This article presents a method for computer-aided tutor evaluation: Bayesian Networks are used for organizing the collected data about tutors and for enabling accurate estimations and predictions about future tutor behavior. The model provides indications about each tutor's strengths and weaknesses, which enables the evaluator to exploit strengths to the benefit of the university and offer advice for tutors' improvement. It also allows the evaluator to make hypotheses about potential tutor approaches and test the effect of such approaches on the educational procedure in advance. The article briefly discusses Bayesian Networks and introduces a model that has been used at the Hellenic Open University for aiding tutor evaluation.