A paper (really more a news piece) in Nature launches a major attack on the commonly used and taught statistical methods.
The focus is on Fisher and the T test, the pervasive “p” values students from high school to grad school learn to sue as a measure of reality. Everyone from politicians advocating medical marijuana to teachers unions opposing charter schools misunderstands or intentionally misuses “P” values to prove things.
The issue with replacing P=.05 with a more powerful test, are very real. The Wall Street math whizzes no t and have better tools, the climate models depend on better tools, but the general public, including most scientists do not understand these tools. Everyone who works in science, social science, environmental science, poly sci ………….. should read this:
http://www.nature.com/news/scientific-method-statistical-errors-1.14700