0
$\begingroup$

I am looking at the relationship between two variables using linear regression (Ordinary least squares method) and would like to compare this relationship between different subpopulations. How can I determine if the differences between regression coefficients are significant. Is ANCOVA the appropriate analysis?

Improve this question
$\endgroup$
1
  • 1
    $\begingroup$ "OLS module" ... ? This "module" is a module of what, exactly? It's hard to understand what you're referring to there. $\endgroup$
    – Glen_b
    Commented Oct 18, 2015 at 2:04

1 Answer 1

Reset to default
1
$\begingroup$

You can do this by adding a variable that corresponds to the subgroup and then looking at interactions between that variable and the main independent variable.

Suppose, for example, your DV is weight and your IV is height and your two populations are men and women. Then you run a regression:

weight ~ height + sex + height*sex

with sex dummy coded.

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.

Not the answer you're looking for? Browse other questions tagged or ask your own question.