The New Yorker has an article out discussing a troubling trend in the reversal of previously verified results of a "scientific method". The abstract I've linked to here is fascinating in itself; I would love to read the whole article, but I can't figure out how to buy it in a reasonable manner (more on this below).
Here's a great quote from the abstract:
The most likely explanation for the decline is an obvious one: regression to the mean.
Yet the effect’s ubiquity seems to violate the laws of statistics.
I know what "regression to the mean mean's, but I'm trying to apply it to this idea of rigorous testing of the effect of a drug, etc, and how those "results" can deteriorate over time.
Anyway, the New Yorker helpfully has the following options to get access to the full article: