By starting with a rating scale, then adding an open-ended follow-up question, I push users to reflect and elaborate on their first answer. Users were more likely to give a gut response, meaning we can be confident that we haven't received biased answers.