A fascinating article appeared on fivethirtyeight.com showing just how easy it is to manipulate results in scientific experiments. The article doesn’t just tell you how this is done; it lets you do it yourself. It gives you a set of data about politics and the economy and challenges you to find correlations between Republicans or Democrats being in power and the influence on the economy. Use employment and GDP as your economic factors, and Republicans have a negative effect on the economy. Use inflation and stock prices, and Republicans have a positive effect on the economy.
It’s fascinating and you should try it for yourself:
What does this have to do with medical school admissions? It shows why the number of retractions of scientific papers has risen by 1000% between 2001 and 2009: researchers have an incentive to manipulate their p-value (a measure of statistical significance) in order to achieve publishable results. The choices they make along the way completely influence the outcome (whether they intentionally do so or not).
In other words, you need to take the most recent scientific findings with a grain of salt. Moreover, just because an effect is proven in one study does not mean it will hold up in future results. So as you go through your research and medical career, understand that the latest research is often wrong.
The whole article is a worth-while read.