| Popis: |
We Japanese are generally too sensitive to the statistical significance level (α) and very credulous to believe that 5 percent should be an absolute standard under any condition. Although there is no room for discussion on the importance of type I error；α, type II error；β or power；1-β should be also taken into consideration. Recently, the importance of statistical power has been pointed out even in the field of medical science. We are therefore requested to refer to power or the sample size, when we especially conduct a clinical trial. Here, the authors put an emphasis on 2 points as follows：1)Methods for computing power；1-β exactly after testing by significance level；α. 2)How to estimate the sample size satisfying the significance level and power, before starting a clinical trial or an animal experiment. |
