10 years ago
1
Is it time to up the statistical standard for scientific results?
A new analysis suggests a very simple explanation for some of the problems: our statistics are weak. A statistician has figured out how to compare Bayesian statistics to those normally used in scientific tests of significance. By comparing the two, he finds that researchers are often accepting numbers that any good Bayesian would consider to be weak evidence.
Continue Reading http://arstechnica.com
Join the Discussion
This article makes a very good point, but I think that statistical error is the tip of the iceberg. A much bigger problem is outright fraud, bought studies by corporations, and a lack of interest in reproducing others' experiments. Incorrect, fraudulent, and/or otherwise misleading studies are printed all the time and are never retracted. The one that are retracted often remain available in online archives, where they are read and cited by new researchers.
TL;DR I think scientists have been put under too much pressure to produce and publish and the work has gotten sloppy.