Trends in driver education, evaluation and development

By: Lawrence P. Lonero, Northport Associates

Date: Thursday, 22. November 2007

Abstract

The aims of the article are to provide a richer understanding of how systematic program evaluation and development can best be understood and improved in the context of driver education policy, program planning, and program management. The article provides an overview of the conclusions, limitations and implications of the evaluation literature for driver education program practice and development, as well as a discussion of potential effectiveness of trends in driver education.

Introduction

For the purposes of this article, driver education means beginner, pre-licensing driver instruction. Most types of education are considered successful if students learn what they are supposed to and pass tests. Driver education is expected to actually change subsequent behavior, and change it sufficiently to produce measurable effects on crashes. Most researchers are skeptical about its success so far, but driver education has strong ï¿½face validityï¿½ among parents, who think it makes their children safer drivers (Fuller and Bonney 2003, 2004; Plato and Rasp 1983). Driver education has become a substantial industry, although a highly fragmented one, particularly in the US. There is commercial and other interest in encouraging student participation. Driver education is widely supported by insurance premium discounts and driver licensing provisions.

The novice driver problems that driver education hopes to mitigate are stioll somewhat controversial ï¿½ advocate can argue endlessly over the relative importance of skills versus motives. Inadvertent errors resulting from skill deficiencies and unsafe choices resulting from incautious motives probably both contribute to young novice driversï¿½ excess risk, albeit perhaps not in the same proportions for all crashes and at all times (Mayhew, Simpson, and Pak 2003; McKnight & McKnight, 2003; Wilde, 1994). To achieve major safety improvement for new drivers, it has become clear that more comprehensive approaches to research and program development are needed (Lonero, Clinton, Brock, Wilde, Laurie, & Black, 1995; National Research Council, Institute of Medicine, and Transportation Research Board, 2007).

Program Trends

Driver education in North America went into a decline after the late 1970s, when the federally-funded driver education experimental study in DeKalb County, Georgia failed to produce unequivocal positive safety results. More than a decade ago, the need for renewal of driver education was recognized. The National Highway Transportation Safety Administration (NHTSA) initiated renewed development with a national workshop and a research agenda for driver education and graduated/provisional licensing (GPL) (Smith, 1994) . In 1995, the AAA Foundation for Traffic Safety (AAAFTS) supported a project to develop programs to ï¿½reinventï¿½ driver education. The principal product was a draft curriculum outline, intended to lead to a more intensive and comprehensive form of driver education for the 21st Century (Lonero et al. 1995) . A decade later the AAA Foundation initiated a project to ï¿½reinventï¿½ evaluation research in driver education, resulting in the recently published guidelines (Clinton and Lonero, 2006a). Evaluating Driver Education Programs: Comprehensive Guidelines and two companion documents are intended to encourage better and more constructive evaluations of driver education programs (Clinton and Lonero, 2006b; Lonero and Clinton, 2006).

As the result of a fatal crash of a driver education car, the National Transportation Safety Board (NTSB 2005) held hearings and produced recommendations for a renewal of federal support for driver education development. It is possible that this will add impetus to the slow turn around in driver education that appears to be taking place after a long period of relatively little centralized support or direction for R&D.

Driver education traditionally meant instruction only before the new driver was licensed to drive independently. A potentially important form of driver instruction is addressed to drivers after they are licensed to drive independently. In a few jurisdictions, such as Michigan, Finland, and Luxembourg, new drivers are required to take a second stage of training after they have been driving as licensed drivers for a short period of time (e.g., Glad 1988; Katila, Keskinen, Laapotti, and Hatakka 1995; Keskinen, Hatakka, and Katila 1998; Shope and Molnar 2003). Some safety benefits of these second-stage programs have been observed, although in a limited range of evaluations. Graduated driver education seems a logical extension of graduated licensing, but no other GDL jurisdiction has yet followed Michiganï¿½s lead.

While good instructional methods can facilitate better cognitive and psychomotor skills, these do not automatically lead to drastically reduced crashes. There was gain in knowledge and skills in the DeKalb study (Stock et al. 1983). Stronger improvements in safety probably require safer driving choices and habits; drivers must actually drive differently than they would otherwise. People do not always use what they know to minimize risk. Individual, social, and cultural factors are important but not yet well addressed.

