Objective: The primary goals of this analysis are to expand on and clarify the findings of the 2007 NHTSA analysis by using a greater variety of vehicles and several additional years of crash data. This analysis will also evaluate electronic stability control (ESC) effectiveness in all police-reported crashes.
The principal evaluation questions are:
Methods: Percent effectiveness of ESC was estimated by comparing the types of crashes that vehicle models experienced immediately before and immediately after the introduction of ESC. Because optional ESC generally cannot be identified from the VIN, only models that transitioned from no ESC system to a standard ESC system were included in this analysis. Effectiveness estimates were computed for different crash types relative to a control group of low-speed and similar crashes that are unlikely to be affected by ESC. The estimates should be interpreted as the reduction in the likelihood of a vehicle being involved in a specific type of crash as a result of ESC being added to that vehicle.
Results: When a vehicle is equipped with ESC, it has a smaller likelihood of being involved in a crash than a similar vehicle without ESC. Overall, ESC was associated with a six percent decrease in the likelihood that a vehicle would be involved in any police reported crash and an 18 percent reduction in the probability that a vehicle would be involved in a fatal crash. For passenger cars, the reductions are 5 percent and 23 percent, respectively; for LTVs, 7 percent and 20 percent. Each of these reductions is statistically significant except for the 5 percent overall effect in cars. More specific crash types were also analyzed and these results are presented in the body of the paper.
Discussion: Estimates of effectiveness were especially large for crash types involving loss of vehicle control. Passenger cars and LTV’s do not show large differences in effectiveness and show more similar results here than in previous analyses. The effect of ESC on collisions with pedestrians, bicyclists and animals, if any, is still unclear and will be monitored as more data becomes available.