Popular belief suggests that private feedback is a more accurate reflection of the feedback giver’s true preferences than public feedback. However, our study challenges this assumption. Given the anonymity associated with private feedback, we anticipate that it will be prone to extremes. We assume, however, that these extreme responses will be symmetrical, providing true feedback on average. Are these common-sense assumptions correct? In this experimental study, we investigate whether the mode of collecting feedback influences the outcome in the context of teacher-student interactions. To account for the inherent power distance and perceived costs and benefits associated with repeated teacher-student interactions, we solicit feedback from students on one-time guest lecturers. We found that private feedback had a higher variance than public feedback, as expected. Private feedback, however, has a higher mean, median, and mode, contrary to popular belief. Because average student feedback is given significant weight in teacher performance evaluations, our findings have implications for higher education administration.