Near-peer mentoring in data science: Two experiences at Stanford University

2022-11-16

AI Summary

This study addresses inequitable access to undergraduate research opportunities in data science—particularly for underrepresented students and those from non-R1 institutions—alongside insufficient pedagogical training for graduate student educators. We designed and implemented a “Near-Peer Mentorship” dual-track model, delivered via summer research programs and for-credit courses, integrating experiential learning, structured workshops, synchronous online one-on-one mentoring, and cross-institutional mentor–mentee matching. The model introduces a novel framework that embeds socially impactful practice with systemic, equity-centered mentor training. Results demonstrate significant gains in undergraduates’ research engagement and sense of academic belonging, while simultaneously strengthening graduate students’ teaching leadership, culturally responsive mentoring competencies, and instructional self-efficacy. The resulting inclusive data science education prototype is scalable, reusable, and empirically grounded—offering both theoretical insights and an actionable pedagogical paradigm for equitable, high-engagement undergraduate research training across STEM disciplines.

Abstract

Universities have been expanding the data science programs for undergraduate students, with the simultaneous goal of reaching and retaining students from underrepresented groups in the data science workforce. The set of new programs also offer opportunities to involve graduate students, fostering their growth as future leaders in data science education. We describe two programs that use the near peer mentoring structure to provide pathways for graduate students to develop teaching and mentoring skills, while providing research and learning opportunities for undergraduate students from diverse backgrounds. In the Data Science for Social Good Summer program, graduate students mentor a group of undergraduate fellows as they tackle a data science project with positive social impact. In the Inclusive Mentoring in Data Science course, graduate students participate in workshops on effective and inclusive mentorship strategies. In an experiential learning framework, they are paired with undergraduate students from non-R1 schools, who they mentor through weekly one-on-one on-line meetings. These initiatives offer a prototype of future programs that serve the dual goal of providing both hands-on mentoring experience for graduate students and research opportunities for undergraduate students, in a high-touch inclusive and encouraging environment.

Problem

• Graduate students mentoring undergraduates in data science
• Strengthening teaching and mentoring skills through near-peer programs
• Providing research experiences for diverse undergraduate backgrounds

Innovation

• Near-peer mentoring structure implementation
• Self-paced mentor training guide development
• Experiential learning framework with remote meetings