Cause, effect and regression in road safety: a case study.

Researchers use various ways to determine what change in safety is caused by some treatment. One way is to fit regression equations to cross-section data. Can this work? Another way is to do a before-after study. Is this better? I examine these questions in the setting of a case study. The treated units are rail-highway grade crossings, the treatment is the replacement of 'crossbucks' by 'flashers', and as evidence serve published papers and reports. The results of regression studies are all over the place. Still, one cannot be sure whether this is a sign that the regression failed to capture cause and effect or a sign that the effect of this treatment depends strongly on the conditions in which it is applied. As different regressions use different variables, they cannot corroborate or negate each other's results. This is deeply troubling. The results of before-after studies, in this case, are very consistent. Unfortunately the results do not apply to specific conditions and are therefore of limited practical use. In this respect crash modification functions derived from regressions would have an inherent advantage over those from before-after studies provided they captured cause and effect. There is, at present, little ground for the belief that they do.

[1]  Gary A Davis,et al.  Possible aggregation biases in road safety research and a mechanism approach to accident modeling. , 2004, Accident; analysis and prevention.

[2]  Education Welfare.,et al.  Smoking and Health. Report of the Advisory Committee to the Surgeon General of the Public Health Service. , 1964 .

[3]  E Hauer.  Should stop yield? Matters of method in safety research , 1991 .

[4]  D. Rubin.  Estimating causal effects of treatments in randomized and nonrandomized studies. , 1974 .

[5]  Frank F Saccomanno,et al.  A Model for Evaluating Countermeasures at Highway-Railway Grade Crossings , 2005 .

[6]  Fedel Frank Saccomanno,et al.  Collision Frequency Analysis Using Tree-Based Stratification , 2005 .

[7]  J Hitz,et al.  RAIL-HIGHWAY CROSSING RESOURCE ALLOCATION PROCEDURE, USER'S GUIDE , 1982 .

[8]  M. Bunge.  Causality and modern science , 1979 .

[9]  W. G. Cochran.  The Planning of Observational Studies of Human Populations , 1965 .

[10]  G A Davis,et al.  Accident reduction factors and causal inference in traffic safety studies: a review. , 2000, Accident; analysis and prevention.

[11]  P. Holland.  Statistics and Causal Inference , 1985 .

[12]  Rune Elvik.  Operational Criteria of Causality for Observational Road Safety Evaluation Studies , 2006 .

[13]  Ronald W Eck,et al.  FURTHER INVESTIGATION OF THE EFFECTIVENESS OF WARNING DEVICES AT RAIL-HIGHWAY GRADE CROSSINGS , 1985 .

[14]  P Mengert.  RAIL-HIGHWAY CROSSING HAZARD PREDICTION RESEARCH RESULTS , 1980 .

[15]  Janet Coleman,et al.  INVESTIGATION OF ACCIDENT DATA FOR RAILROAD-HIGHWAY GRADE CROSSINGS , 1976 .

[16]  J. Bennett,et al.  Enquiry Concerning Human Understanding , 2010 .

[17]  Francisco J. Ayala,et al.  Science, evolution, and creationism , 2008, Proceedings of the National Academy of Sciences.

[18]  Fedel Frank Saccomanno,et al.  Evaluating factors affecting safety at highway-railway grade crossings , 2005 .

[19]  E Hauer,et al.  HOW TO ESTIMATE THE SAFETY OF RAIL-HIGHWAY GRADE CROSSINGS AND THE SAFETY EFFECTS OF WARNING DEVICES , 1987 .

[20]  J Morrissey.  THE EFFECTIVENESS OF FLASHING LIGHTS AND FLASHING LIGHTS WITH GATES IN REDUCING ACCIDENT FREQUENCY AT PUBLIC RAIL-HIGHWAY CROSSINGS, 1975-1978 , 1980 .