More Money, Fewer Problems?The Effects of Community School Funding on Student Behaviors in New York State

• In Event: District Policy Toward Educational Equity

Abstract

As policymakers increasingly invest in whole-child approaches like community schools, evidence on their impact—particularly on non-academic outcomes— remains limited. This study evaluates whether such investments improve student behavior, an essential but often overlooked dimension of school success. Using data from New York State public schools and a fuzzy difference-in-discontinuities (diff-in-disc) design., the study examines need-based funding targeted toward disadvantaged communities. It contributes to existing debates on whether “money matters” in education by extending the focus beyond test scores to behavioral outcomes and highlighting the broader role of community schools in promoting educational equity.

Since 2016, New York has allocated over $250 million annually to high-need districts through the Community School Set-Aside (CSS) fund, supporting students’ all-around development. This funding allows schools to implement community school models that connect students, families, and community organizations to address the multifaceted challenges faced by students with limited resources. Allowable expenditures include site coordinators, academic support, physical and mental health care, counseling, dental services, and coordination with service agencies. Eligibility is determined by a state funding formula classifying districts as “high-need” based on a composite index of student poverty, disability, and English learner status. Despite these substantial investments, limited research has explored whether such funding strategies—targeted for disadvantaged communities— affect non-academic outcomes. To address this gap, the current study examines the causal effect of CSS funding on student behavior, focusing on violent and disruptive incidents.

We construct a panel dataset of approximately 690 New York State school districts (excluding NYC) from 2011 to 2023, combining the School Safety and Educational Climate (SSEC) reports with financial data from the State Education Department and the Comptroller’s Office. The empirical strategy uses a fuzzy diff-in-disc design that leverages: (1) an arbitrary eligibility threshold; and (2) changes in funding availability over time. While funding amount is determined by the state funding formula, CSS eligibility depends on a threshold in the Need/Resource Capacity Index—a measure of a district's ability to meet the needs of its students with local resources.

In the first stage, we regress CSS funding on the interaction of post-treatment and above-threshold indicators, controlling for index score, district, and year fixed effects. The second stage regresses student behavioral outcomes on predicted CSS funding, using the same controls. This approach provides more flexibility over standard regression discontinuity (RD) designs and weakens the necessary assumptions for identification.

Preliminary descriptive analysis shows that districts receiving CSS funding serve significantly higher-need student populations than those that do not. They also face greater demands regarding student behavior, with lower attendance rates and higher suspension rates compared to their counterpart districts. We have secured via FOIL the necessary Need/Resource Capacity Index scores and are ready to proceed to estimation of impacts of CSS funding.

This research makes two main contributions. First, it brings behavioral outcomes into education finance debates, expanding the literature that is often focused solely on academic outcomes. Second, it provides practical policy insights for government agencies seeking to implement whole-child approaches for school improvement.

Authors

• Miyeun Jung, University at Albany
  Presenting Author

• Lucy C. Sorensen, University at Albany
  Non-Presenting Co-Author