Personalized Adaptive Learning in Undergraduate Mathematics: A Meta-Analysis

The University Of Central Florida Board Of Trustees, Orlando FL

Investigators

Abstract

This project aims to serve the national interest by examining prior research on personalized adaptive learning (PAL) specific to mathematics curricula, to identify where, when, and for whom PAL is helpful toward improving undergraduate students’ mathematics performance. PAL uses intelligent learning systems, incorporates preferences of the learner, and analyzes and incorporates individual learning data to create a unique learner pathway to maximize the chance of student success. This project will examine PAL within the population of undergraduate students enrolled in mathematics courses broadly along with a more focused study of its use in algebra. Each study will examine the overall student population as well as underserved students. Using meta-analytic procedures, the proposed research will investigate the extent to which PAL is effective in improving student learning outcomes in undergraduate mathematics courses and will provide insight on PAL for improving mathematics instruction, including mathematics course design and redesign. The meta-analytic approach will examine published literature, synthesizing findings across single studies using quantitative statistics to compute an overall measure of effectiveness. The results of this study are expected to help increase undergraduate student success in mathematics, undergraduate student retention and degree completion, and access to career opportunities and economic stability. The meta-analysis will begin with a thorough literature review, conducted using comprehensive keywords. Studies identified in the literature review will be examined to determine if they meet inclusion criteria. Abstracts of included studies will be screened by two coders, followed by full-text reviews. At the full-text stage, quantitative data needed for computing effect sizes will be extracted and then analyzed using multilevel meta-analysis. Moderator variables coded will include: student, type of PAL, class, and institutional characteristics. Robust variance estimation will be used to estimate standard errors and hypothesis tests, addressing dependencies of effect sizes within studies. Subgroup analyses will be used to evaluate the relationship between moderator variables and the magnitude of the effect size. Cochran’s nine standard criteria for assessing risk of bias in studies with a separate control group will be applied. Reporting bias will be assessed visually with funnel plots which assist in identifying small-study effects (of which, non-reporting bias may be the culprit). Dissemination efforts will aim to reach general audiences, practitioners, and researchers/scholars. The results have implications for improving faculty professional development, mathematics course redevelopment and redesign, course support services, and student advising, all of which may help to improve student success in mathematics. The NSF IUSE: EHR Program supports research and development projects to improve the effectiveness of STEM education for all students. Through its Engaged Student Learning track, the program supports the creation, exploration, and implementation of promising practices and tools. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.