CONFIDENCE-INTERVALS AND THE SCIENTIFIC METHOD - A CASE FOR HOLM ON THE RANGE

Based on principles of modern philosophy of science, it can be conclud ed that it is the magnitude of a population effect that is the essenti al quantity to examine in determining support or lack of support for a theoretical prediction. To test for theoretical support, the correspo nding statistical null hypothesis must be derived from the theoretical prediction, which means that we must specify and test a range null hy pothesis. Similarly, confidence intervals based on range null hypothes es are required. Certain of the newer multiple comparison procedures a re discussed in terms of their applicability to the problem of generat ing confidence intervals based on range null hypotheses to control the familywise Type I error rate in multiple-sample experiments.