Project Title

Hybrid Human-Agent Tutoring: Accelerating Middle School Math Achievement for Low Income Students

Project Summary

High-impact human tutoring is a proven, effective strategy to improve math learning outcomes. The main challenge; however, is how to scale while maintaining quality of service. This intervention is a hybrid human-AI tutoring (HAT) platform that supports critical components of tutoring work such as centering rich, collaborative math discourse and building caring relationships in order to improve tutoring processes and consequently student learning outcomes

The University of Colorado Boulder is developing HAT in the context of Saga, a large non-profit provider of tutoring services to Title 1 schools. Saga has already demonstrated the potential to double student math achievement for these students; and HAT is helping it scale by providing tutors with data-driven, job-embedded, high-quality professional learning to enhance their practice.

HAT analyzes small-group tutoring sessions for evidence of tutors’ use of high-impact discourse practices (based on research-based frameworks) and provides feedback visualizations on these practices to enhance professional learning with an instructional coach. Tutors who act on this feedback should improve their tutoring practices, leading to enhanced student achievement outcomes. 

HAT’s Impact

  • HAT was made available to 37 coaches who supported 172 tutors and over 4300 ninth-grade students. A study measured changes in tutors’ discourse practices before and after HAT was introduced, finding that coaches’ use of HAT led to long-term statistically significant improvements in three of six measured key tutor talk moves. 
  • Data collected from 1080 Saga students across 46 tutors and directly linked rigorous thinking tutor talk moves with math learning outcomes. 


What have tutors and coaches said about HAT?

  • Saga Coach, “[HAT] helps me, in a shorter amount of time, to give specific and directed feedback that is tangible for [tutors] to work on and see growth in.”
  • Saga Tutor, “As a tutor, I have found the AI hints and reminders before and after sessions helpful. Seeing a prompt right before the session gives a skill to focus on and be mindful of while tutoring. After sessions, it is helpful to see areas to focus on in later sessions and create a cycle of self analysis.”


Media

Learn more at:

or contact Peter Foltz at peter.foltz@colorado.edu or Brent Milne at bmilne@sagaeducation.org

Team Members

Sidney D’Mello
Professor in the Institute of Cognitive Science and Department of Computer Science at the University of Colorado Boulder
Tammy Sumner
Director of the Institute of Cognitive Science and Professor in Computer and Cognitive Science at the University of Colorado
Jennifer Jacobs
Associate Research Professor at Institute of Cognitive Science at the University of Colorado Boulder
Peter Foltz
Research Professor at Institute of Cognitive Science at the University of Colorado Boulder