HOME-STUDY DRIVER EDUCATION
DMV-Proctored Exit Examination
After students in the study completed their driver education course, they were required to return to the provider’s school site within 2 weeks to take a DMV-proctored exit examination before receiving a Certificate of Completion of Classroom Driver Education (OL 237 or DL 387). The exams were proctored every 2 weeks at most school sites and every week at the others. The proctors were selected from DMV employees at field offices near the schools and, in most cases, were trained in-person. The exit exams were not graded by the proctors and there was no “failing” score for the exam. However, every effort was made to ensure that the providers and proctors did not make this fact known to the students.
The provider schools were not allowed to see the content of the exit examinations to prevent them from coaching their students on the exam content. This 60-item examination was created by the department for use in the study and contained three different sets of items: (a) 40 items to measure knowledge of rules of the road and safe driving practices as presented in the standardized curriculum, (b) 15 items to measure driver attitudes along an internal-external locus of control continuum, and (c) 5 items to evaluate the students’ opinions about the courses. The exit test represented the best and most reliable criterion for evaluating the relative effectiveness of the different courses
because of its close proximity to administration. A copy of the exit
course completion and examination is shown in
high level of Appendix D.
Knowledge Items A pool of 120 knowledge items was written for the exit examination based on the material presented in the standardized driver education curriculum. The items were written so that the content covered by the exit examination would proportionately represent the content covered in the standardized driver education curriculum. That is, longer chapters had more knowledge questions on the exit examination than did shorter chapters. In addition, the items covered the general safety-related material in each chapter rather than esoteric details that would be idiosyncratic to the standardized curriculum such as specific word choice.
Three forms or versions of the exit examination with 40 knowledge items each were pilot tested at the beginning of the first day of classroom driver education at a local (Sacramento) licensed driving school in January 2001. The purpose of the pilot was to identify items that 40% or more of first-day driver education novices answered incorrectly to determine whether the exit examinations could discriminate between those who were knowledgeable of the subject matter and those who were not. Some items were modified or replaced based on the results of the pilot study to make them more difficult for novices. Form 1 of the revised exit examination was pilot tested again in February 2001 at the beginning of the first day of driver education at a different local licensed driving school. Based on the results of this second pilot study, a final Form 1 of the exit examination was created consisting of 40 knowledge items and the other items discussed in the next two sections.
The internal-consistency reliability of the exam was computed using the Kuder- Richardson formula (K-R 20). This type of reliability indicates the degree of uniformity among exam items and the extent to which the exam items measure a common domain