Students’ miscalibration is the inaccuracy in judgment between their perceived academic performance and actual performance, including over- and underestimation. Empirical evidence shows the prevalence of the tendency of students to be overconfident. In this paper, we analyze two types of interventions aiming to improve students’ accuracy of guessing their grades. Monetary incentives and item-specific individual feedback are implemented, sole or combined, as treatment variables in a randomized field experiment designed to analyze students’ calibration of their own academic performance. In this field experiment, students enrolled in a Microeconomics course are offered the possibility of guessing their academic performance in three course assessments involving 10 four-choice questions corresponding to the course topic recently taught. We elicit two types of guesses, depending on the timing in which students’ forecasts are requested: predictions (elicited immediately before the exam) and post-dictions (elicited immediately after they have completed the exam). These two guesses allow us to analyze two types of miscalibration: students’ miscalibration of their potential academic performance (defined as the difference between the obtained grade and the grade predicted) and students’ miscalibration of their actual academic performance (interpreted as the difference between the accomplished grade and the grade post-dicted). In addition to the two treatment variables, we control for individual characteristics considered as potential factors driving students’ miscalibration. These characteristics are: (1) the potential skill (by means of a reasoning ability test), (2) the actual skill (measured through the academic record), (3) the risk aversion (elicited using the Holt and Laury (2002) Lotteries task), (4) the personality traits (measured using the Big Five Model), and (5) the gender. The results point out that, although students' miscalibration does not depend on treatment variables, this is significantly reduced to a greater extent in the subsequent tests compared to the first one when individual feedback and monetary incentives are provided together than when they are not offered.