استفاده از رگرسیون لجستیک به منظور برآورد تاثیر عوامل تصادف در شدت تصادف

Logistic regression was applied to accident-related data collected from traffic police records in order to examine the contribution of several variables to accident severity. A total of 560 subjects involved in serious accidents were sampled. Accident severity (the dependent variable) in this study is a dichotomous variable with two categories, fatal and non-fatal. Therefore, each of the subjects sampled was classified as being in either a fatal or non-fatal accident. Because of the binary nature of this dependent variable, a logistic regression approach was found suitable. Of nine independent variables obtained from police accident reports, two were found most significantly associated with accident severity, namely, location and cause of accident. A statistical interpretation is given of the model-developed estimates in terms of the odds ratio concept. The findings show that logistic regression as used in this research is a promising tool in providing meaningful interpretations that can be used for future safety improvements in Riyadh.

Accident severity is of special concern to researchers in traffic safety since this research is aimed not only at prevention of accidents but also at reduction of their severity. One way to accomplish the latter is to identify the most probable factors that affect accident severity. This study aims at examining not all factors, but some believed to have a higher potential for serious injury or death, such as accident location, type, and time; collision type; and age and nationality of the driver at fault, his license status, and vehicle type. Other factors were not examined because of substantial limitations in the data obtained from accident reports. Logistic regression was used in this study to estimate the effect of the statistically significant factors on accident severity. Logistic regression and other related categorical-data regression methods have often been used to assess risk factors for various diseases. However, logistic regression has been used as well in transportation studies. A brief literature review follows of the use of this type of regression in traffic safety research.

Since the response variable is of a binary nature (i.e. has two categories — fatal or non-fatal), the logistic regression technique was used to develop the model in this study. The intent was to provide a demonstration of a model that can be used to assess the most important factors contributing to the severity of traffic accidents in Riyadh. On the basis of traffic police accident data, nine explanatory variables were used in the model development process. Using the concept of deviance together with the Wald statistic, the study variables were subjected to statistical testing. Only two variables were included in the model, namely, accident location and accident cause. The observed level of significance for regression coefficients for the two variables was less than 5%, suggesting that these two variables were indeed good explanatory variables. The results presented in this paper show that the model provided a reasonable statistical fit. Stratifying location-related data into two classes, the model revealed that the odds of a non-intersection accident being fatal are higher. This finding might lead to a greater focus on road accident sites other than intersections, which should help agencies focus their safety improvements more cost-effectively. However, it should be said that not only the relative danger as expressed by the odds ratio, but also the absolute density of accidents with regard to location should be taken into account in order to develop cost-effective strategies. The odds presented in this paper can be used to help establish priorities for programs to reduce serious accidents. For example, since the odds of being involved in a fatal accident at a non-intersection location because of a wrong-way violation are relatively higher than those for any other violation, drivers should be warned in a specific awareness program about the possible lethality of such a violation. The same can be said of the impact of running a red light on the odds of being involved in a fatal accident. Presentation of odds in a matrix format, as described in this study, provides a simple method for interpretation. The columns and rows of the matrix correlate the factors in the logistic model, and each cell shows the impact of a certain factor on the odds with respect to another factor (a corresponding factor). It is important to note that the odds described in this paper were computed with no consideration for traffic exposure or the data that are not available or difficult to obtain in Riyadh. However, the findings of this study can be considered as guidance for a future study when such data become available.