Behavior change toward habitual lower risk behaviors is much harder to accomplish than is generally understood (Green and Kreuter 1999; Lonero and Clinton 1998; Lonero et al. 1994; Sleet and Lonero, 2002; Lonero, Clinton and Sleet 2006). More education is always a popular prescription for improving safety, but demonstrated effectiveness in improving safety performance solely through educational measures of any form is relatively rare.

An interesting development in recent years has been the entry of well-funded private corporations into driver education. For example, an automotive insurer (MetLife Casualty) supported the start up of Top Driver, which made inroads into consolidating the US private driving school industry, in partnership initially with General Motors and later with equity participation from Ford Motor Company. In the 1990s, Ford took a majority ownership in Young Drivers, a large and long-established international driver education provider based in Canada. The auto companies have since backed away from beginner driver education. AAA developed a new curriculum approach called Licensed to Learn, a comprehensive risk-based approach that incorporated new technology and interactivity, and expanded its direct delivery of courses. AAA has embarked on a new initiative that could lead to a national network of AAA-approved driving schools. There is renewed development of program standards among driver education organizations ( Highway Safety Center Indiana University of Pennsylvania 2002; Road Safety Educatorsï¿½ Association and Driving School Association of the Americas 2005).

There are also trends in instructional methods and program delivery. Traditionally, all driver education activities involved face-to-face interaction between instructor and learner, although classroom instruction was often supported with film and video media and sometimes with simulators. More recently, self-instruction, computer-based instruction, simulation, and web-based instruction are becoming prevalent. These have produced profound changes in the technological, business arrangement, regulatory, and driver licensing environments in which driver education operates. It is not yet clear whether these changes will improve the safety effectiveness of driver education. Most changes are directed to delivery efficiency, and they are largely technology driven and entrepreneurial, rather than systematic and evidence-based. However, studies are showing positive learning effects of computer-based instruction and simulation (Fisher, Laurie, Glaser, Connerney, Pollatsek, Duffy, and Brock 2002). A California study found some benefits of CD-ROM and internet/workbook presentation of basic driver education material (Masten and Chapman, 2003; 2004).

Driver education is now highly diverse. Some public programs involve many thousands of students each year and receive strong support from the state or province. Other jurisdictionsï¿½ programs are small and receive little support. Most commercial driving schools are relatively small, many having only a single location. A few driving schools have many locations and teach thousands of students each year. Web-based programs also teach many thousands of students. Driving school operating standards can range from none to strict centralized control. One large driving school (Young Drivers) maintains ISO quality standard certification. Instructor qualifications across different driver education programs range from minimal to highly qualified professional teachers.

From one perspective, the growing diversity of driver education programs can be seen as an indication of vigorous development that may lead to greater effectiveness. On the other hand, it may also result from many programs and jurisdictions remaining static or declining, while a few leading program providers or jurisdictions build new and better programs. The nature and trends in program diversity need investigation, another call for more and better evaluation.

Programs have to be more carefully designed, implemented, managed, and, most especially, coordinated with other behavioural influences. In addition to the modest existing level of epidemiological knowledge about young beginnersï¿½ risk, it is important to know more about critical personal, social, and cultural characteristics and specific aspects of studentsï¿½ attitudes, skills, and driving performance during and after training.

Past Driver Education Evaluation

Compared to other health and education fields, driver education has seen limited evaluation research. The largest experimental and most influential driver education evaluation (ï¿½DeKalbï¿½) involved randomly assigning 16,000 U.S. high school student volunteers to three groups ï¿½ intensive training, minimal training, or no formal driver education. The results failed to show a dramatic, long-term benefit of a special course (Davis 1990; Lund, Williams, and Zador 1986; Ray, Sadof, Weaver, Brink, and Stock. 1980; Smith 1983; Smith and Blatt. 1987; Stock, Weaver, Ray, Brink, and Sadof, 1983; Weaver 1987). Reactions to the results had profound effects on driver education; many high school driver education programs were dropped (Simpson, 1996).

Clinton and Lonero (2006a) identified three key areas where driver education evaluation has been found lacking:

Program theory: the logic model that justifies thinking a program work. There has been little evaluation of the theory underlying various driver education programs.

Formative evaluation: applying evaluation tools to improve the content and delivery of a program. Driver education courses vary greatly in quality, and there has been limited evaluation of program differences.

Methodological soundness: problems of scope, design, statistics, and sampling limit unequivocal conclusions about driver educationï¿½s present ultimate value and how its impact might be improved in the future.

