Gender, knowledge and resilience in evaluation processes

Abstract

This study examines gender differences in the impact of low-stakes test penalties on academic performance for a first-year undergraduate applied economics course at a public university in Spain using a quasi-experimental design. The study explores how the presence or absence of these penalties for incorrect answers influences students’ grades and whether this impact differs by gender. Statistical analyses include descriptive statistics, permutation tests, bootstrap simulations and linear regression models with interaction terms to explore the complex relationships between test penalties, gender ratios and academic outcomes. The results reveal that removing penalties for incorrect answers increases students’ overall grades significantly and female students benefit relatively more from non-penalising environments. Furthermore, the interaction between the penalty level and the proportion of female students in the class shows that increased penalties affect their grades disproportionately, indicating a gender-specific sensitivity to risk. We also demonstrate that while the presence of female students has a positive influence on overall grades, the combined effect of high penalties and a higher proportion of female students results in diminished performance, underscoring the impact of assessment strategies on gender disparities in academic outcomes. These conclusions are based on simulated classroom configurations derived from resampled individual-level data. The observed interaction patterns indicate differentiated adaptive responses to institutional stressors.