Confound it! Correlation is (usually) not causation! But why not?

gwern branwen, U.S. Department of Education, Jul 27, 2014
Commentary by Stephen Downes
files/images/cancernet.JPG

When somebody proposes a simple mechanism to improve (say) learning outcomes, they're most always wrong. But why? It's because they have ascribed a simple cause-effect relation onto a complex phenomenon. But why should complexity impact causation? Complex phenomena are densely connected networks where correlations are increasingly likely to be the result of underlying conditions rather than the result of one thing causing another. This article makes the point, with mathematics, and a good example drawn from the literature on cancer research.

Views: 0 today, 113 total (since January 1, 2017).[Direct Link]
Creative Commons License. gRSShopper

Copyright 2015 Stephen Downes ~ Contact: stephen@downes.ca
This page generated by gRSShopper.
Last Updated: Oct 17, 2017 02:03 a.m.