Developing an AI-Supported Approach to Identify Instructional Groupings in Early Childhood Education Classrooms. Technical White Paper
Saved in:
| Title: | Developing an AI-Supported Approach to Identify Instructional Groupings in Early Childhood Education Classrooms. Technical White Paper |
|---|---|
| Language: | English |
| Authors: | Aravind Sundaresan, Leigh Ann DeLyser, Gullnar Sy, Sarah Gerard, John Niekrasz, SRI International |
| Source: | SRI International. 2025. |
| Availability: | SRI International. 333 Ravenswood Avenue, Menlo Park, CA 94025. Tel: 650-859-2000; e-mail: customer.service@sri.com; Web site: https://www.sri.com/ |
| Peer Reviewed: | N |
| Page Count: | 20 |
| Publication Date: | 2025 |
| Sponsoring Agency: | Gates Foundation |
| Document Type: | Reports - Research |
| Education Level: | Early Childhood Education Preschool Education |
| Descriptors: | Artificial Intelligence, Grouping (Instructional Purposes), Technology Uses in Education, Early Childhood Education, Video Technology, Preschool Education, Documentation, Classification |
| Abstract: | A high-quality early childhood classroom is alive with activity. Children may be rotating from whole-group morning meeting to free play centers and then to small-group math instruction, all before lunch. Pre-K teachers looking to make the most of these instructional groupings--and those who support them, such as instructional coaches--can benefit from understanding how classroom time is spent to best focus instructional goals. Many early childhood providers use video to support teacher development, classroom observations, or even opportunities for parents to look in on their children during the school day. Researchers from SRI, supported by the Gates Foundation, explored opportunities to leverage AI and early childhood classroom videos to develop foundational frameworks and AI-supported approaches that promote high quality teaching and learning. After interviews with curriculum providers, early childhood education (ECE) leaders and educators, and funders, the research team selected instructional groupings as a fundamental building block that could be used to support coaching, curriculum implementation at scale, and data for ECE providers for review of implementation. This technical white paper describes the research team's approach to automatically identifying instructional groupings, the data used, and initial performance of tested models on the dataset. We close with a discussion of future implications of this work. |
| Abstractor: | As Provided |
| Entry Date: | 2026 |
| Accession Number: | ED679863 |
| Database: | ERIC |
| Abstract: | A high-quality early childhood classroom is alive with activity. Children may be rotating from whole-group morning meeting to free play centers and then to small-group math instruction, all before lunch. Pre-K teachers looking to make the most of these instructional groupings--and those who support them, such as instructional coaches--can benefit from understanding how classroom time is spent to best focus instructional goals. Many early childhood providers use video to support teacher development, classroom observations, or even opportunities for parents to look in on their children during the school day. Researchers from SRI, supported by the Gates Foundation, explored opportunities to leverage AI and early childhood classroom videos to develop foundational frameworks and AI-supported approaches that promote high quality teaching and learning. After interviews with curriculum providers, early childhood education (ECE) leaders and educators, and funders, the research team selected instructional groupings as a fundamental building block that could be used to support coaching, curriculum implementation at scale, and data for ECE providers for review of implementation. This technical white paper describes the research team's approach to automatically identifying instructional groupings, the data used, and initial performance of tested models on the dataset. We close with a discussion of future implications of this work. |
|---|
