Phase I CAMEL-CN: Building Integrated Data for Education, AI, and Learning to Support K-12 Mathematics Research and Practice (IDEAL-Math)

Early mathematical learning builds an important foundation for later math achievement and STEM careers. Large datasets play a unique and essential role in advancing our understanding of students’ mathematics learning and how teachers and schools can support that early learning. This project integrates six existing datasets spanning 90,000 elementary school students and their 3,000 teachers into a combined dataset focused on numerosity and mathematical operations (e.g., addition, subtraction). A team of science of learning researchers, data scientists, artificial intelligence/machine learning researchers, and educational practitioners and policy makers work together to integrate the dataset, curate the data, and make the data available for broader use in mathematics learning and education applications. The project aims to create an accessible, AI-ready, integrated, dataset that represents a large number of elementary students and their teachers. The integrated dataset is grounded in a model in which student math achievement is influenced by teacher professional development, teacher knowledge, instructional practices, classroom environment, and student math attitudes. The data include student and teacher surveys and assessments, classroom videos, and student think-aloud videos, all with many overlapping measures across datasets, covering teacher professional development, teacher math knowledge, instructional practices and class environment, student math attitudes, and student math strategies and math achievement. The first goal is to harmonize and curate the data from six research projects. Members of the interdisciplinary network plan to identify several research questions of interest that can be answered with the integrated data, which will guide the harmonization and documentation. The other goals relate to increasing access and usability of the dataset, with plans to develop a dashboard and hold workshops for potential users of the dataset. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria. NSF Award ID: 2621428 | Program: 4082PYXXDB NSF TRUST FUND | Principal Investigator: Colleen Ganley | Institution: Florida State University, TALLAHASSEE, FL | Award Amount: $1,509,240 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2621428 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2621428.html