Determining Differences in Rates Corresponding to a Given Significance Level

摘要

The well-known Chi-square test can be used to determine whether the difference in two rates is statistically significant, and if so, at what level of significance. This paper discusses the related question of how large a difference in rates must be (when one rate is held constant) in order to show statistical significance at a given level of significance. An illustrative example, adapted from data appearing in the biostatistical literature, is provided.