PSY 310 Study Guide - Midterm Guide: Null Hypothesis, Missing Data, Repeated Measures Design
22 Mar 2017
School
Department
Course
Professor
Document Summary
Important distinction: between-subject comparison: contrast between groups that were randomly assigned, within-subject comparison: contrast between groups of matched pairs, natural pairs, or through repeated measures. For 2 independent group comparisons df = n-2, n = number of subjects. Positive illusions: null hypothesis: ratings about likelihood of divorce for people themselves vs. average person will not differ, experimental hypothesis: ratings about the likelihood of divorce for people themselves vs. average person will differ. Iv = subject of question: levels = self and other, groups = each student is in both conditions. 160: design = 2-group design with repeated measures, statistic = t-test. Have 2 ratings from each person not independent groups; they are correlated groups. Need to use t-test for correlated samples (a. k. a. dependent t-test, within groups t-test, or paired t-test) t(159) = 17. 50, p < . 001. For correlated sample comparisons df = n-1, n = number of pairs.
