Long-term trends in traffic safety are usually modeled by regressing the fatal accident rate on time. Safety interventions can then be assessed by evaluating their impact relative to these baseline trends. However, this approach confounds the accident rate with the fatality rate and introduces a number of statistical problems. Methods for avoiding these problems are illustrated by modeling California traffic safety data for the years 1939 to 1973. These models reveal a substantial reduction in the proportion of accidents involving fatalities. No such reduction is evident in accident frequency, after changes in vehicle mix and the driver age distribution are taken into account.
Get full access to this article
View all access options for this article.
References
1.
Box, G. E. P., and Jenkins, G. M. (1976). Time series analysis: Forecasting and control. San Francisco: Holden-Day.
2.
Chu, B. Y., and Nunn, G. E. (1976). An analysis of the decline in California traffic fatalities during the energy crisis. Accident Analysis and Prevention, 8, 145-150.
3.
Durbin, J., and Watson, G. S. (1951). Testing for serial correlation in least squares regression, II. Biometrika, 38, 159-178.
4.
Freeman, M. F., and Tukey, J. W. (1950). Transformations related to the angular and the square root. Annals of Mathematical Statistics, 21, 607-611.
5.
Gibbons, J. D. (1971). Nonparametric statistical inference. New York: McGraw-Hill.
6.
Heckard, R. F., Pachuta, J. A., and Haight, F. A. (1976). Safety aspects of the national 55 mph speed limit (Report No. FHWA-RD-76-191). Washington, DC: U.S. Dept. of Transportation.
7.
Helwig, J. T., and Council, K. A. (1979). SAS user's guide, 1979 edition. Cary, NC: SAS Institute.
8.
Kemper, W. J., and Byington, S. R. (1977). Safety aspects of the national 55 mph speed limit. Public Roads, 41, 58-67.
9.
McCleary, R., and Hay, R. A., Jr. (1980). Applied time series analysis for the social sciences. Beverly Hills, CA: Sage.
