1
$\begingroup$

I'm planning on running a hierarchical multiple regression. In the first step, I would like to enter demographic characteristics, second step continuous predictor variables of interest, and third step interactions between the continuous predictor variables. However, the issue I've run into is that some of the demographic variables are more than two levels and either nominal (e.g., education level: less than BA, BA, graduate degree) or ordinal (e.g., age: 20-30, 31-40, 41-50, 51+). My dependent variable is continuous.

My question is: can I dummy code the nominal/ordinal variables and then run the multiple regression as I normally would? Or, would you recommend handling the data differently? I'm using SPSS, so I appreciate any help/input. Thanks!

Improve this question
$\endgroup$

1 Answer 1

Reset to default
2
$\begingroup$

"Hierarchical" multiple regression here is just a fancy way of saying you're fitting three models, each nested inside the subsequent ones, & comparing them. So there's no need to code predictors differently from the way you normally would. Clearly you'll want to include all the dummies for any one categorical variable in the same step.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.