IMS_Blog

Because I forget stuff. Part of norcimo.com

Note: It appears you must have reached this page by a deep level URL. In general this site is currently down and unmaintained. See here

About This Post

Originally posted March 31 2004 at 13:03 under General and Physics. 0 Comments. Trackbacks Disabled.

Lies, Damn Lies and Statistics*

Very good piece on the Risks of Quantitative Studies. We get statistics thrown at us all the time, by politicians, the press or the bloke down the pub. I’ve always remembered my statistics lecturer in my 1st year of being an undergraduate trying to instill in us how unrealiable most statistics in the press are (if only I could actually remember most of what he actually taught us). Possibly the most compelling quote from the above piece is the following:

Researchers often perform statistical analysis to determine whether numeric results are “statistically significant” By convention, they deem an outcome significant if there is less than 5% probability that it could have occurred randomly rather than signifying a true phenomenon.
This sounds reasonable, but it implies that one out of twenty “significant” results might be random if researchers rely purely on quantitative methods.

In other words one swallow does not a summer make, or one study does not a result make. The importance of repetition is often lost on the general public (as is the importance of sample size and the myriad of pitfalls the article touches on) and on the press themselves, often leading to scare mongering (just look at the triple vaccine/autism “scandle”). One result could be a complete freak. Be careful with those statistics, >90% of people don’t really understand them.

*Originally Benjamin Disraeli, not Mark Twain who merely made it a popular phrase in the US.

Comments (0):

Post a comment

Name and email address are required. Email address is never shown. If you enter a URL your name will be linked to it (this and other links will have the rel attribute set to contain nofollow). Markup allowed: <a href="" title="" rel=""> <em> <strong> <abbr title=""> <acronym title=""> <p> <br />. Anything else is stripped; please be valid. Single linebreaks automatically convert to <br />, double to <p>'s. Additionally anything that looks like a bare URL should get automagically linked. Many acronyms and abbreviations are also automagically handled.

Please note this blog's comment policy

Remember me?
Forget Me..

Trackbacks (0):

Trackback URL: http://www.norcimo.com/MT/mt-tb.cgi/38

Advanced...

This Crazy Fool

Who:
Dr Ian Scott
Where:
Croydon (and Gateshead), United Kingdom
Contact:
ian@norcimo.com
What:
Bullding Services Engineer (EngDesign), PhD in Physics (University of York), football fanatic (Newcastle United), open source enthusiast (mainly Mozilla)

More about me [Disclaimer]

You may subscribe to IMS_Blog using the RSS Feed, the Atom Feed or by email.

Creative Commons License

© Ian Scott. Powered by Movable Type 3.2. This blog uses valid XHTML 1.0 Strict and valid CSS. All times are local UK time. For further details see the IMS_Blog about page.. All my feeds in one.