1
$\begingroup$

I am pretty familiar with linear regression analysis, but not about how to generate plots using it (aside from residuals, etc.). I'd like to create a scatter plot of two variables, say depression and gratitude for example, where the values are adjusted for demographic factors (e.g. age, sex, education). I know I can find the relationships using a partial correlation or linear regression, but I'd specifically know whether it's possible to generate a scatterplot when the demographics have been controlled for. I'm using the most recent version of SPSS.

Thank you.

$\endgroup$

1 Answer 1

1
$\begingroup$

Yes, this is possible. It is generally called a partial residual plot and is helpful in assessing the linearity assumption for a given covariate.

First, fit a partial regression model including age, sex, education, etc as covariates.

Then, create a scatter plot using the residuals from the partial regression on the y axis and gratitude on the x axis.

This seems to be possible in SPSS

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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