Statistical Methods in Small Animal Experimentation

Every confirmatory statistical experiment must start with a hypothesis expressed by one criterion. The blood count is no such criterion, but the hematocrit can be chosen if it is relevant for the effect of the treatment in question. Usually, an experimenter objects to the choice of just one criterion, arguing that all parameters are equally important. He or she should select one which can be expected to show an effect. For example, in an experiment about cell injury, edema could be the key parameter, because if there is no edema, the cell is not injured very much. A criterion may, of course, consist of several repeated measurements of the same parameter, as there are tests which calculate one simultaneous p value from all values.