Hj. Keselman et al., The pairwise multiple comparison multiplicity problem: An alternative approach to familywise and comparisonwise type I error control, PSYCHOL MET, 4(1), 1999, pp. 58-69
When undertaking many tests of significance, researchers are faced with the
problem of how best to control the probability of committing a Type I erro
r. The familywise approach deals directly with multiplicity problems by set
ting a level of significance for an entire set of related hypotheses; the c
omparison approach ignores the issue by setting the rate of error on each i
ndividual hypothesis. A new formulation of control, the false discovery rat
e, does not provide control as stringent as that of the familywise rate, bu
t concomitant with this relaxation in stringency is an increase in sensitiv
ity to detect effects relative to the sensitivity of familywise control. Ty
pe I error and power rates for 4 relatively powerful and easily computed pa
irwise multiple comparison procedures were compared with the false discover
y rate procedure for various 1-way layouts by use of test statistics that d
o not assume variance homogeneity.