HOME-STUDY DRIVER EDUCATION
information such as whether the student had access to a computer that met the minimum hardware and software requirements for the CD course.
School Project Log
When the completed Parent/Guardian Release Statement was returned to a provider along with the course fee, the schools entered the student’s information on a School Project Log (shown in Appendix C), including the student’s true full name, birth date, whether or not the student had access to the computer hardware and software necessary to run the CD program, and whether or not the student would be enrolled in driver training during enrollment in the study. Students who simultaneously enrolled in both driver education and driver training are eligible to take the DMV written license test before completing driver education, and therefore were excluded from the analysis comparing the four instruction methods on DMV written test outcomes. The coding for simultaneous enrollment on the School Project Log was used to identify students who needed to be excluded for this reason. These students were still to be included in various analyses of the study exit examination and DMV drive test results, although separate analyses are presented in the Results section to determine if simultaneous enrollment interacted with instruction method.
Random Assignment of Students
After entering the student’s information on the School Project Log, the provider school determined which type of driver education instruction the student was to receive using a random assignment method created by the department. The provider schools were not allowed to divulge the random assignment scheme, nor to tell a student what type of instruction he or she was going to receive until after collecting the course fee and a signed Parent/Guardian Release Statement. The course assignment was based on the day of the month of the student’s birth and whether or not the student’s parent or guardian indicated (by a checkmark on the release statement) that he or she had access to a computer with the necessary hardware and software for the CD program. If the student had access to a computer that would allow the use of the CD program, he or she was eligible to be randomly assigned to any of the four instruction methods using Random Assignment Method I. Otherwise, he or she was eligible to be randomly assigned to one of the three instruction methods other than the CD program using Random Assignment Method II. After determining the type of instruction the student was to receive, the school indicated the selection by checking the appropriate box and writing the date that the assignment was made on the School Project Log. The home- study kits were then distributed to the students, or they were enrolled in the next available classroom course (if they were assigned to classroom instruction).
The three home-study kits were completed on different dates and therefore were distributed to the provider schools at different times during the study. As a result, there were actually three different pairs of random assignment schemes used during the course of data collection. The workbook was the first completed kit, and was distributed in December 2001. The CD program was distributed in February 2002 and the PEN course was distributed in July 2002. Hence, the random assignment schemes changed to include the new methods of instruction as they became available. The