Choice and Consequence: Assessing Mismatch at Chicago Exam Schools

This quasi-experimental study by Joshua D. Angrist, Parag A. Pathak, and Román Andrés Zárate examines the impact of affirmative action for students from low-income neighborhoods by comparing students just above and just below the cutoff for selective exam schools. Students just above the cutoff are more likely to attend an exam school than their peers below the cutoff and, as a result, have lower high school math scores. The authors find that this is because students below the cutoff are more likely to attend Noble Network charter schools which significantly increase math scores. This result supports the idea that affirmative action causes student-school mismatch and harms student performance, but the authors did not find evidence that this mismatch was related to the student’s race or to being academically unprepared for the exam school.