Authors

Ketan Sinha, Carol Kakileti

Research Question

This project rebuilds and formally expands a data model that was originally developed from data collected in 2019. While the initial scope of the project was based on work from Raj Chetty on the importance of neighborhood conditions, this project is pushing forward both the theoretical work on the relationships between neighborhood factors and academic outcomes and to drive programmatic implementation of interventions to adjust structural obstacles that people experience daily but may struggle to articulate.

Background

ImpactTulsa is a collective impact organization that turns data into systemic change by convening community sectors through the shared vision that all children in Tulsa receive every opportunity to succeed, from cradle to career.

This project uses socio-economic, population health, safety/crime, neighborhood access including walkability, and housing data compiled at a census tract level to determine relationships in cooccurring structural barriers across Tulsa County. Relationships between these neighborhood conditions are examined through exploratory factor analysis and possible further structural equation modeling. Geographical patterns in the analysis are assessed through census tract mapping of the conditions. Further visualizations involving the combinations of various factors and their geographic relationship are being explored throughout the project and developed into visualizations for inclusion on the ImpactTulsa data dashboard.

Methodology

The original modeling resembles work from Raj Chetty and the "Zipcode Matters" documentary adjusting outcome variables of interest from future earnings to student testing outcomes and performance, suspensions, and neighborhood homicides.

Deliverables

• Development of a model, including exploratory factor analysis and structural equation modeling.
• Geographical mapping and visualization practices targeted toward engaging non-expert audiences.
• Development of a data dashboard.

Data Sources

The student team is responsible for all external data sourcing with publicly available datasets. ImpactTulsa handles modeling and connection to protected data sources from school districts, the City of Tulsa, and the police department.