Date of Award

2026

Degree Type

Open Access Dissertation

Degree Name

Education, PhD

Program

School of Educational Studies

Advisor/Supervisor/Committee Chair

Guan K. Saw

Dissertation or Thesis Committee Member

Mallika Scott

Dissertation or Thesis Committee Member

Emilie Reagan

Dissertation or Thesis Committee Member

Thomas Luschei

Terms of Use & License Information

Creative Commons Attribution 4.0 License
This work is licensed under a Creative Commons Attribution 4.0 License.

Rights Information

© 2026 Kimberly A. Megyesi-Brem

Keywords

international comparative education, math and gender, math self-concept, multilevel structural equation modeling, socio-ecological framework

Subject Categories

Education

Abstract

In Japan and the United States, women’s underrepresentation in STEM fields in tertiary education limits opportunities for social mobility and economic stability. Efforts to address this issue need to consider the role played by the intractable girls’ inferiority in math (GIM) myth that contributes to girls’ lower engagement in math through lower math self-concepts. Prior research shows that perceived teacher support through is one area where the GIM myth relates to girls’ math self-concepts through teacher bias about girls’ math competence. However, research has predominantly been focused at the individual level, including studies that conclude that girls’ lower math self-concepts are related to girls being more self-critical, using more social comparison, and being more sensitive to feedback. Systems-level research is needed which considers predictors of middle school girls’ math self-concepts as interconnected and in relationship to the intractable and pervasive GIM myth. The present dissertation study utilizes the structure of Louie and Zhan’s (2022), socio-ecological framework to explore interrelationships among middle school girls’ math self-concepts across five levels. The theoretical framework of status construction theory (Ridgeway & Correll, 2006) facilitates a consideration of how sociocultural structural conditions may relate to middle school girls’ math self-concepts in the classroom by assigning status to competence based on social characteristics such as gender. Multiple group multilevel structural equation modeling in Mplus was employed with data from the Trends in International Mathematics and Science Study (TIMSS 2019) to consider student- and classroom-level predictors including math competence affirming feedback, classroom math achievement, classroom SES, and female representation. Key findings include that math competence affirming feedback partially explained middle school girls’ math self-concepts in both Japan and the U.S. This supports research that the GIM myth may have the potential influence girls’ math self-concepts through teacher bias in a range of educational contexts. In addition, lower math achievement relative to male classmates partially explained girls’ math self-concepts in the U.S., although girls’ overall math achievement was not significantly different from boys. Differences in girls’ and boys’ math self-concepts were not directly associated with classroom-level factors, although all students were more likely to have higher math self-concepts when they had a math competence affirming teacher and when they had more female classmates, suggesting that math self-concepts may be connected to class climate, for example possibly through an increased emphasis on collaboration over competition. Contextual effects demonstrated that in both countries, both girls and boys reported lower math competence beliefs in higher-achieving math classes, although lower effects than other big-fish-little-pond effect studies supports prior research suggesting that perceived teacher support may attenuate this effect. In the U.S., a small negative direct effect between class SES and affective math self-concept was observed after accounting for the positive indirect effect through math achievement. Prior research has found decreased interest in math in wealthier countries related to increased gender alignment through self-expression. It is possible that an increased pressure to align with stereotyped interests might also play a role in math self-concepts through an association with collective SES at the classroom level. Exploratory analyses considered country-level differences in the inner-level empirical results in Japan, a country with less educational stratification and tighter gender norms, and the U. S., a country with more educational stratification and looser gender norms. For example, in the U.S., the negative relationship between class math achievement and students’ math self-concepts might have been associated with increased social comparison in a stratified educational system where math class labels are salient in day-to-day school activities. However, in Japan, where ability grouping does not begin until after middle school, lower math self-concepts in higher achieving math classes might instead reflect increased social comparison related to collective high school entrance exam pressure at the classroom level. Recommendations to educators and policymakers to limit the potential influence of the GIM myth on middle school girls’ math self-concepts include the importance of distributing resources equitably, including through efforts to address teacher bias in math competence affirming feedback, minimizing social comparison in educational settings, and adopting a more expansive notion of mathematical competence. Future research is needed, including to understand possible connections between educational stratification and tighter gender roles at the macro level with girls’ lower math self-concepts. It is hoped that reconsidering middle school girls’ math self-concepts as related to a broader GIM myth will provide educators and advocates with new avenues for supporting girls’ future opportunities for economic stability.

ISBN

9798247942030

Download

Included in

Education Commons

Share

COinS