STAT 301 Study Guide - Midterm Guide: Test Statistic
Document Summary
Assuming that n 30 (or x has a normal distribution i. e. , normal population distribution) p-value. If ha: > 0, then p-value = p(z > ztest) If ha: < 0, then p-value = p(z < ztest) If ha: 0, then p-value = p(z > |ztest|) Assumption is still n 30 (or x~normal) Follows t-distribution with n 1 degrees of freedom p-value. If ha: > 0, then p-value = p(tn-1 > ttest) If ha: < 0, then p-value = p(tn-1< ttest) If ha: 0, then p-value = 2 p(tn-1 > |ttest|) Recall that the t-distribution is symmetric and based on some degrees of freedom. Table gives probabilities where p(t > ttest) can be found without any change, as long as ttest > 0. Note that p(tn-1 < -ttest) = p(tn-1 > ttest) i. e. p(tn-1 < -3. 4) = p(tn-1 > 3. 4) If p(tn-1 > 2. 37), we can bound it such that p(tn-1 > 2. 4) < p(tn-1 > 2. 37) < p(tn-1 >
