1 Department of Pediatrics, Children's Hospital of Michigan, Wayne State University, Detroit, MI 48201
In Reply.
We thank Boynton and Merritt for pointing out the error in the figure in our paper.1 The abscissa should be labeled "Total number of united pairs" as they have indicated. When using sequential analysis, a plan is constructed that takes into account the probability of erroneously rejecting the null hypothesis ( ), the probability of erroneously accepting the null hypothesis (
), and the degree of change sought (0).2 Only pairs in which the outcomes are different ("untied pairs") are plotted on the sequential plan.