The present study aimed to compare attributional styles among male and female secondary school students with academic failure in mathematics and to examine differences in the dimensions of effort, ability, conditions, luck, and task difficulty. This study employed a non-experimental causal-comparative design. The statistical population consisted of male and female second-grade high school students in Behbahan City studying in mathematics and experimental sciences who experienced academic failure in mathematics. The sample included 143 students (83 females and 60 males) selected through census sampling. Students with scores below 10 in mathematics during the first academic term of the 2025–2026 academic year were included. Data were collected using the Mathematics Attribution Questionnaire, which consisted of 17 four-point items assessing five attribution dimensions: effort, ability, luck, conditions, and task difficulty. The questionnaire demonstrated a Cronbach’s alpha reliability coefficient of 0.83. Descriptive statistics and independent samples t-tests were used for data analysis. The findings indicated that both male and female students used all attribution dimensions to explain their academic failure in mathematics. Among male students, effort obtained the highest mean score, whereas among female students, environmental conditions and task difficulty showed the highest means. Independent samples t-test results revealed significant gender differences in effort, conditions, luck, and task difficulty (P<0.001), while no significant difference was observed in the ability dimension (P>0.05). The findings further demonstrated that male students relied more on internal attributions, whereas female students relied more on external attributions to explain their mathematics failure. The findings suggest that attributional styles play a significant role in explaining academic failure in mathematics. Male students’ attribution of failure to effort may promote responsibility and motivation for academic improvement, whereas female students’ attribution to environmental conditions and task difficulty may contribute to reduced motivation and learned helplessness. Therefore, modifying maladaptive attributional styles and strengthening internal and controllable attributions may help reduce mathematics academic failure among secondary school students.