fought the construction of low-income housing projects and probably most were apolitical. That various people involved had different goals does not mean that many individuals were torn in the way described above. The fallacy of composition is to attribute a common view to each member of a diverse population. Better to think about variation: imagine the different attitudes as circles on a complicated Venn diagram.

[Cross-posted at The Monkey Cage]

Our ideas can save democracy... But we need your help! Donate Now!

Andrew Gelman is a professor of statistics and political science and director of the Applied Statistics Center at Columbia University.