Are empirical questions inherently non-philosophical? If answers to those questions can be determined by polling or science, should philosophers never address them?

Your question touches on a current debate within philosophy. You can find more about the debate by searching under "experimental philosophy" and "x-phi" on the web. Regardless of which side one takes, however, it's always important to know which kind of question (empirical, conceptual, logical, normative, or a mixture of those) one is trying to answer: the answer to "Which kind of question is that?" is a philosophical matter.

In my experience, philosophers too often fail to recognize that the question they're asking has empirical aspects -- aspects that, as philosophers, they're not trained to investigate. If a question can be answered by polling or some other empirical method, then any philosopher who tries to answer it had better be properly trained in the relevant empirical method. Once the empirical results are in, the implications of those results are something that philosophers, as such, are well-equipped to work on.

Read another response by Stephen Maitzen
Read another response about Philosophy, Science