I poked some fun at Charles Murray yesterday, but it’s worth noting the dishonesty of writing something like “I can’t vouch for the representativeness of the sample, but at about eight o’clock last night in the St. Denis area of Paris.” Saint Denis is a notoriously low-income high-crime banlieue that’s right outside Paris. It’s the kind of place a foreign researcher might visit to get a look at a low-income, high-crime former industrial town near Paris—a legitimate and interesting thing to take a look it, but something you would undertake knowing full well that you were getting an unrepresentative sample. It’d be the equivalent of going to Anacostia in DC and then professing ignorance as to how representative what you saw was.
On a related note, I was thinking about Banlieue 13 the other day and was frustrated that we don’t have a good English language translation for “banlieue.” In school I was taught to translate it as “suburb,” but that conveys a very misleading impression. Rendering “Banlieue 13″ as “District B13″ is sort of nonsense, but rendering it as “Suburb 13″ would induce confusion. The term “Outer Boroughs” from New York probably conveys a more similar idea.