PSYC 301 Study Guide - Midterm Guide: Flatline, Conditional Variance, Type I And Type Ii Errors

• School: Simon Fraser University
• Department: Psychology
• Course: PSYC 301
• Professor: Michael Maraun

Document Summary

Step 1: specify the h0 and h1 pairs/make a testable claim (can be nondirectional or directional). The pair should also exhaust all possible values/states of nature. *note: we never know which h is true, because we do(cid:374)"t ha(cid:448)e a(cid:272)(cid:272)ess to a(cid:374) e(cid:374)ti(cid:396)e populatio(cid:374). We a(cid:396)e simply making decisions, and they may/may not be right. Step 2: specify the population distribution, something along the lines of x~n(cid:894) , 2) Step 3: what are the assumptions of step 2 (normality and homogeneity of variances most common) If h0 and assumptions are true, then test statistic distributed on the null distribution x ~ x. Step 7: type ii control = for a fixed , we seek to minimize the p(type ii) by maximizing the sample size: 2 power games/ways to control for type ii. There are a lot of ways in which h0 can be false, for each of these departures from the null there is a specific p(type ii).