Skip to content

Prather

Project Overview

The Prather project explored new computational approaches to understanding how children learn mathematics by creating models that capture the many interacting factors influencing learning—from cognitive processes and executive function skills (EFs) to students’ home environments, family supports, and classroom experiences.

The team explored a rich data set with longitudinal mathematics assessment data for more than 8,000 students in grades 4-8 and modeled the relationships between learners’ internal cognitive processes, socio-emotional factors, and external contexts. Using this dataset, they conducted exploratory analyses to test approaches for representing learners within a multidimensional “context space” that reflects the interaction of these various factors. This work was complemented by qualitative interviews that explored how families, teachers, classrooms, and home environments impact executive functioning, social emotional development, and aspects of attitudes towards mathematics.

Together, this work contributed early insights into how complex models that incorporate cultural, developmental, and societal contexts might be used to better understand patterns of productive struggle in math learning and to inform the design of more equitable educational tools and supports.

Lead Organization University of Maryland, College Park
Funding Period May 2020 – November 2023

Project Approach

The Prather project’s approach combines computational modeling, qualitative insights, and equity-focused design to help educators and researchers understand how cognitive, socio-emotional, and contextual factors interact to shape students’ math learning.

Computational Modeling

A computational model of learners to evaluate how contextual variables may be associated with math outcomes.

Student and Family Perspectives

Qualitative interviews to understand how students’ home environments, family support, and classroom experiences can shape the development of EF skills and mathematical identity.

Designed for All Learners

A holistic approach to modeling that accounts for internal cognitive processes, affective processes, and the learner’s external environment.

Key Insights and Innovations

Context space modeling reveals the complexity of math learning

The Prather team developed exploratory approaches to representing students within a multidimensional “context space” that captures how cognitive processes, socio-emotional factors, and environmental contexts (such as school and community) interact to shape mathematics learning. By analyzing data from the Trends in International Mathematics and Science Study (TIMSS) for over 8,000 students, alongside contextual information from teachers, students, and principals, the Prather team tested approaches for representing learners within a multidimensional “context space” that reflects the interaction of these various factors.

Interviews with families and students provide essential context for quantitative findings

Through qualitative interviews with families and students in grades 3-5, including problem solving sessions and discussions about their attitudes towards mathematics, the Prather team gained insights into how home environments, family support, and classroom experiences can shape the development of EFs and mathematical identity. These interviews explored the cognitive processes and decision-making strategies students use when solving math problems, bringing forward context that computational models alone cannot capture, and centering the strengths and successes students bring to mathematics learning.

Project Team

Cameron Johnson

Research Assistant

Josh Medrano

Graduate Student Researcher

Joshua Taton

Educator Leadership Council

Lauren Kendall-Brooks

Postdoctoral Researcher

Richard Prather

Principal Investigator

Explore Our R&D Projects

The EF+Math portfolio of R&D projects developed innovative math learning products and advanced research on mathematics learning, executive function skills, and equitable learning experiences using inclusive R&D methods.