THE P-P PLOT AS A METHOD FOR COMPARING TREATMENT EFFECTS

This article examines the use of the probability-probability plot (p-p plot) as a method for comparing treatment effects. To begin in the co ntext of three examples the p-p plot is contrasted with the quantile-q uantile plot (q-q plot), which is an alternative means of describing t reatment effects. In these examples it is shown that p-p plots represe nting different experimental conditions or patient populations allow s cale-invariant comparisons of treatment effects but q-q plots do not; that the presentation of the treatment effect by the p-p plot is not o bscured by outliers, whereas it may be in the q-q plot; and that the p -p plot encompasses information in the control distributions that is i mportant for the assessment of treatment effects but that is not incor porated in the q-q plot. Theoretical considerations are presented that show that under appropriate assumptions, the p-p plot is a maximal in variant and contains all the information necessary to make scale-invar iant comparisons of treatment effects. Further, statistical methods fo r assessing patterns observed in the p-p plots are presented and illus trated in two examples.