I'd like to do a robust hierarchical regression using bootstrapping because regression diagnostics indicate that assumptions of ordinary regression have been violated. I've made a start using the bootstrapping function in SPSS. I'm confused. SPSS provides a "Bootstrap for Coefficients" output (which mirrors the regular "Coefficients" output) but it doesn't provide a "Bootstrap for Model Summary" output (to mirror the regular "Model Summary" output). That is to say, SPSS gives me a bootstrap estimate of whether each of the variables is making a significant contribution to the prediction of Y by testing its regression coefficient, but SPSS doesn't give me a bootstrap estimate of whether the set of variables taken together predict Y at better-than-chance levels. Isn't this a bit of problem? I want a robust estimate of the overall model precisely because my data violate assumptions of normal least square regression, and for this reason I can't trust the standard "Model Summary" that SPSS provides. But if no "Bootstrap Model Summary" is provided in its place it would seem that I don't have a robust estimate.

Any clarification about whether 1) I am confused about bootstrapping, or 2) SPSS doesn't provide seemingly crucial output would be much appreciated.


Your Answer

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

Browse other questions tagged or ask your own question.