Massachusetts Institute of Technology
Department of Urban Studies and Planning

11.220 Quantitative Reasoning and Statistical Methods for Planning

SPSS

11.220-HOME   |   TEST-OUT   |   BRUSH-UP   |   Brushup-DATA

1. We'll try to improve on our model for predicting annual income by adding a variable for gender. We already created a variable called "male" that is coded 0 for female and 1 for male. This is called a "dummy variable." We will add the "male" dummy variable to our model. Go to Analyze->Regression->Linear and add the variable "male" under "Independent(s):"

2. Check the result. Compare the difference between the coefficients and significance of this regression and the previous bivariate regression.

Our R2 is now raised to .262. It is nearly twice the R2 from the bivariate model, but is still pretty low.

The F statistic is similar to the bivariate model; not too high, but significant.

The contribution of years of education is a little more than in the first model and the significance has improved. But, as you can see, being male contributes about $14,286.22 to your annual income. The t test for this coefficient is also significant to a value of .017.

Created by Myounggu Kang on January 25, 2004. Edited by Rhonda Ryznar on January 20, 2005.