PSYC 301 Lecture Notes - Lecture 14: Null Hypothesis, Asteroid Family, Analysis Of Variance
Document Summary
Across all the scores in our experiment, we can calculate total variability. We can break this apart into variability between treatments and within treatments. Between-treatment variance: provide a measure of the overall differences between treatment conditions. Within-treatment variance: provides a measure of the variability inside each treatment condition. Naturally occurring, random, and unsystematic differences that exist between one sample and another (i. e. , sampling error) To test our hypothesis, we need to know if treatment differences are bigger than that expected from sampling error. Everyone within a treatment condition has the same treatment so any variability in their scores cannot be due to the treatment. Use within-treatment variance to estimate how big the mean differences would be due to sampling error (i. e. , when the null hypothesis is true, how much variability can be expected) Because f-ratios are computed from two variances, f values are always positive (variances are always positive)