SOCIOL 3FF3 Chapter Notes - Chapter 8: Squared Deviations From The Mean, Total Variation
7 May 2019
School
Department
Course
Professor
Document Summary
Computing a bunch of t tests would increase the probability of making a type 1 error. The more statistical tests we do the increase. Need a test that holds type 1 error at a constant level by making a single overall decision to whether there is significant difference. The analysis of variance yields an f ratio whose numerator represents variation btwn groups and denominator reps variation within groups. F ratio indicates size of variation btwn groups relative to the size of variation within each group. The larger the f ratio (larger the variation btwn relative to within) the greater probability of rejecting the null. Reps initial step for measuring total, within, and btwn variations. Squaring deviations from mean and adding the squared deviations. Find the total sum of squares (defined as the sum of the squared deviation of every raw score from the total mean)
