Evaluation of electronic collection of vehicle crash data in Iowa

Yerdelen, Turhan
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Journal Issue

As the primary source of highway safety data, vehicle crash data may not always be timely or meet the needs of analysts. Collecting and processing crash data is also time consuming and labor intensive for agencies and states. Emerging technologies are suggested to have potential to facilitate and shorten the data collection process, increase officer efficiency, and improve the quality of crash data collected. To that effect, the Iowa Department of Transportation, in cooperation with other agencies and states, developed the Mobile Accident Reporting System (MARS) in 1995. In 1997, the system was chosen by the Federal Highway Administration (FHWA) as the National Model for the Statewide Application of Data Collection & Management Technology to Improve Highway Safety, now known as Traffic and Criminal Software (TraCS). The objective of this research is to evaluate the efficiency of Iowa's electronic crash data collection system (TraCS) through field studies and database analyses based on the latest data and knowledge, and to document whether the system meets expectations such as better quality crash data (more accurate, complete, consistent, timely data), reduced data collection time in the field, and other suggested benefits. For this purpose, three studies were performed: Attribute Quality Assessment-whether electronic collection of crash data helps improve the quality of attributes (accuracy, completeness, consistency, legibility); Location Accuracy-whether the system improves accuracy of crash location; Report Completion Time-whether the system helps save officer time at scene. The attribute quality validated randomly selected paper reports with TraCS and compared the information on the same paper reports to those of the Iowa database. The location accuracy study was a before and after study on the crash database for unlocated and mislocated crashes for two location processes, node based versus GIS based. The report completion time study was performed in ten law enforcement agencies in Iowa. With the participation of 47 officers, report completion times for paper and electronic reporting processes were measured based on a hypothetical crash scenario created for this purpose.

Community and regional planning, Transportation