It would be equally unrealistic to suppose that employers can eliminate, or discover and explain, the myriad of innocent causes that may lead to statistical imbalances in the composition of their work forces.
Dr Hand admires the meticulous work of Steve McIntyre, a mining consultant and blogger, who unearthed statistical problems in another climate analysis.
This is why there are good statistical reasons to be cautious about jumping on the Avastin bandwagon for wet AMD just because it appears to work as well as Lucentis: Simply, the trend for adverse effects from Avastin is greater when the sample of patients is bigger.