Tell me more ×
Cross Validated is a question and answer site for statisticians, data analysts, data miners and data visualization experts. It's 100% free, no registration required.
up vote 1 down vote favorite

I am looking at the relationship betwwen two interval level variables, being income and subjective well-being and I want to test the overall correlation as well as the correlation in three professional groups, namely doctors, high school teachers, and shop assistants.

If I use bivariate correlation, I will have to run two tests, one for the overall relationship and the other for pofessional groups. I am using SPSS.

Can this be done in one setup?

Sorry, I am a newbie so regression may be an option but my focus is on correlation.

share|improve this question

1 Answer

up vote 2 down vote accepted

The "split file" function allows you to do this. It will calculate the correlation for every distinct level of the variable "profession". So you'd run your overall correlation first (without using split file), and then proceed with the split file, rerun your correlation, and there you go.

share|improve this answer
Thanks. Wouldn't this be two tests (one overall and one for the professional group), for which I have to correct the alpha level. Could this be done as one "big" test? – Amarald Apr 12 '12 at 22:11
Nope. To do this all at once (take into account the professional group and calculate the association between your 2 other variables), you will need to use linear regression. – dominic999 Apr 12 '12 at 22:26
Thanks. How do I interpret a correlation (coefficient) in a linear regression? I just want to see the relationship between my two variables. – Amarald Apr 13 '12 at 0:03
You won't have a "correlation" coefficient but rather a "slope". You can search for "intro to linear regression" and you'll get plenty of useful info to get you started! – dominic999 Apr 13 '12 at 2:03

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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