Mills lists a variety of ways that data can be manipulated (“tortured”) to achieve a goal by a dishonest researcher. One approach is termed Procrustean data torturing.


Procrustes was a Greek bandit who physically altered travelers until they fit the length of a metal bed.


Starting point: hypothesis, often one that is popular (and thus less likely to be closely scrutinized)


Process: Data is altered until it fits a hypothesis, then the manipulated data is presented as proof of the hypothesis.


Various approaches:

(1) changing inclusion criteria to include subjects whose data supports the hypothesis

(2) changing exclusion criteria to remove anyone whose data does not support the hypothesis

(3) changing the analysis until the data fit the hypothesis

(4) changing cutoffs or the definitions for what is normal or abnormal


To read more or access our algorithms and calculators, please log in or register.