POLS 2400 Chapter Notes - Chapter 8: Squared Deviations From The Mean, Total Variation, Null Hypothesis

• School: Northeastern University
• Department: Political Science
• Course: POLS 2400
• Professor: Maness

Document Summary

To conduct an analysis we treat the total variation in a set of scores as being divisible into two components. The distance of deviation of raw scores from their group mean-> variation within groups (denominator) The distance or deviation of group means from one another-> variation between groups (numerator) The analysis of f ratio, whose numerator represents variation between the groups compared. The larger the f ratio-> greater likelihood of rejecting the null. The initial step for measuring total variation, as well as a variation between and within groups. There is more than one type of sum of squares-> each type represents the sum of squared deviations from a mean. Sums of squares tens to become larger as variation increases. Sum of squares also gets larger with increasing sample size. Sum of squares cannot be regarded as entirely satisfactory pure measure of variation. Need to control the number of scores involved.