PSY-PC 2120 Lecture Notes - Lecture 1: Effect Size, Statistical Process Control, Hyperplane
4 Oct 2018
School
Department
Course
Professor
Document Summary
Magnitude measures strength, sign indicated direction, must be between -1 and 1. Will not be influenced by scales of measurement because it is standardized. Usually thought to apply to situations where x1 and y are constant variables that are interval or ratio scales. We want to make the residuals small in order to provide the best approximation of the observed y values, so we define an aggregate measure of size of residuals in the sample. These following equations provide the linear regression intercept and slope for slr that minimize the sum of squared residuals. Relationship between raw and standardized regression slope coefficients. Total variance of y can be partitioned into variance explained by predictor and error variance. Squared correlation - x explains __% of the variance in y. Reject the null hypothesis that there is no linear effect of x on y in the population .
