Some Comments on the Use of Magnitude-of-Effect Estimates

In a recent article Murray and Dosser (987) discussed the interpretation and potential misuse of magnitude-of-effect (ME) estimates. In this article I expand on two issues discussed by Murray and Dosser. First, it is demonstrated that ME estimates vary in their susceptibility to sample-size bias depending on whether they are directional or nondirectional estimates. Second, it is demonstrated that study characteristics that influence the size of ME estimates can be explicitly taken into account when comparing studies. Both issues underscore the need to consider very carefully the many study features that can distort ME estimates and make study comparison hazardous.