January 30, 2002
From the desk of Mindles H. Dreck:Stats for Iain
Iain Murray makes a very kind comment about me on his fine site. Seeing it sent me to my clippings for something statistically-oriented. I could punt, and send you to Sophie's Mama, but I actually did have something tucked away.
Call this a "lost opportunity" post. A writer lost the chance to make his point in a truly vivid fashion. What a pity.
I noticed this little argument going on in the "Letters" section of the Wall Street Journal. Apparently, a George Pearsall wrote in to apply some politically correct logic to the profiling debate. He claimed that the probability of an Arabic Muslim being a terrorist was three in 100,000. He calculated this figure by dividing the number of terrorists acting on September 11 to the number of Arabic Muslims in the U.S.
A Mr. Richard Porter wrote in on Monday to point out that one cannot infer the probability of an event based on a single occurrence. He notes that by the same reasoning, the probability that a terrorist is anything but an Arabic Muslim is zero.
Well noted, Mr. Porter, but let's not stop there. Reductio ad Absurdum can be more fun than that. For instance, one could extrapolate from September 11 that Arabs who go to strip bars have an overwhelming chance of committing terrorist acts the very next day. Furthermore, the chances of the passengers foiling a highjacking hover around 25%.
But let's not forget the best example: Based on September 11, the chances of a terrorist being dead by noon are 100%.
After all, all known suicide bombers are dead. Isn't that a relief?
Posted by Mindles H. Dreck at January 30, 2002 09:47 PM | Technorati inbound links