PSYC 394 Lecture Notes - Lecture 7: Exercise Intensity, Null Hypothesis, Family-Wise Error Rate
Document Summary
All assumptions met, equal sample sizes: regwq or tukey hsd. Controls for unequal sample sizes (small differences): gabriel"s. Controls for unequal sample sizes (large differences): hochberg"s gt2. Calculate the amended alpha level by dividing the standard alpha of 0. 05 by the number of tests: ex) Bonferroni = / number of tests. Then, run t-tests between the variables using the associated p-value. Step 1) compute mean differences between all groups. Step 3) compare obtained hsd difference with mean differences. [end of material for quiz 2; covers chapters 9 and 11. ] Factorial anova tests are used when an analysis contains several independent variables. Anova test with two independent variables would be called a "two-way independent anova". (the term independent implies that there are different participants in each condition. ) Generalizability: results apply to a wider range of entities. Economy: fewer participants necessary to achieve same level of power compared to separate one-way anova tests. Learning task independent variable task 1 through task 5.
