Via The Incidental Economist, I learned that JAMA is rolling out a new feature–a guide to statistics and methods. The neat thing about they way they’ve organized this is that each article will discuss a statistical method actually employed in a paper appearing in the same issue of the journal. So, this first installment is on intention-to-treat analysis, which is employed in a study of erythropoietin use and tranfusion thresholds in traumatic brain injury (and it turns out that neither Epo nor maintaining a hemoglobin > 10 g/dL affects neurological outcome).
The Incidental Economist post referenced above also has a link to a great little paper explaining p values.
And for good measure, I wanted to highlight a really interesting book I just read, titled “The theory that would not die: How Bayes’ rule cracked the enigma code, hunted down Russian submarines, & emerged triumphant from two centuries of controversy.” It’s a fascinating tale of how Bayes’s theorem was actually fleshed out by Laplace, ridiculed and banished by the frequentist giants Fisher and Pearson, used to great effect by the Allied militaries during WWII, ignored due to security classifications in the war’s aftermath, but found to be useful in many other applications (including medicine) thereafter.