# A Hierarchical Bayes Approach to Adjust for Selection Bias in Before-After Analyses of Vision Zero Policies

## TL;DR

This paper introduces a Bayesian hierarchical model to correct for selection bias in evaluating traffic safety policies, revealing that previous estimates of policy effectiveness may be significantly overstated.

## Contribution

It develops a novel hierarchical Bayesian approach to adjust for selection bias in before-after analyses of traffic safety interventions.

## Key findings

- Overestimation of policy effects in prior analyses
- Realistic effect size is approximately two-thirds of previous estimates
- Method improves accuracy of policy evaluation

## Abstract

American cities devote significant resources to the implementation of traffic safety countermeasures that prevent pedestrian fatalities. However, the before-after comparisons typically used to evaluate the success of these countermeasures often suffer from selection bias. This paper motivates the tendency for selection bias to overestimate the benefits of traffic safety policy, using New York City's Vision Zero strategy as an example. The NASS General Estimates System, Fatality Analysis Reporting System and other databases are combined into a Bayesian hierarchical model to calculate a more realistic before-after comparison. The results confirm the before-after analysis of New York City's Vision Zero policy did in fact overestimate the effect of the policy, and a more realistic estimate is roughly two-thirds the size.