Authors

Mengfei Gao, Tim Yupeng Sun, Ziyi Wang

Research Question

How do different urban features influence depression rates through causal mechanisms?

Background

Racial disparities remain a persistent concern in the U.S. criminal justice system. Recent efforts to reform practices such as traffic stops show significant disparities by race. Connecticut has emerged as a leader, pioneering legislative reforms that mandate the collection and public disclosure of traffic-stop data, while also restricting the grounds on which law enforcement can initiate stops. This project examines racial, ethnic, and gender-based disparities in traffic stops in Connecticut.

Methodology

Using a dataset spanning 2016 – 2022, this research investigates whether minority groups experience disproportionately higher rates of stops, searches, and adverse outcomes. The methodology includes data collection, cleaning, and analysis. Data preprocessing involved standardizing place names, addressing missing values, and geocoding stop locations using GEO API. Data analysis includes the application of statistical and spatial techniques. Descriptive Statistics are used to examine demographic patterns in stop rates. Logistic Regression Models are employed to assess whether race, ethnicity, or gender significantly predict stop outcomes. Additionally, ArcGIS is used to map traffic stop locations, identify high-incidence areas, and generate heat maps highlighting potential biases.

Deliverables

• Interactive Dashboard for trend analysis
• Technical Report

Datasets