Mark Kleiman's post on meth is really quite good—among other things, he points out that drug-use surveys tend to understate the number of heavy users, who for some mysterious reason either can't be found or aren't entirely forthcoming when the poll guy comes a-knockin'. In hindsight, John Tierney should have just syndicated and re-printed the Slate piece on meth that he cribbed his New York Times column off of; the problem was that in the process of rewriting it in his own words Tierney removed all the qualifiers and started calling meth a "fad". Yikes. A game of pundit 'telephone' gone horribly awry.

(Previous posts on this topic, with a bit more data, here, here, and here.)