Static Stability as a Predictor of Overturn in Fatal Motor Vehicle Crashes

Fatal Accident Reporting System (FARS) files for the years 1981-1984 were examined for rollover crash involvement of 15 utility and passenger vehicle make-models for which static stability values (1/2 track width divided by height of center of gravity) were published. The values ranged from highs of 1.57-1.62 for the pre-1979 Ford LTD, the pre-1979 Chevrolet Nova, and the pre-1982 Pontiac Firebird to lows of 1.01-1.07 for the Jeep CJ-5 and CJ-7 and the pre-1978 Ford Bronco. Rollover as the first harmful event and as the most harmful event per 100,000 vehicles registered was strongly predicted by stability. Stability was unrelated to nonrollover crashes. The low-stability vehicles were much more likely to roll over on the road rather than after leaving the road. Other road, driver, and environmental risk factors recorded in the FARS files were not correlated to stability in such a way as to explain the high rollover of low-stability vehicles. Using Federal Highway Administration vehicle mileage estimates, calculations were made of the mileages under various conditions which the vehicles with low stability values would have to have been driven if mileage or hazardous-condition differences rather than stability differences accounted for their substantially higher fatal rollover fatal crash rates. This analysis indicates that fatal rollover rates of low-stability vehicles could not have occurred at reasonable mileage.