PSYC 3000 Lecture Notes - Lecture 3: Explained Variation, Analysis Of Variance, F-Test

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

Document Summary

An anova with only two groups, it is basically a t test. S pooled is characterising unexplained variation within a group. Variation between groups is caused by a successful treatment. We use the variance of all the groups. If the means are all very separate and the variance will be big. Once anova is completed and a difference between groups is notice, then t tests are used. If i take variance of means and multiply it by sample it turns into estimate of population variance. Both numerator and denominator are estimates of population variance. So they should be the same if the null hypothesis is true. Hence, f = 1 if ho is true. The anova test should reveal that the null hypothesis is false. So further testing must be done to find out which groups are significantly different. F needs to have a critical point in which we can have limits of accepting the null hypothesis.