The determination of college students' academic performance is an important issue in higher education. Whether students' attendance at lectures affects students' exam performance has received considerable attention. The authors conduct a randomized experiment to study the average attendance effect for students who choose to attend lectures, which is known in program evaluation literature as the average treatment effect on the treated. This effect has long been neglected by researchers when estimating the impact of lecture attendance on students' academic performance. Under the randomized experiment approach, the results suggest that class attendance has a positive and significant impact on college students' exam performance. On average, the effect of attending lectures corresponds to a 9.4 percent to 18.0 percent improvement in exam performance for those who choose to attend classes. In comparison, the improvement is only 5.1 percent, using the empirical method of existing studies, which measures the overall average attendance impact. (Contains 4 tables and 3 notes.)