Th e current research explores the infl uence of multiple factors (such as class attendance, previous academic performance, in-class involvement and attention, class schedule, gender and other control variables) on the academic performance of public-administration undergraduate students. Th e regression models developed based on the academic literature were tested on a sample of 1 st -year students (N = 115) enrolled in the Public Administration bachelor program of Babeș-Bolyai University (Cluj-Napoca) in order to explain their performance (grade) in a fi nal examination. Since none of the variables included in the model are self-reported (i.e. classic self-administered questionnaires) we have reduced the potential that social desirability bias could infl uence our results, thus strengthening the reliability and robustness of our fi ndings.Our results show that the main factors which can infl uence students' academic performance are attendance at seminars and in-class involvement. All other factors which were included in the regression model (age, gender, distance between class site and their home, residence in urban or rural environment, attention and motivation, class schedule and four proxies / measures for previous performance) were either not statistically signifi cant in any of the models or showed an inconsistent / unstable infl uence on academic performance.Th ese results can be of interest not only for Public Administration scholars, but also for university decision makers. As such, starting from the aforementioned