Digital Learning Challenge
Phase
Concluded
Finalist Teams
In August 2022, the independent judging panel selected three teams to move forward as Finalists in the last phase of the competition, the Demonstration Phase. These three teams, listed below, will spend the next six months scaling their solutions further and continuing to demonstrate the capabilities of their systems. The end of the phase – in March 2023 – also marks the end of the challenge at which point a Grand Prize winner will be determined and awarded $500,000 (USD).
Scroll to see more
| About the Team | Team Name | Location | Website | ||
|---|---|---|---|---|---|
| Adaptive Experimentation Accelerator (formerly known as HINTS-IAI) | Raleigh, NC | Link | |||
| The cross-platform infrastructure supports both traditional and adaptive experiments. Adaptive experiments change how often conditions are assigned as data is collected, such as using increasing evidence to give a condition more frequently to future students. This can accelerate the use of data to enhance and personalize learning for future students, by integrating machine learning algorithms with statistical models and human decision-making. Our team integrates expertise in education research, experimental psychology, human-computer interaction, large-scale software development, educational data mining, statistics, and machine learning. It includes professors Thomas Price from North Carolina State University, Joseph Jay Williams from University of Toronto, John Stamper from Carnegie Mellon, Norman Bier from CMU's Simon Initiative and the Open Learning Initiative (OLI). Graduate students include Ilya Musabirov, Steven Moore, Koby Choy, Pan Chen, Jiakai Shi, Harsh Kumar, & Mohi Reza. We have collectively deployed over 150 field experiments with more than 300 000 learners from 2012 to 2022, working with over 50 instructors and 40 researchers. | |||||
| Terracotta | Bloomington, IN | Link | |||
| Terracotta is a web application created with the goal of making experimental research easy and accessible across education levels, student populations, and learning materials. Terracotta is designed to be flexible and robust, enabling a range of experiment designs that reflect the complexity of instructional practice. Terracotta integrates with the Canvas learning management system (LMS) and handles the practical details of embedding experiments; it manages participant recruitment, random assignment, implementation, and de-identified data exports across multiple classes. By allowing teachers and researchers to embed studies in LMSes, where many learning activities already take place, Terracotta can advance our understanding of what works in student learning on a broad scale. | |||||
| UpGrade by Carnegie Learning (formerly known as Upgrade) | Pittsburgh, PA | Link | |||
| UpGrade by Carnegie Learning is composed of the nation's foremost learning engineers, software developers and educators who are passionate about using data to improve outcomes for students. Carnegie Learning is a leading provider of K-12 education technology, curriculum and professional learning solutions. With the highest quality offerings for K-12 math, literacy, world languages, professional learning, tutoring, and more, CL is changing the way we think about learning and creating powerful results for teachers and students alike. Born from more than 30 years of learning science research at Carnegie Mellon University, the company has become a nationally-recognized leader for its ability to harness the power of data to improve student performance. Carnegie Learning’s K-12 solutions are renowned for their basis in research and proven effectiveness in rigorous studies. Driven by our commitment to continuous improvement, our team has developed UpGrade, a free and open-source platform for conducting evidence-based testing in educational software. UpGrade will enable the research and educational technology community to better understand the factors that lead to improved effectiveness, and then to apply these best practices for optimal results. For the XPRIZE Digital Learning Challenge, the team/Carnegie Learning will use the UpGrade platform to conduct scalable, reproducible experiments within our MATHia software. Our experiments will test whether using localized character names within math word problems leads to a greater sense of belonging and math success, particularly among students in minoritized communities where students may have less-common first names. Headquartered in the historic Union Trust Building in Pittsburgh, Carnegie Learning's 700+ employees across the US and Canada – the majority of whom are former teachers – are passionate about partnering with teachers in the implementation of effective, student-centered instructional strategies, and supporting them in the classroom. For more information, visit carnegielearning.com and follow us on LinkedIn, Twitter and Instagram. | |||||