LONGITUDINAL STUDY OF CALIFORNIA DRIVER ACCIDENT FREQUENCIES I: AN EXPLORATORY MULTIVARIATE ANALYSIS

The longitudinal study attempts to develop an optimum accident prediction system using nine year driving history records of a large sample of California drivers and biographical and mileage exposure data collected through a questionnaire survey. An exploratory non-concurrent regression analysis on three year accident interval produced a prediction equation with 29 predictor variables and a multiple correlation coefficient of .271. Among the variables which discriminated between accidents and non-accident drivers were prior conviction frequencies and prior accidents. Drivers with prior accidents and convictions were more likely to have accidents in subsequent time periods.

• Supplemental Notes:
  • Sponsored in part by Federal Highway Administration, Washington, D.C.
• Corporate Authors:

  California State Department of Motor Vehicles

  Development and Consultation Section
  Sacramento, CA  United States  95818

  Federal Highway Administration

  Research Development and Consultation Section, 1200 New Jersey Avenue, SE
  Washington, DC  United States  20590
• Authors:
  • Kwong, K W
  • KUAN, J
  • Peck, R
• Publication Date: 1976-6

Media Info

• Pagination: 57 p.

Subject/Index Terms

• TRT Terms: Crash rates; Data collection; Driver records; Drivers; Forecasting; Hazards; Licenses; Mathematical prediction; Multivariate analysis; Prevention; Questionnaires; Regression analysis; Risk assessment; Safety; Traffic crashes
• Uncontrolled Terms: Motor vehicle accidents
• Geographic Terms: California
• Old TRIS Terms: Reggression analysis
• Subject Areas: Data and Information Technology; Highways; Safety and Human Factors; Security and Emergencies;

Filing Info

• Accession Number: 00166273
• Record Type: Publication
• Source Agency: National Technical Information Service
• Report/Paper Numbers: CAL-DMV-RSS-76-55 Final Rpt.
• Files: TRIS, USDOT, STATEDOT
• Created Date: Sep 20 1978 12:00AM