by Colonel Sun » Wed Nov 10, 2010 2:38 pm
It's an article in the general press, so I wouldn't read too much into it.
However, the most common problems in medical research are that
1/ n is to small. Biological systems have a notoriously large natural variability and the sample sizes for the control and experimental are typically too small.
The size of the groups is often limited by cost and the time taken to obtain such groups, for example, is a double blind drug study.
2/ the experimental design does not properly account for confounding effects other than the effect under study
Never criticize anyone until you've walked several kilometres in their shoes.
Because
1. You're now several kilometres away; and
2. You've got their shoes.