Evaluation is a potentially valuable and crucial tool for improving driver education, but it seems clear that it has been too narrow to be fully constructive and to achieve its full potential.

Individual Evaluations of Driver Education Programs

The great majority of driver education programs have never been formally evaluated, including some of the apparently best developed and managed programs. Most existing evaluations are severely limited in scope, power, and scientific rigor. Clinton and Lonero (2006a) reviewed a number of individual evaluations of driver education programs, which represent either fairly recent work, or older studies of special importance. The approach was critical and primarily methodological, in aid of the evaluation guidelines. The strengths and limitations of three basic types of evaluation studies were identified:

Experimental studies

ï¿½students randomly assigned to training conditions;

Quasi-experimental studies

ï¿½naturally occurring groups compared; and

Ecological studies

ï¿½ assessment of changes in requirements; or program differences across different jurisdictions.

The now aging DeKalb study has been considered to be the most extensive and rigorous experimental evaluation, but even this study had serious limitations, and its conclusions are still controversial. Most other evaluation studies were more limited in scope and scale. Experimental evaluations typically have found no statistically significant effects of driver education on crash records, but one California study did. Quasi-experimental and ecological studies of widely varying quality and results have been conducted. Two large-scale ecological evaluations showed positive effects of driver education, but one early study did not. Table 1 provides a summary of findings of the individual studies reviewed in more detail in Clinton and Lonero (2006a).

Table 1. Summary of Selected Individual Evaluation Studies

Reference	Design	Results	Methodological Strengths/Limitations
*Experimental Studies*
Dreyer and Janke 1979 California	2,057 students randomly assigned to two training conditions	Those receiving range practice had fewer recorded crashes, but tests scores were no different	Randomized control trial Intermediate measures No follow-up survey for exposure & behavioral measures
Ray et al. 1980 Stock et al. 1983 DeKalb County, Georgia	Intensive, minimal, and no driver education groups About 6,000 students randomly assigned to each group	Intensive training (SPC) drivers had better skills and fewer crashes during first 6 months, but not beyond Effects were complex; see text	Comprehensive randomized controlled trial Long follow-up ï¿½ 6 years Formative evaluations and intermediate outcomes measures
Wynne-Jones and Hurst 1984 New Zealand	788 students, 561 received course, 227 family/friend taught Random assignment	No reduction in collisions for driver education group	Adequate design Small control group No formative evaluation or intermediate outcomes
Gregersen 1994 Sweden	850 students received driver education course compared to controls Random assignment	Driver education group significantly worse first year, significantly better second year	Longer follow-up ï¿½ 2 years Reasonable sample size
Masten and Chapman 2003; 2004 California	1,300 students randomly assigned to one of four instructional settings	Home-based methods better for 1 knowledge and attitude test, classroom better for DMV knowledge test	Sample size adequate Well planned and controlled Psychometric measures only
*Quasi-Experimental Follow-Up Studies*
Forsyth et al. 1995 United Kingdom	Survey of 15,000 new drivers	Longer time learning to drive associated with fewer crashes for males More driving education was associated with more crashes	Several follow-ups over time Self-selection bias Self-reported data only
Howarth et al. 2000 Australia	Self-report crash effects for in-car training effects	Substantial differences, but not significant	Sample size too small
McKenna et al. 2000 Pennsylvania	Survey and crash records Random sampling for survey	Driver education not associated with lower crashes or convictions	Multi-variate statistical analysis used to control for confounding variables SES missing from control variables
Lonero et al. 2005 Manitoba	Survey and crash records Random sampling for survey	Driver education not associated with lower crashes or convictions	Multi-variate statistical analysis used to control for confounding variables
Wiggins 2005 British Columbia	Cohort record study Case control study with survey and records	New graduated license holders who took driver education had 26% more crashes	Multi-variate statistical analysis used to control for confounding variables
Zhao et al. 2005 Ontario	Self-report survey of high school students	Driver education associated with fewer crashes for learners	Multi-variate statistical analysis used to control for confounding variables
*Ecological Studies*
Robertson and Zador 1978 27 U.S. States	Modeling study of driver education and fatal crash rates	No relation between proportion taking driver education and fatality rates	Not program specific
Robertson 1980 Connecticut	School boards with and without driver education	For school boards without driver education, total licensing and crashes of 16- and 17-year-olds decreased by 10-15%	Not enough data analysis presented
Potvin et al. 1988 Quï¿½bec	Mandatory driver education introduced in Quï¿½bec for all (formerly just 16-17 year olds)	Increased number of young driver crashes due to increased number of licensed females aged 16-17	Large sample size Different timeframes for treatment and control groups
Levy 1990 USA 47 States	Large-scale modeling study of effects of mandatory driver education	Small but significant beneficial effect on fatal crashes	Not program specific
Carstensen 2002 Denmark	Mandatory driver education, new curriculum	Reduced crashes	Large sample size No control of confounding variables

