Korreferat zum Beitrag von E. Eschwege und D. Schwartz
The paper by E. Eschwege and D. Schwartz points out a very important feature of clinical trials: “that problem formulation is essential”. Different clinical trials may be concerned with different problems and need different methods for analysis. It is not appropriate to restrict the analysis of all clinical trials primarily to acception or rejection of a null-hypothesis at a preassigned significance level. This was already stated by M. Zelen during a Biometric Seminar of the National Institutes of Health in Bethesda 1965. He said in the discussion: “We have learned one has only to determine whether to reject at the 5 per cent or 1 per cent level. Then the statistician can grandly draw obvious conclusions about data from any scientific field by proclaiming significance or non-significance. Such nonsense is taught usually by professors who have had minimal contact with the applications of statistical methods to scientific problems” (Cutler et al. (1966), page 873).
Unable to display preview. Download preview PDF.