PSYC 2002 Lecture Notes - Lecture 6: Multicollinearity, Regression Analysis, Standard Score
Document Summary
Regression and correlation: correlation is the measure of the relationship between two variables. Indicates how strong a relationship is between two variables: regression is how well variation in one variable can predict variation in another variable. Predictor and criterion variables: predictor variable is the variable that is used to do the predicting, denoted as x, high school grades, the criterion variable is what you are trying to predict, denoted as y, university grades. For regressio(cid:374), y is repla(cid:272)ed (cid:449)ith , the predi(cid:272)ted (cid:448)alue of y. For a line, a negative means that the line is going down to the right. Mathematical relationships: there is a function where the relationship between correlation and regression becomes clear. Standardized regression formula: (cid:2207)=(cid:2182)(cid:2206: beta is the standardized regression coefficient. Beta = r: (cid:455) is the predi(cid:272)ted ) s(cid:272)ore. Select a value of x and transform it into its corresponding z-score.