4
$\begingroup$

I am being provided z-scores of dependent and independent variables. I was checking if it can analyzed as such as raw data?

$\endgroup$
4
$\begingroup$

It's not unusual to do this with the independent variables: subtract the mean and divide by the standard deviation. See the great discussion about that here. I've not heard of doing it for dependent variables, though it seems at first glance that it wouldn't matter.

$\endgroup$
2
$\begingroup$

Yes. This is how you get standardized regression coefficients. Here is another discussion of them. Basically, the point is to remove the unit of measure from the variable. When you only have one predictor in your model, your standaridized regression coefficients are equivalent to correlation coefficients. A 1 standard deviation increase in your $x$ variable leads to a $\beta$ times 1 standard deviation increase in your dependent variable. When you have more than one predictor, the interpretation is similar, but it is a marginal effect. So a change of 1 SD in your predictor leads to an expected change of $\beta$ times 1 SD in your dependent variable, assuming all the other variables in your model are held constant.

$\endgroup$
1
  • $\begingroup$ "Your regression coefficients are analogous to correlation coefficients" - could you elaborate a little bit more on this? This is true only for univariate regression and so statement like this could be misleading. $\endgroup$ – Tim Apr 15 '15 at 16:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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