Programming Errors and Academic Performance in an Introductory Data Structures Course: A Per Gender Analysis

Computer Science studies is among the primary fields usually dominated by male audiences. But can the empirical data explain this preference or are other types of factors responsible for its perpetuation? This study aims to contribute by examining the student performance in a per gender manner from an introductory Data Structures course taught in the second semester of a university’s undergraduate program. The years whose data was used are 2021 and 2022. Visualization and statistical analysis tests are applied on the programming errors and grades per student as an attempt to monitor said performance per gender throughout the semester and determine if any differences arise. Association rule mining is also used in order to uncover the role of the students’ different attributes in shaping their course pass status. The findings suggest that the student’s gender does not considerably affect their performance, while the two genders’ results rarely were statistically different. Moreover, in all the cases where differences emerge, women are the gender with the higher academic performance.