Today I am sharing a really useful post - Small samples, and the margin of error from the blog: What’s new
Updates on my research and expository papers, discussion of open problems, and other maths-related topics. By Terence Tao
It is pretty complex in some places, but skip the equations and read the text, which is quite digestible.
As Constellation conducts random sample research in an event marketing measurement context, these are things we worry about and try to point out. Sometimes we see a client take results and run with them that may have significant weaknesses. For example, very high level executives are not nearly as likely to respond to a poll because they are very busy or they have an assistant screen their correspondence, so they are under-represented in the result.
Also relevant, people tend to believe what is reported on TV or in the press. The myriad of polls reported daily have many of the issues explained in this post. I believe that is why networks famously called states incorrectly on election day. I would suggest we are usually looking at close to 10% possible (not probable) error on most of the better political polls reported on TV.
Thanks to Terrence for a very thoughtful article. If he could dumb it down a bit, it should be required reporting on every television network and print media outlet.
Read the Post - Small samples, and the margin of error
10 October, 2008 in expository, math.ST, non-technical | by Terence Tao