Unpacking Education Choice: Leveraging Machine Learning to Explore School and Homeschool Decisions (Poster 4)

• In Event: Division D Graduate Student Committee In-Progress Research Gala Poster Session

Abstract

This study examines how family demographics (e.g., income, location) and preferences (e.g., safety, curriculum, special education) shape parental school choice across public, private, and homeschooling options. Using NHES: 2019 Parent and Family Involvement Survey data, it integrates homeschooling into the broader school choice research. Employing advanced machine learning methods— K-Nearest Neighbors, Decision Tree, Random Forest, and Multinomial Logistic Regression—it identifies key predictors, including safety, instructional quality, race/ethnicity, and parental language. Results highlight stronger predictive accuracy for homeschooling and challenges for other options, offering a holistic view of school choice dynamics to inform policymakers and stakeholders within education.

Authors

• Cammie Justus-Smith, Virginia Commonwealth University

• Tzu-Wei Wang, Virginia Commonwealth University

• Georgia Heyward, Virginia Commonwealth University