PSYC 3000 Lecture Notes - Lecture 12: Repeated Measures Design, Statistical Significance, Statistical Hypothesis Testing

• School: Carleton University
• Department: Psychology
• Course: PSYC 3000
• Professor: Bruce Hutcheon

Document Summary

With 80% power it means you will make a type 2 error 20% of the time. Which means you will say there is no effect when there actually is one. Depending on what type of test it is, it"s not uncommon to see a higher power being used. With a medium cohen"s effect size, you need 64 participants per groups. Which is much higher than the normal psychological study. Cohen"s d should be done before the data is gathered, that way you know how large your sample should be. Four most important factors to think about when dealing with stats test. Often times statistical significance is irrelevant for real life changes. This list is relevant for any type of statistical test (one sample, two sample, anova etc) Power can be greatly enlarged compared to other t tests. This test is used in repeated measures design. (ex: treated before and after the treatment).