Document Type


Publication Date



Bus drivers -- Training of -- Oregon, Bus driving -- Safety measures, Bus drivers -- In-service training -- Evaluation


This report provides an evaluation of the Oregon Department of Transportation-Driver and Motor Vehicle (DMV) Services Driver Improvement Program (DIP), which was substantially changed in 2002. Prior to 2002, the DIP was organized around four progressive steps involving advisory letters, warning letters, probation, and suspension. The current program is organized around two steps: restriction and suspension. The timeline to the steps in the current program have also been shortened. To evaluate the current program, driver records of persons suspended between January and July of 2004 were examined in relation to a sample of Oregon’s driving population. The incidence of crashes and traffic offense convictions of DIP subjects in the 18-month period prior to suspension was compared to the incidence of these events among the driving population. A similar comparison was also made for the 18-month period following suspension. A substantial reduction in the relative incidence of crashes and convictions among DIP subjects following suspension was observed. This finding is subject to the effects of regression-to-the-mean. An approximation of regression-to-the-mean effects was made based on prior evaluations of Oregon’s DIP that employed a true experimental design. A regression analysis was also undertaken using driver record information from the period prior to suspension to estimate the likelihood of post-suspension crash and traffic offense conviction involvement. The estimated likelihood of post-suspension crash involvement was significantly affected by the frequency of pre-suspension crashes, but not by the frequency of pre-suspension convictions. Conversely, the estimated likelihood of post-suspension convictions was significantly affected by the frequency of pre-suspension convictions, but not by the frequency of pre-suspension crashes. Two changes in the DIP are suggested in the concluding section of the report. The first change involves re-instituting warning letters, given their demonstrated cost effectiveness in the driver improvement literature. The second change involves the assignment of greater weight to crashes in triggering license actions, based on the regression findings.


Report no. FHWA-OR-RD-07-08; Grant no. SPR 634.

Persistent Identifier