PSYC 200W Chapter 11: ch11
Document Summary
Analyzing experimental data: an intuitive approach to analysis, the problem: error variance can cause differences b/w means. Confound variance: means may differ even if independent variable does not have an effect. Null hypothesis: independent variable did not have an effect on the dependent variable. Reject v fail to reject null: never accept null. Alpha: prob of making a type 1 error. Beta: prob of making a type 2 error. Power: prob that a study will correctly reject null when null is false (1-beta) Increase n = increase power: related to # participants in study, generally set at 0. 8, power analysis: to determine # participants that is needed in order to detect the effect of a particular independent variable, effect size. Strength of the obtained effect proportion of variability in the dependent variable that is caused by the independent variable (0-1: omega-squared (w^2, eta-squared (n^2)