This unique book explains how to fashion useful regression models from commonly available data to erect models essential for evidence-based road safety management and research. Composed from techniques and best practices presented over many years of lectures and workshops, The Art of Regression Modeling in Road Safety illustrates that fruitful modeling cannot be done without substantive knowledge about the modeled phenomenon. Class-tested in courses and workshops across North America, the book is ideal for professionals, researchers, university professors, and graduate students with an interest in, or responsibilities related to, road safety. This book also: · Presents for the first time a powerful analytical tool for road safety researchers and practitioners · Includes problems and solutions in each chapter as well as data and spreadsheets for running models and PowerPoint presentation slides · Features pedagogy well-suited for graduate courses and workshops including problems, solutions, and PowerPoint presentations · Equips readers to perform all analyses on a spreadsheet without requiring mastery of complex and costly software · Emphasizes understanding without esoteric mathematics · Makes assumptions visible and explains their role and consequences

This report summarizes the work done by a team of experts on the subject of efficient management of safety research. It shows that decision analysis methods can also be applied to reveal the "value" of a research project, derived from the project's potential to correct a wrong decision on a counter-measure's implementation, together with the estimated money cost of that wrong decision. It also shows how the techniques can be applied to determine the appropriate size of a field experiment with a countermeasure, and how they might indicate priorities among a list of potential measures. It concludes with suggestions of further research directions.

This report presents the results of research that performed a well-designed before-after evaluation of the safety effects of providing left- and right-turn lanes for at-grade intersections. Geometric design, traffic control, traffic volume, and traffic accident data were gathered for a total of 280 improved intersections, as well as 300 similar intersections that were not improved during the study period. The types of improvement projects evaluated included installation of added left-turn lanes, added right-turn lanes, and extension of the length of existing left- or right-turn lanes. An observational before-after evaluation of these projects was performed using several alternative evaluation approaches. The three contrasting approaches to before-after evaluation used were the yoked comparison or matched-pair approach, the comparison group approach, and the Empirical Bayes approach. The research not only evaluated the safety effectiveness of left- and right-turn lane improvements, but also compared the performance of these three alternative approaches in making such evaluations. The research developed quantitative safety effectiveness measures for installation design improvements involving added left-turn lanes and added right-turn lanes. The research concluded that the Empirical Bayes method provided the most accurate and reliable results. Further use of this method is recommended.