Education

Equitable Access to Quality Schools

Map of a single family’s “choice basket” from the Home-Based assignment system’s algorithm.

I have worked with my colleague Nancy Hill at Harvard Graduate School of Education and Boston Public Schools (BPS) to use data-driven approaches to identify and address inequities in academic opportunities. This has taken the form of two projects. First, in 2017 we constructed and implemented an Opportunity Index that quantifies both individual- and neighborhood-level impacts on achievement, producing an algorithm that is used to distribute $6M in resources to schools. Second, we evaluated BPS’ school algorithm-based choice and assignment system known as the Home-Based Assignment Plan (HBAP), finding that algorithms can only do so much to overcome inequities in the distribution of high-quality schools in a de facto segregated city. This work has been supported by the Boston Foundation and others.

Publications (students in bold)

  • O’Brien, D.T., Hill, N.E., Contreras, M. Community Violence and Academic Achievement: High-Crime Neighborhoods, Hotspot Streets, and the Geographic Scale of “Community”. 2021. Public Library of Science One. 16: e0258577.

  • O’Brien, D.T., Hill, N.E., Contreras, M., Sidoni, G. An Evaluation of Equity in the Boston Public Schools’ Home-Based Assignment Policy. 2018. Report from the Boston Area Research Initiative.

Grants

  • Boston Schools Fund. Support for a project titled, “Improving Equitable Access to High Quality Schools in Boston: Algorithms, Choice Patterns and School Options.” $96,137. PI, with co-PI Nancy E. Hill.

  • Boston Public Schools. Support for a bid titled, “Equity Analysis of the Home-Based Assignment Plan.” $56,289. 2017-2018. PI, with Nancy Hill.

  • Boston Public Schools, Lynch Foundation, The Boston Foundation. Support for a proposal titled, “Collaboration between Boston Public Schools and Boston Area Research Initiative on the Validity and Utility of the Opportunity Index.” $69,912.27. 2017. PI, with Nancy Hill.

Previous
Previous

“Seeing” Neighborhoods through Big Data