August 22, 2007

Innumeracy Cannot Be Overestimated

The innumeracy that Mark has discovered among public relations professionals just scratches the surface. Some years ago I was staying with friends who are schoolteachers. On the desk in the spare bedroom was a handout, five or six pages long, from a workshop that a visiting reading specialist had presented. The subject of the handout was how to calculate the error rate on a test of reading ability. This is not hard. You take the total points possible and subtract the points received for correct answers, yielding the points incorrect. Dividing this by the total points possible yields the fraction incorrect. Multiplying this by 100 yields the error rate as a percentage. One subtraction, one division, one multiplication. Something that with perhaps minor variations (such as computing the percent correct rather than the error rate) surely most teachers have done many times. The math involved is, I believe, typically covered in North American schools in grade 5 (when students are around 10 years old).

It should perhaps be shocking that the presenter felt it necessary to prepare a handout on such a topic, but what is REALLY shocking is that the description of the procedure in the handout was wrong, and that the error could not be attributed to a simple typographical error.

The appalling level of mathematical knowledge on the part of a very large fraction of those who do not specialize in a field with a strong mathematical orientation helps to explain why otherwise intelligent people are so readily deceived by such things as claims about language and gender or crank hypotheses in historical linguistics.

Update: Posts on related topics can be found here, here, and here.

Posted by Bill Poser at August 22, 2007 01:29 PM