Kacprzyk, JoannaJoannaKacprzykParsons, Martin E.M.Martin E.M.ParsonsMaguire, Patricia B.Patricia B.MaguireStewart, GavinGavinStewart2022-01-132022-01-132019 Educa2019-07-25Irish Educational Studies0332-3315http://hdl.handle.net/10197/12724The optimum assessment structure measures student knowledge accurately and without bias. In this study, the performance of the first-year undergraduate science students from the University College Dublin was evaluated to test the gender equality of the assessment structure in place. Results of male and female students taking three life science modules were analysed, for two academic years, with assessment structure based on a combination of three types of evaluation: continuous assessment and multiple choice questions (MCQ) exam scored with/without negative marking. We found no significant gender effect associated with performance in continuous assessment, or MCQ exams scored without negative marking. However, a significant bias against females was consistently observed for the same cohort of students in the MCQ exams with negative marking of 0.25 points. This bias was at least partially linked to a gender difference in willingness to guess and preliminary data suggest that it disappears after removal of negative marking from the MCQ exams. Our results support the view of a diverse assessment structure being fairer to the students. Moreover, caution is advised while using negative marking, and regular reviews of assessment strategy should be implemented by higher education institutions to ensure gender-bias free evaluation of students’ performance.enThis is an Accepted Manuscript of an article published by Taylor & Francis in Irish Educational Studies on 25 July 2019, available online: http://www.tandfonline.com/10.1080/03323315.2019.1645721.Gender-biasUndergraduate science assessmentNegative markingAchievementPersonalityPerformancePreferenceBiologyJournal Article38446748010.1080/03323315.2019.16457212022-01-04ERC-2012-StG311000https://creativecommons.org/licenses/by-nc-nd/3.0/ie/