As the reviews and table show, there have been quite a number of program evaluations over recent years, but the findings are quite variable and the question of whether driver education programs enhance or detract from safety remains highly controversial. Findings to support either side can be found, although the larger number of studies fall on the negative side. No one study design is perfect, and progress will likely develop on a ï¿½weight-of-evidenceï¿½ basis over numerous studies of different types.

There are also many other questions that need to be considered. Do some types of driver education programs lead to better educational and safety outcomes than others? Can we identify which components of driver education programs do and do not work? Do programs meet their learning objectives? How can driver education programs be improved in order to yield safer young drivers? These questions have been left partially or completely unanswered. Methodological weaknesses have plagued past evaluations, and a lack of systematic research progress has held back the development of more effective programs.

Reviews of Driver Education Evaluations

In the last decade or so, quite a number of broad scholarly reviews of driver education evaluation have appeared; seemingly outnumbering the evaluation studies themselves (Christie 2001; Engstrï¿½m et al. 2003; Lonero et al. 1994, 1995; Mayhew and Simpson 1997; Mayhew, Simpson and Robinson 2002; McNeil in press; Nichols 2003; Siegrist 2003; Smiley, Lonero, and Chipman 2004; Woolley 2000). These reviews have usually addressed driver education in conjunction with other forms of driver instruction or graduated driver licensing (GDL). In addition to the broader scholarly/analytic reviews; two ï¿½systematicï¿½ reviews ( Roberts and Kwan 2004; Vernick et al. 1999 ) , and a single quantitative meta-analysis (Elvik & Vaa, 2004) have appeared.

In the most inclusive and detailed of the earlier reviews, Mayhew and Simpson (1997) set up a score card of findings: 7 studies showed a positive effect, 16 showed no effect, and about 7 others showed a negative safety effect. The authors, consistent with others reviewers, conclude that, on balance, the literature is negative toward driver education effectiveness.

The Elvik and Vaa meta-analysis is potentially the most interesting of all, but unfortunately, the only report on the meta-analysis is a very brief one in the 2004 edition of the Norwegian Handbook of Road Safety Measures, although it is available in English. The combined data of 16 studies indicated that driver education graduates have 1.9% fewer crashes per driver (confidence interval, -3.8%; 0%). The overall difference is borderline statistically significant, as the combined results had the power to detect a significant difference (Î±=.05) if the result would have been 2% (rather than 1.9%). Per kilometer driven graduates had 4% lower crash rate (6%; -2%). When the combined results were limited to the experimental studies, a different picture emerged. No difference per driver surfaced (+/- 4%). Per kilometer driven, graduates had 11% more crashes (+8%; +15%). The authors conclude that the combined evaluation results do not indicate that driver education reduces crashes over the first couple of years of driving.

All the earlier reviews were addressed to finding whether the evaluation literature suggests that driver education ï¿½works.ï¿½ As part of their methodologically oriented review, Clinton and Lonero (2006a) categorized the earlier reviews and discussed critically the reviewsï¿½ conclusions and limitations. There is a strong trend toward systematic review, particularly in the health fields, but the often narrow orientation of this approach to research review seems to limit its applicability to driver education evaluation. The more conventional scholarly reviews vary greatly in scope and quality. Some point to ways to improve driver education, but they all failed to provide much critique of the evaluations, and they do not provide much guidance as to how evaluation research might be improved.

Improving Driver Education Evaluation

A serious lack of statistical and research design sophistication was evident in the early evaluation studies (Nichols 2003). Many of the later studies have used better designs and statistical methods. No perfect study exists. Even the ï¿½gold standardï¿½ random controlled trials have had weakness in terms of maintaining clean assignment of students to training conditions and clear comparisons between assigned groups.

