Power computations in time series analyses for traffic safety interventions.

Area of Science:

• Road safety research
• Transportation policy analysis
• Statistical methodology

Background:

• Evaluating traffic safety interventions requires administrative time series data, like monthly motor vehicle collision data.
• Collecting this data can be difficult and expensive.
• Sufficient statistical power is essential for policy decisions based on intervention analysis.

Purpose of the Study:

• To present a simple methodology for sample size determination in traffic safety intervention analysis.
• To ensure adequate data collection for detecting meaningful changes in road safety.
• To provide a useful tool for researchers and policymakers.

Main Methods:

• A straightforward methodology for sample size determination is proposed.
• The method focuses on ensuring sufficient time series data collection before and after an intervention.
• The approach is illustrated using a traffic safety study funded by the National Institutes of Health (NIH).

Main Results:

• The presented methodology offers a practical approach to sample size determination.
• It addresses the challenge of data collection costs and difficulties.
• The method is designed to enhance the statistical power of intervention analyses.

Conclusions:

• The proposed methodology is expected to be valuable for various traffic safety intervention applications.
• It aids in optimizing data collection strategies for road safety studies.
• This approach supports evidence-based policymaking in traffic safety.