Using GLM Univariate to Perform an Analysis of Covariance

Proponents of a government works program want to see if it helps people into better jobs, controlling for their salary before entering the program. A sample of potential program participants were followed, some of whom were randomly selected for enrollment in the program, while others were not.

This information is collected in workprog.sav. See the topic Sample Files for more information. Use the GLM Univariate procedure to perform an analysis of covariance (ANCOVA) on the incomes after the program. An extra assumption of ANCOVA is that there is no significant interaction between the covariate and factor, so begin by fitting a model with an interaction term.

Next