Solving the Gender Equation? Measuring and Refuting Teachers' Math-Gender Misconceptions

Abstract

This cumulative dissertation contains three publications/manuscripts, researching (1) teachers' math-gender misconceptions (specific, plausible-sounding yet scientifically unsupported knowledge conceptions that contradict empiric evidence and/or established theoretical frameworks, e.g., "Girls think more empathically than boys, but lack systematic thinking abilities and are hence not as good in mathematics."), their relation to their explicit math-gender stereotype (the mere connotation of mathematics as male), (2) their relation to teachers' awareness of math-gender stereotype-reinforcing behaviors in the classroom (e.g., stereotypical task selection, hampering female learners' self-efficacy), and (3) the effectiveness of personalized refutation texts in reducing teachers' math-gender misconceptions (in comparison to non-personalized refutation text, common expository text and personalized expository text). Study 1 introduces and validates the Math-Gender Misconception Questionnaire (MGMQ), with three distinct but related misconceptions (Empathizing-Systemizing, Girls' Compensation and Girls' Non-Compensability). Nearly half of the 303 researched pre-service teachers endorsed at least one misconception. These misconceptions were correlated with the explicit math-gender stereotype, suggesting that math-gender misconceptions and the math-gender stereotype are two related but distinct constructs and that math-gender misconceptions contribute to the manifestation of the math-gender stereotype. Study 2 examines 278 (student and in-service) teachers’ awareness of stereotype-reinforcing teaching behaviors in the classroom using video vignettes depicting a male vs. female model teacher. Teachers reliably detected explicit but rarely implicit biased behaviors. Teachers were more aware of a male vs. female model teacher's math-gender stereotype-reinforcing behaviors. Furthermore, model-observer-gender-similarity improved awareness. Awareness correlated negatively with misconception endorsement, suggesting that misconceptions may influence teachers’ interpretation of, and potentially their engagement in, classroom behavior. Study 3 tests the effectiveness of personalized refutation texts (vs. standard refutation text, personalized expository text and standard expository text) in promoting conceptual change among 336 student teachers and in-service teachers. Personalized refutation fostered the strongest misconception reduction among teachers who initially held math-gender misconceptions, but also elicted backfire effects—ironic misconception reinforcement—in participants without prior misconceptions. Effects on broader attitudes (e.g., math-gender stereotypes) were limited, though reduced misconceptions were associated with more gender-equitable views, suggesting that future interventions should not only be personalized but tailored to teachers' prior knowledge. Methodologically, the dissertation contributes two validated tools—the Math-Gender Misconception Questionnaire and a video-based awareness measure—that enable precise assessment of misconceptions and teachers’ sensitivity to stereotype-reinforcing behaviors, and that can also be meaningfully applied in teacher training. Theoretically, the findings clarify how misconceptions are embedded within broader stereotype systems and identify conditions under which personalized refutation texts promote or hinder conceptual change. Practically, the results underscore the importance of explicitly addressing math-gender misconceptions in teacher education and show that well-aligned, personalized interventions can effectively reduce misconceptions, whereas misaligned feedback may trigger defensive reactions.