The AAA Foundation recently began a research project, Risky Driving Behaviors and Safety Outcomes, in conjunction with Virginia Tech Transportation Institute.

In the recently concluded $3 million "100-Car Naturalistic Driving Study" conducted for the National Highway Traffic Safety Administration and the Federal Highway Administration, the Virginia Tech Transportation Institute performed real-time observations of drivers to learn about rear-end crashes. The final sample consisted of 109 predominantly high-mileage, relatively young, mostly male drivers in the Washington, DC/Northern Virginia area, and they were each observed for one year. Each vehicle was instrumented with five video cameras and an electronic data sensor (i.e., “black box”) that was used to detect various driving behaviors (e.g., speed, safety belt use), thus providing both video and quantitative data for analysis.

For this study, the researchers will perform additional analysis of the database compiled in the earlier study, to conduct a comprehensive analysis of various risky driving behaviors and their relationship to crashes, near crashes, and crash-relevant conflicts. This observation method has several strengths. It will analyze directly observed driver behavior, rather than relying on driver self-reporting or police reporting. Along with enhancing credibility, direct observation also allows near crashes and conflicts to be considered as outcomes, rather than relying solely on infrequent crash events.

Specifically, the project will:

• Assess the relationship between various risky driving behaviors (e.g., speeding, talking on a cell phone) and crashes, near crashes, and crash-relevant conflicts;

• Determine if the relationship between risky driving behaviors is similar across all levels of event severity (crashes, near-crashes, and crash-relevant conflicts);

• Use logistic regression to determine if the prevalence of risky driving behaviors during ‘normal, baseline driving' can predict the frequency of crashes, near crashes, and crash-relevant conflicts;

• Determine if the relationship between these risky driving behaviors and crashes, near crashes, and crash-relevant conflicts differ as a function of age and gender;

• Enhance understanding of the relationship between risky driving behaviors and crashes, near-crashes, and crash-relevant conflicts;

• Help direct future research and interventions aimed at reducing risky driving behaviors.