Related Experiment Videos
A new method for making interstate comparisons of highway fatality rates
1Injury Control Center, Harvard School of Public Health, Boston, MA 02115.
Accident; Analysis and Prevention
|February 1, 1993
Summary
This study introduces a new method to identify highway safety performance, accounting for external factors. It reveals many states perform differently than expected, highlighting the need for nuanced policy evaluation.
Area of Science:
- Transportation safety
- Public health policy
- Statistical modeling
Background:
- Highway fatality rates exhibit significant jurisdictional variations.
- Attributing these differences solely to policy is challenging due to exogenous factors like weather.
- A robust method is needed to isolate policy impact from uncontrollable influences.
Purpose of the Study:
- To develop and apply a novel statistical method for identifying states with unexpectedly high or low highway fatality rates.
- To differentiate actual state performance from outcomes influenced by external factors.
- To inform evidence-based transportation safety policy.
Main Methods:
- A fixed effects linear model was employed to estimate state-specific fatality rates.
- Exogenous factors were statistically controlled for in the analysis.
- The method was applied to highway fatality data from the 48 contiguous US states (1975-1986).
Main Results:
- Many states demonstrated significantly higher or lower highway fatality rates than predicted by crude measures.
- The analysis identified specific states performing better or worse than expected after accounting for external variables.
- Results challenge simple rankings based on raw fatality data.
Conclusions:
- Crude highway fatality rates can be misleading for policy evaluation.
- The developed method provides a more accurate assessment of state-level highway safety performance.
- Findings have implications for targeted policy interventions and resource allocation in transportation safety.