Project Activities
The team will complete the Phase 1 scope in five tasks:
- Educator focus groups to identify a need and define indicators of success. These focus groups with teachers and instructional coaches will inform our needs hypothesis and key performance indicators.
- Educator co-development session to refine key performance indicators, improve theory of change, and identify core components. They will collaborate with an educator advisory board to refine our ideas before we begin development.
- Iteration on a minimum viable product (MVP). They will iterate on an MVP as we conduct user interviews with educators and then reflect on findings with our educator advisory board.
- Iteration on a prototype and market analysis. The team will build our MVP into a prototype, further iterating via user interviews with educators and leaders and then reflect once more with our advisory board.
- Capacity-building activities hosted by the Accelerate, Transition, Scale Initiative Hub.
These tasks will inform the project's proof-of-concept report, Phase 2 R & D plan, and early-stage prototype.
Key outcomes
REVEAL is intended to improve instructional quality and personalization by helping teachers to effectively and efficiently select appropriate, high-quality videos to use in instruction. Not only are videos engaging for children, but high-quality educational videos have been shown to support mathematics and literacy learning gains both in and out of the classroom. Most elementary teachers report they use YouTube videos in their instruction. Unfortunately, teachers find it difficult and time-consuming to curate high-quality online resources and to personalize instruction.
REVEAL will support teachers’ curation and personalization efforts by using machine learning to recommend high-quality, publicly available educational math and literacy videos targeted to individual students’ learning progress. The project team believes this will support students’ literacy and mathematics learning. REVEAL capitalizes on a unique interdisciplinary collaboration between machine-learning and education experts to combine multimodal content classification and a deep, granular understanding of relevant qualities of educational media. It will be the first standards-aligned, machine-learning video content classification and recommender system, to the team's knowledge.
People and institutions involved
IES program contact(s)
Project contributors
Questions about this project?
To answer additional questions about this project or provide feedback, please contact the program officer.
