Right now I am reading a book, Proofiness, The Dark Arts of Mathematical Deception by Charles Seife. I will be returning to the public library in a few days.
The author is critical of how numbers are manipulated and used in polls or just regular statements and how people without question or thought will accept things as long as a number is attached. He is especially critical of journalists who use poll results or studies as a basis of a story without critically analyzing the data.
Now today, I came across this article in the Globe and Mail theglobeandmail.com/news/nat … le1931149/
First of all the map is quite difficult to read but Prince Rupert appears to be yellow so that’s good but I am wondering about the study.
It talks about 52000 neighbourhoods. How many people were involved in the study to get an accurate reading on 52000 communities. Well it says they “combined 2006 census demographics with data from the 2003 International Adult Literacy and Life Skills Survey.” How many people wrote the 2003 International Literacy and Lifeskills Survey? Wouldn’t we need one per neighbourhood at least. And is one a sufficient sample size. Is 10? Did a half million adult Canadians fill out this survey. Did anybody here fill out that survey?
And then the article includes this gem of a quote:
“The financial crisis of 2008 is often blamed solely on the banking world’s irresponsibility, but individual decisions by people struggling to understand their mortgages, or the true cost of a loan or a debt, arguably helped bring world economies to their knees.”
Nope, the fault didn’t lie with the brilliant accountants, bankers, stock brokers, company presidents and other financial wizards with their multitude of math degrees. It was the fault of the poor schmucks who can’t add.
Proofiness would say that the study gave the journalist a story to write about the decline in math skills without the journalist using whatever math skills he may or may not have to question it.