生物统计学10.pdf

生物統計學 10 Hypothesis Testing – Two-sample Testing (兩個樣本檢定) 陳光琦助理教授(Kuang-Chi Chen) 陳光琦助理教授(Kuang-Chi Chen) chichen6@.tw, Office: H603-2 chichen6@.tw, Office: H603-2 .tw/ .tw/ I. Basic Steps of a Hypothesis Test Step 1: Formulate the two hypotheses. Step 1: Formulate the two hypotheses. H : null hypothesis. H0 : null hypothesis. 0 H : alternative hypothesis, is contradictory to the null H1 : alternative hypothesis, is contradictory to the null 1 hypothesis. hypothesis. Step 2: Determine the random variable and the distribution for Step 2: Determine the random variable and the distribution for the test to assess the evidencefor or against H . the test to assess the evidencefor or against H0 . 0 Step 3: Using the statistic, calculate thep -value. Step 3: Using the statistic, calculate thep -value. A smaller p -value indicates stronger evidence against H . A smaller p -value indicates stronger evidence against H . 0 0 I. Basic Steps of a Hypothesis Test (cont’d) Step 4: Make a comparison of the p -value with a fixed and Step 4: Make a comparison of the p -value with a fixed and preconceived significance level, α. Then make a decision: preconceived significance level, α. Then make a decision: - If p -value ≤ α, then reject the null hypothesis. - If p -value ≤ α, then reject the null hypothesis. - If p -value α, then do not reject the null hypothesis. - If p -value α, then do not reject the null hypothesis. Step 5: Write an appropriate conclusion to the hypothesis test. Step 5: Write an appropriate conclusion to the hypothesis test. -