Note: Once in a while, I will provide links to other blogs that contain relevant materials, with limited commentary. In so doing, I also hope to introduce some of the blogs I admire.
Is it okay to ignore results from people you don't trust? (Bad Science)
This is from Ben Goldacre's blog, Bad Science, which debunks charlatans and bad medical advice. Goldacre is a UK-based writer, trained as a doctor. I'm not interested in his lede question but extremely grateful that he introduced the paper by Cataldo et. al.
The paper reviews all available research on the link between smoking and Alzheimer's, for each author, determines whether he or she has links to the tobacco industry, and discovers that "the risks of Alzheimer's associated with smoking reported by [those researchers linked to tobacco] were on average about a third lower than those conducted by other researchers."
In a recent interview, I said: "It’s really not just about the numbers. It’s about the people who are using these numbers." This research demonstrates why we need to pay attention.
***
Eggs-actly what ARE the chances... (Daily Mail)
I can't believe I am linking to the Daily Mail. I found this on David Spiegelhalter's Understanding Uncertainty blog so that's my excuse. This report of someone discovering a box of six eggs all of which had double yolks understandably caused a stir. In this article, someone who knows probability theory explains why this happens with a higher chance than we expect. David has a few more comments on his blog.
The lesson: in real life, the beloved mathematical assumption of independent events rarely holds.
Comments