Do Grades Enhance learning? Evidence from a Natural Experiment in a Liberal Arts College

Abstract

This paper examines how grades affect students' learning using data from a liberal arts college in Singapore. We examine a natural experiment where, in 2017, the grading policy for an introductory statistics module was exogenously changed from letter grade to pass/fail, but the content, format, difficulty, and modes of assessment remained exactly the same. Our dataset includes all results from individual twice-weekly quizzes and two exams for five cohorts | three under letter-grade and two under pass/fail. Our results show that pass/fail students initially work as hard as letter-graded students, but tend to lower their effort as the course goes on | with a sharp decrease right after the first exam. This suggests that, as soon as students realize they have secured a \pass", they choose to allocate their time and effort to other subjects or activities. We find that intending to major closely related to the subject does not explain relatively better performance under a pass/fail scheme. This suggests that students who exert consistent effort regardless of the grading scheme are either those who are naturally gifted or those who are driven by personal ambition (whether to learn, score well, or outperform their peers). Our results suggest that pass/fail policies could safely be implemented more extensively: the loss in acquired knowledge is marginal and could easily be compensated in terms of personal time and mental health.

Keywords: Education; Grades; Effort; Learning