GSE/RES: Promoting Competence in Mathematics Through Collaboration, Reflection, and Role Models

University Of Southern California, Los Angeles CA

Investigators

Abstract

University of Southern California is studying new tools for enabling math learning in high school students. The tools are particular features built into an intelligent tutor: chat for collaboration, capturing text for reflection on problem solving, and mathematics mentoring through virtual role models based on real scientists representing a range of gender, ethnic, cultural, and language backgrounds. The features are introduced to address learning preferences of female students and to appeal to ethnic minority students. A certain level of mathematical competence is required for participation in many STEM fields of study, and high-stakes test such as the GRE-Math and SAT-Math serve as "critical filters" for access to such programs. Unfortunately, many students, including young women as well as students from ethnic groups traditionally under represented in science and engineering, are at a disadvantage due to poor performance on mathematics achievement tests. Recent research points to differences in the strategies that males and females adopt in mathematics problem solving. Strategies that are used more often by males predict higher scores on achievement tests. On the other hand, other research suggests that there are significant learning benefits to collaborative learning, inquiry and discovery-oriented activities, and interaction with mentors and role models: strategies that female students more often adopt and prefer. Therefore, the central research question addressed in the proposed project is to evaluate the benefits of female-preferred learning strategies for mathematical competence in high school girls through innovative software tools integrated into Wayang Outpost, an intelligent tutoring system providing individualized instruction in mathematics problem solving. In addition, the project will evaluate the benefits of the enhanced software for students (males as well as females) who find potential failure in mathematics to be threatening, particularly those from ethnic groups that have been traditionally under-represented in STEM participation. The three specific objectives focus on the creation, implementation and evaluation of innovative new software tools for promoting collaboration, reflection, and mathematics mentoring through virtual role models. Evaluation studies will focus on the effects of different versions of the tutoring software on student learning outcomes in mathematics and on attitudes towards STEM careers. Participants in the evaluation studies will be students attending high schools in Pasadena CA, representing the great diversity of urban California - nearly 50% Hispanic, 30% African American, 33% with limited English proficiency, and 66% receiving subsidized lunches. The results of the project will have broad societal impact by enhancing the ability of students from traditionally under represented groups to participate in STEM fields of study through freely-accessible web-based tutoring. The intellectual merit of the research will be the careful empirical evaluation of the benefits of instruction that supports learning styles and strategies preferred by females and students vulnerable to stereotype threat in mathematics learning, and the evaluation of the potential benefits of virtual role model characters representing real STEM professionals for student learning outcomes in mathematics.