Return to Website

Number Watch Web Forum

This forum is about wrong numbers in science, politics and the media. It respects good science and good English.

Number Watch Web Forum
Start a New Topic 
Author
Comment
View Entire Thread
Re: 59% less likely

Depends on lots of things. Are they reporting the true relative risk or has the journalist got the wrong end of the stick? And was it a long-term interventional study (unlikely) with a pre-defined endpoint or was this simply among the best results from a data-dredge, thus worth nothing more than "interesting result that might or might not be worth some follow-up".

Of course in the event of correlation it could be simply something else, such as men who drink 6 or more cups of coffee get rather more of life's greatest pleasure as well, which apparently (according to equally credible studies) reduces the risk of prostate cancer as well. And is that because forming the two backed beast genuinely reduces prostate cancer risk or because those who engage more in hanky-panky are younger and fitter on average than those who do not?

Re: 59% less likely

James, thank you for the explanation. It's not as simple as the writer would have us believe.