Studies have shown that numerical errors are among the most common mistakes in newspapers and other types of media. Apart from reading books like this, this and this, British journalists out there should take advantage of an upcoming free workshop for the press being offered by the Royal Statistics Society. (Thanks to Adrian Monck, who writes a wonderful blog, for spotting this.) Here are the details:
Focusing on real stories, topics include:
- surveys - are the conclusions reached really supported by the evidence?
- pilot studies - how can you know if a piloted policy really has made a difference?
- league tables - are the data any good? Is there really any news behind the figures?
- sample sizes, randomisation, confidence intervals - what the technical language means
You will get tips on asking searching questions and how to avoid statistical pitfalls and fallacies - all from leading statisticians with wide experience of working with the media.
Date: Monday, 26 November 2007
Time: Either 09:30 to 12:30
Or 14:00 to 17:00There is no charge.
To register call Andrew Garratt on 020 7614 3920 or email a dot garratt AT rss dot org dot uk
2 Comments
Well, the Language Log boys[1] will be *really* happy to hear about this one…
[1]and whatever colloquial term is currently in use for females that they’re not offended by
My dad gave me a book called How to Lie with Statistics by Darrell Huff when I was young and though it always seemed like a strange gift to me, I certainly did dig it.
http://tshirtinsurgency.com/