Interaction between a categorical and a continuous variable

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

Christian25
Posts: 7
Joined: Wed Dec 07, 2011 4:50 pm

Interaction between a categorical and a continuous variable

Postby Christian25 » Mon Feb 13, 2012 6:38 pm

Dear board,

I have to analyze whether the effect of one independent continuous variables on a dependent continuous variable is different for two different values (gender: males vs. females) of a categorical independent variable. So my question: how do I best do that?

Is it better to group my dataset according to gender (coded as 0 and 1) and run correlations for males and females separately? If that's a good idea, how do I then test, whether the correlation coefficients that result for the two regressions differ significantly from each other?

Is there a more elegant way in which I can directly test for an interaction of the categorical and the continuous variable?

Thank you very much!

Best regards,

Chris
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: Interaction between a categorical and a continuous varia

Postby GerineL » Tue Feb 14, 2012 9:04 am

use split file to do correlations separately for males and females.
after that, use fisher z: http://faculty.vassar.edu/lowry/rdiff.html
Christian25
Posts: 7
Joined: Wed Dec 07, 2011 4:50 pm

Re: Interaction between a categorical and a continuous varia

Postby Christian25 » Wed Feb 15, 2012 3:32 pm

Okay thank you; I'll do it that way!

Who is online

Users browsing this forum: No registered users and 2 guests

cron