Tree diagram to illustrate the false discovery rate in significance tests. This example considers 1,000 tests, in which the prevalence of real effects is 10\%. The lower limb shows that with the conventional significance level, p=0.05, there will be 45 false positives. The upper limb shows that there will be 80 true positive tests. The false discovery rate is therefore 45/(45+80)=36\%, far bigger than 5\% (from Colquhoun, 2014).
More from Chalkdust
Things to make
Platonic shapes from tube maps, flexagons, and more mathematical craftWhat’s hot and what’s not
Optimal Pac-Man
Ghostbusting with graph theoryReinventing 2D
What a difference a matrix makesYour social networks
How similar are you to your friends?Art meets maths