Re: Please kindly advice

Posted by Bruce Weaver on Jun 30, 2011; 11:32am
URL: http://spssx-discussion.165.s1.nabble.com/Please-kindly-advice-tp4537676p4538509.html

Assuming one of those variables is seen as dependent on the other, you can use REGRESSION with values of 0 and 1 for gender.  Focus on the coefficient for the explanatory variable of interest rather than the correlation.

Out of curiosity, what is your sample size?  I.e., is it large enough to support a model with 3 explanatory variables?

HTH.

maggieho wrote

Hi,

I am analyzing the relationship between HbA1c (measure in %) and skin thickness (measure in mm) using pearson's correlation but I would also like to know whether gender(nominal data) and diabetic duration(years) are covariate.  Can you please advise on which test I should use to control for gender and diabetic duration in this case? Thank you for your time.

Regards,
Maggie