Evaluation of Driver Education Teaching Mode in North Carolina: Final Report

Executive Summary

In the spring of 2012 the School of Government was approached by the North Carolina Department of Public Instruction (DPI) about conducting an evaluation of the different teachings modes used for driver education in North Carolina. The School was also asked to include in the comparison, to the extent possible, a fully interactive on-line option. Finally, we were asked to assess the use and perceived impact of the optional student fee for driver education. We were unable to find other, similar evaluations specifically focused on driver education. In addition, a brief examination of the literature shows that while on-line programs had been described recently in a report by the National Highway transportation Safety Association (NHTSA) of the U.S. Department of Transportation, no evaluations of on-line driver education programs exist. The focus of the evaluation covered three broadly defined teaching modes: traditional classroom instruction; traditional classroom augmented with various technological tools, which we refer to as blended instruction; or completely web-based on-line instruction. For each mode, we examined two evaluation criteria: (1) quality as defined by overall pass rates, test scores, and test frequencies for the Knowledge test offered for licensure by the Division of Motor Vehicles (DMV), and (2) cost per student. We received an extraordinarily large and complete dataset of 2011-2012 DMV Knowledge test information for all students in North Carolina, allowing for a robust analysis for our first part of the evaluation. We conducted a state-wide survey of schools to obtain information on how driver education was taught and the use of the optional student fee. We also had access to limited information on driver education costs as self-reported by Local Education Agencies (LEAs) for other DPI purposes. We found little difference in the performance of traditional classroom instruction versus blended instruction as measured by overall pass rates, test scores, or test frequencies for students taking the DMV Knowledge test. North Carolina has conducted a pilot of a web-based, on-line education model, but the earliest graduates of that program will take the Knowledge test in 2013, and therefore we did not have test results from them to include in this analysis. Finally, we were unable to assess cost per student across the three teaching modes because of concerns over the comparability of the available cost information. A budget audit or other method to determine school-level driver education costs may be required to accurately capture the cost-per-pupil averages between the blended and traditional methods.