Previous studies have found student procrastination to compromise learning outcomes using initial test scores to control for the influence of unobserved ability. The validity of such analysis rests on the assumption that students do not react to initial test scores. Utilizing daily information on student behavior, this paper shows that feedback effects were negligible in a student sample from a university second-language course. The paper then objectively quantifies the degree of procrastination, and finds evidence for detrimental effects of procrastination on test scores, corroborating previous studies. The result lends confidence to the value-added specification of the education production function.