Beyond Truth-Telling: Preference Estimation with Centralized School Choice and College Admissions

The authors propose new approaches for estimating students’ preferences using data from matching mechanisms, in particular the Gale-Shapley deferred acceptance mechanism. Even though this mechanism is strategy-proof, assuming that students truthfully rank schools in their applications can prove restrictive. The authors show that when students are strictly ranked according to a priority index known ex ante (for example, exam results), stability constitutes a plausible and weaker assumption, implying that each student is matched to their preferred school among those to which they are admissible ex post. These methods are illustrated through simulations and applied to school choice in Paris. The authors examine the circumstances under which each approach is most appropriate in real-world situations.