"The Mathematics of Narcissism" in Slate.com

Anyone who has tried to compile statistics for a large group of humans has been confronted with the issue of which of a large collection of measurements are the most important. It's useful to remember that how we choose the key data is worth revisiting occasionally. Jordan Ellsberg discusses this issue, and how it was resolved in two very different contexts, in a Slate post called, "The Mathematics of Narcissism," available here.

The two contexts are the decision of the committee writing DSM V to exclude "narcissistic and paranoid personality disorder" from the list of psychiatric diagnoses, and the decision of the authors of the National Research Council's committee ranking graduate schools essentially to throw up their hands instead of agreeing on a single ranking system.

I can think of lots of contexts -- New York City's ranking of its schools with a letter grade comes to mind -- in which this issue could use revisiting. Sometimes the contentious discussion just tells you more than the single grade or ranking does.