How do beliefs and feedback affect decision making among students? Evidence from a field experiment

Abstract:
Evidence from laboratory experiments shows that individuals hold biased beliefs about their own ability. However, little is known about how beliefs impact real-life decision making, which often requires assessing own ability relative to others. I conduct a field experiment with students preparing for a college entrance exam at a test preparation center in Colombia to document biases in beliefs and to test the impact of providing relative-performance feedback on academic choices. Over 10 rounds, I elicit beliefs about relative performance in weekly practice tests and provide feedback to treated students about their actual standing in the score distribution. Overall, I find that over 50% of students either over- or under-estimate relative performance in math and reading, and that receiving feedback makes high-performers more accurate in their performance assessments. University administrative records show that high- and low-performing treated students are 5.8 and 9.5 pp less likely to take the entrance exam, respectively. The evidence suggests that low-performers realize they may need additional time to prepare and may be more resilient. Those who took the exam but were not admitted are 12 pp more likely to register for the next admissions process than low-performers in the control group.