« More Lunacy | Main | The Mahdi on Population »

September 10, 2007

Do They Mean Polly?

Interesting question in The Guardian:

Would journalists benefit from numbers training?

I wonder who they could be referring to?

September 10, 2007 in Media | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c2d3e53ef00e54eef2b1e8834

Listed below are links to weblogs that reference Do They Mean Polly?:

Comments

If they did mean polly it could be because of the Fact Checking Pollyanna blog:

http://factcheckingpollyanna.blogspot.com/

Posted by: gordon-bennett | Sep 10, 2007 11:56:34 AM

If they did mean polly it could be because of the Fact Checking Pollyanna blog:

http://factcheckingpollyanna.blogspot.com/

Posted by: gordon-bennett | Sep 10, 2007 11:57:02 AM

If they did mean polly it could be because of the Fact Checking Pollyanna blog:

http://factcheckingpollyanna.blogspot.com/

Posted by: gordon-bennett | Sep 10, 2007 11:57:12 AM

If they did mean polly it could be because of the Fact Checking Pollyanna blog:

http://factcheckingpollyanna.blogspot.com/

Posted by: gordon-bennett | Sep 10, 2007 11:57:19 AM

If they did mean polly it could be because of the Fact Checking Pollyanna blog:

http://factcheckingpollyanna.blogspot.com/

Posted by: gordon-bennett | Sep 10, 2007 11:57:34 AM

If they did mean polly it could be because of the Fact Checking Pollyanna blog:

http://factcheckingpollyanna.blogspot.com/

Posted by: gordon-bennett | Sep 10, 2007 11:57:42 AM

If they did mean polly it could be because of the Fact Checking Pollyanna blog:

http://factcheckingpollyanna.blogspot.com/

Posted by: gordon-bennett | Sep 10, 2007 11:58:08 AM

The problem is not just basic numeracy, it's also in the practical application thereof. Basic economics, statistics, science etc etc.

But then incomprehension isn't just limited to numbers. Although not a serviceman nor an expert, I'm interested in things military. The number of times I've read a supposed "defence correspondent" making simple mistakes is stunning. And I'm sure people with other interests could make similar statments about the reporting of the things they know about.

But maybe the Guardian is thinking about this problem from the wrong angle. The discussion is about makinging liberal arts journalists numerate. What about taking numerate people and teaching them how to report? I suspect this might be an easier task and it would certainly cut down on the number of gross errors committed.

Posted by: The Remittance Man | Sep 10, 2007 1:44:37 PM

some of the math complained about ,is done to make the stated problem worse,therefore subject to a new law(ie)

the 8 fold increase in liver disease from drinking ,is done for "gee whiz" purposes not clarity..

Posted by: embutler | Sep 10, 2007 1:51:14 PM

Although I am all for greater journalist numeracy I think there probably is a case for having a 'numbers proof-reader' as well, as we all sometimes make mistakes on magnitudes and so on.

Two which bug me however are 1) using annual data to make short-term points, especially in house prices. Because annual house price inflation is 10% doesn't mean anything about whether the June interest rate rise is biting. You need, taking on board the cautions about using monthly data, to annualise the monthly data. 2) "worst ever" - two things here. One where the data series only goes back five years or so, so the "worst ever" is in fact the worst in five years. Worst than that is when "worst ever" refers to the year-on-year rate of change, as in this classic example

http://observer.guardian.co.uk/business/story/0,6903,1543905,00.html

Posted by: Matthew | Sep 10, 2007 2:17:14 PM

I might need a links proof-reader. I meant this story...

http://news.bbc.co.uk/1/hi/business/4133768.stm

The Observer one is possibly OK if it only meant the footfall data.

Posted by: Matthew | Sep 10, 2007 2:19:53 PM

it's the Times and the ASI you write for, Tim, not the Guardian.

Posted by: dsquared | Sep 10, 2007 2:27:33 PM

Post a comment