multiple regression with interaction terms

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

multiple regression with interaction terms

Marnie LaNoue
Hi All --

Thanks in advance for any help with what may be a simple question. I have
a 'regression framework' research question, which predicts that the
relationships between two continuous predictors and one continuous outcome
will be different across 2 groups. Specifically it predicts that x1 -> Y
will be significant in one group and that x2 -> Y will be significant in
the other group. Is there any reason to prefer a MR with an interaction
term (dummy coding the grouping variable)to 2 separate MR analyses and then
a test of the significance of the difference between the beta coefficients?
Can I gain any additional information one way or the other, or is there
some reason for preference in terms of control of type I error?

Thanks!
Marnie LaNoue, Ph.D.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: multiple regression with interaction terms

Henrik Lolle
Hi Marnie,

The two alternatives will give the same results, and it is just a
question of convvenience which one you choose.

Best,
Henrik

At 23:46 18-06-2006, Marnie LaNoue wrote:

>Hi All --
>
>Thanks in advance for any help with what may be a simple question. I have
>a 'regression framework' research question, which predicts that the
>relationships between two continuous predictors and one continuous outcome
>will be different across 2 groups. Specifically it predicts that x1 -> Y
>will be significant in one group and that x2 -> Y will be significant in
>the other group. Is there any reason to prefer a MR with an interaction
>term (dummy coding the grouping variable)to 2 separate MR analyses and then
>a test of the significance of the difference between the beta coefficients?
>Can I gain any additional information one way or the other, or is there
>some reason for preference in terms of control of type I error?
>
>Thanks!
>Marnie LaNoue, Ph.D.



***********************************************************
Henrik Lolle
Associate Professor
Department of Economics, Politics and Public Administration
Aalborg University
Fibigerstraede 1
DK 9220 Aalborg East
http://www.socsci.auc.dk/institut2/dansk/empl/lolle.htm
***********************************************************
Loading...