STAT-2300 Lecture Notes - Lecture 10: Null Hypothesis, Type I And Type Ii Errors, Alternative Hypothesis
Document Summary
Type one error: reject true null. Type two error: do not reject false null, guilty person acquitted. Sample data consistent with null = do not reject null. Sample data inconsistent with null = conclude alternative. Conditions: randomly selected sample, n is less than . 05n, np(1-p) greater than or equal to 10. Test statistic (z: difference between point estimate and null value (p-hat p) / (square root of (p(1-p)/n)) Small p values = strong evidence against null: p(results or more extreme|null is true) Reje(cid:272)t (cid:374)ull at 5% sig(cid:374)ifi(cid:272)a(cid:374)(cid:272)e level = 95% i(cid:374)terval does(cid:374)"t (cid:272)o(cid:374)tai(cid:374) (cid:374)ull value. Do not reject null at 5% significance level = 95% interval does contain null value. Suspicion were trying to find support for. Conditions: randomly selected sample, n is less than . 05n, n greater than or equal to 30 or normal population. Test statistic (t) (x-bar - mean) / (s/square root of n: n-1 = degrees of freedom.
