SOC202H1 Lecture Notes - Lecture 11: Contingency Table, Linear Regression, Canada 2006 Census
Document Summary
This is the expected value of y when both x1 and x2 are 0. b1 = the partial slope of the first independent variable (x1) This is the expected change in f y when for every 1-unit increase in x1 holding constant x2. b2 = the partial slope of the second independent variable (x2) This is the expected change in f y when for every 1-unit increase in x2 holding constant x1. From our example, we could calculate a multiple regression equation predicting sentence lengths as follows: Beta weights can be used in multiple regression to ascertain which independent variables has the strongest association with y. Multiple regression with standardized partial slopes (beta weights): Zy = (0. 68)z1 + (0. 29)z2: from our example, the beta-weights show that number of prior convictions (x1) has a much larger effect on sentencing lengths than does age (x2).
