Evaluation of New Jersey crash data reported to MCMIS crash file

Abstract

This report is part of a series evaluating the data submitted to the Motor Carrier Management Information System (MCMIS) Crash File undertaken by the Center for National Truck and Bus Statistics at the University of Michigan Transportation Research Institute. Earlier studies showed that reporting to the MCMIS Crash File was significantly incomplete. This report examines reporting from the state of New Jersey. New Jersey Police Accident Report (PAR) files were matched to the MCMIS Crash file to determine the nature and extent of underreporting. In 2003, there were 10,103 crash involvements in New Jersey that were reportable to the MCMIS Crash file. Of these crash involvements, 8,333 were actually reported, resulting in a reporting rate of 82.5%. In addition, 5.3% of MCMIS cases that were reported do not qualify for reporting. Reporting rates vary by crash severity, crash month, and jurisdiction. Unlike many other states in which reporting rates were greater for more severe involvements, in New Jersey reporting rates were lower for more severe involvements. The reporting rates were 67.4% for fatal involvements, 81.5% for injury involvements, and 83.2% for towaway involvements. Underreporting is also related to the time of the year, indicating delays in providing the data. The month of December had the highest percent of unreported cases (21.7%). Data quality is also reviewed. The PAR case number contains inconsistent characters such as dashes, question marks, and asterisks. This appears to be a source of duplicate records. The ambulance run number and the hazardous materials placard variables also contain many inconsistent characters.