Speaking of Science

The Scienticity Blog

Archive for the ‘Gratuitous Quantification’ Category

Jan
31

Exactly How Many?

Posted by jns on January 31, 2005

From The Los Angeles Times story U.S. to Overhaul Training of Iraqi Forces, reporting on the confirmation hearings of Condi Rice as Secretary of State, comes this extraordinary statement:

The Pentagon wants to train about 135,000 police officers, 62,000 national guardsmen, 24,000 army troops and others for a security force totaling 271,041.

What is extraordinary is this number: 271,041. Not, mind you, “271,042″ or “271,043″ or even “about 270,00″, but exactly “271,041″. [N.B. This statement originated with The Pentagon, and not with the LA Times.]
This is a beautiful example of gratuitous quantification and misguided precision. In this context, specifying “…and forty-one” is simply absurd: there is no reason to think that the number could possibly be known to that precision, and suggesting otherwise is misleading , inaccurate, and wrong, as well as being an obvious display of innumeracy that should never appear in Congressional testimony or come out of the Pentagon.
All this would be true, even if the first part of the statement didn’t say

The Pentagon wants to train about 135,000 ….

thus limiting, by saying “about”, the expected precision of the numbers right up front.