top of page
  • Writer's pictureMATPP

Analytic Methods in Accident Research: The Current Methodological Frontier and Future Directions

16 OCT 2012 (THU) | 19:00 - 20:00

Wang Gungwu Theatre, Graduate House, The University of Hong Kong.


Speakers:Prof Fred Mannering Charles Pankow Professor of Civil Engineering, Purdue University, USA

Fred Mannering is currently the Charles Pankow Professor of Civil Engineering at Purdue University with a courtesy appointment in the Department of Economics. He received his BSCE from the University of Saskatchewan, MSCE from Purdue University and PhD from the Massachusetts Institute of Technology. Dr. Mannering's expertise is in the application of statistical and econometric methods to study a variety of subject areas including highway safety, transportation economics, automobile demand, and travel behavior. His body of work has been highly influential and has been cited over twenty-five hundred times in the Institute for Scientific Information (ISI) databases, over twenty-five hundred times in Scopus, and over five thousand times in Google Scholar. Dr. Mannering has published over 100 refereed journal articles, 2 text books, over 60 other publications (conference proceedings, project reports, book reviews and commentaries), and has given over 120 invited lectures and presentations at professional conferences.

His undergraduate textbook, "Principles of Highway Engineering and Traffic Analysis" is now in its fifth edition and has sold over 40,000 copies. He has been principal investigator on 38 funded research projects and has supervised 21 PhD students and 43 MS students. Dr. Mannering has been Editor-in-Chief of Transportation Research Part B: Methodological since 2003. The Journal's 2.856 citation impact factor is currently first among transportation journals (ISI Web of Knowledge, 2011, Journal Citation Reports Social Science Edition) and fourth highest (second highest when excluding self cites) among 118 Civil Engineering journals (ISI Web of Knowledge, 2011, Journal Citation Reports Science Edition). Dr. Mannering's awards include: Arthur M. Wellington Prize, American Society of Civil Engineers, for the best paper in the Journal of Transportation Engineering (2010); James Laurie Prize, American Society of Civil Engineers (2009) "For his outstanding contribution to the advancement of transportation engineering through his influential research and publication in the area of highway safety"; Wilbur S. Smith Award, American Society of Civil Engineers (2005) "For outstanding contributions to the enhancement of the role of the civil engineer in highway engineering through excellence in teaching and research"; National Highway Safety Award (2001) for "A new method for prioritizing intersection improvements"; Harold Munson Award for outstanding teaching, Purdue University (2007); CHOICE Magazine's Outstanding Academic Books Award (1991) for "Principles of Highway Engineering and Traffic Analysis" first edition.



Vehicle crashes and resulting injuries continue to take a terrible economic and emotional toll on societies worldwide. Given these adverse impacts, the analysis of accident data has long been used as a basis for developing appropriate countermeasures that seek to reduce the frequency and severity of transportation accidents. However, accident data present many complex methodological challenges that are often overlooked by transportation agencies and researchers. These methodological challenges include factors relating to unobserved heterogeneity (resulting from the fact traditional databases do not have information on all of the factors influencing accidents), endogeneity and self selectivity (a reflection that accident outcomes are often tied to interrelated processes and decisions), and temporal and spatial correlations (the possibility that accidents may be tied over time and space). In this talk, I will provide numerous examples of these methodological challenges (drawn from a variety of past studies) to show the subtleties of the issues involved, and how these subtleties can affect the inferences drawn from the data if not addressed in a methodologically appropriate way. I will also discuss how methodological challenges will persist with emerging data sources (such as those from naturalist driving and simulator studies), and give an assessment of potentially fruitful directions for methodological development in accident research.



Institute of Transport Studies, HKU (ITS)



Recent Posts

See All


Commenting has been turned off.
bottom of page