In the mid-1980s, the OECD assessed the efficacy of road safety education programs and provided strategies for program development, implementation, and evaluation (OECD 1986). This report outlined issues of effectiveness for program planning and implementation, but its primary focus was evaluation. It was suggested that a program should be seen as effective if it does what it was intended to do, and that it is, therefore, very important to be explicit about educational objectives, which should include measures of intermediate effects and not be solely focused on collisions. Not much happened in response to the OECDï¿½s astute directions.

With notable exceptions, such as the DeKalb experiment, Dreyer and Jankeï¿½s 1979 experimental study, and Gregersenï¿½s 1994 survey study, most evaluations have failed to look at intermediate outcomes. Lacking information on what the students have or have not learned, directions for program improvement are left unclear. Most existing evaluations leave many unanswered questions regarding:

Logical links between curricula and young driversï¿½ needs;
Theories explaining how a program is expected to achieve safer driving;
The quality, comprehensibility, and usability of curriculum products;
How well and how consistently the instructional processes actually deliver the intended learning; and
Which learning and behavioral outcomes result, or fail to result, from the training.

Tracking of learning outcomes is an area where programs could easily build in ongoing evaluation. Knowledge and attitude measures at the beginning and end of the course and at later intervals would help keep contact with graduates and provide feedback for continuous improvement of curriculum and delivery.

Exposure differences are too often ignored in driver education evaluation. In general, simple differences in the amount of driving (exposure to risk) account for much of the differences in crash rates between groups of drivers (Chipman 1982). Qualitative exposure differences, such presence of passengers and trip purposes, also represent different collision risks for young drivers (Preusser 1996; Preusser, Ferguson, and Williams 1997; Williams, 2003).

A key implication of the evaluation literature is that there is no perfect method for evaluation of beginner driver education. A broader and more systematic evaluation approach is needed, which includes various study designs, a wide range of output and outcome measures, and a variety of carefully planned comparisons. Perhaps it is time the ï¿½gold standardï¿½ title is passed along to comprehensive, systematic evaluation that includes various approaches to developing the whole picture of program effects and improving programs.

The final implication of past evaluations of driver education is how little they have contributed to developing and improving the programs. What has occurred has been very unsystematic; no study seemed to relate to or build upon earlier studies. Even credible positive findings, such as those of Dreyer and Janke (1979), were not been followed up. Scientific knowledge usually develops through systematic replication of research, but that has not yet happened in driver education research. Earlier thinking, including that behind the DeKalb project, seemed to favor a single massive development effort to achieve driver educationï¿½s safety goals in a one big step. More recently, however, researchers and theorists have emphasized the importance of incrementally building knowledge gains and other intermediate effects, (Lonero et al. 1994; Woolley 2000), as well as continuously developing and improving programs. Keskinen et al. wrote: ï¿½We have decided ï¿½ that the development of driver education will take place in short steps, with constant evaluations of the results and trying to avoid solutions which are thought to be finalï¿½ (1998, p. 382).

Support for new and more systematic evaluation research has been made available. Projects are being undertaken at the National Highway Traffic Safety Administration (NHTSA). Under funding from the AAA Foundation for Traffic Safety and the National Center for Injury Prevention and Control of the Centers for Disease Control, Northport Associates and the Traffic Injury Research Foundation are implementing a program of ongoing, related and systematic evaluations of a number of driver education programs during 2007-2009. The program initially focuses on three jurisdictions ï¿½ Michigan, Oregon and Manitoba, with other sites intended to be added as broader financial support is found for the Large Scale Evaluation of Driver Education (LSEDE).

Summary and Conclusions

Reviewers of the evaluation literature have typically concluded that beginner driver education has yet to demonstrate clear success in improving the safety of new drivers who receive it, or of youth on a population basis. A few studies have shown positive effects, and not only poorly designed studies. Keys to long-term effectiveness of driver education, if it is to make a net contribution to safety, are likely to be astute evaluation and continuous improvement of program content, delivery, organization, regulation and coordination with other healthy influences.

If driver education is to achieve success in its demanding mission, it needs to be more firmly based, on the one hand, in sound research and theory concerning young driver skills, behavior, motivation, and risk, and on the other in the principles of effective behavior change. Program evaluation is critical for more effective program development in the future, but its limitations in the past must be recognized and corrected for a beneficial effect. A more comprehensive approach to evaluation needs to address program theory, products, processes, and management.

Evaluating intermediate outcomes ï¿½ changes in behavior, knowledge, attitudes, and exposure to risk ï¿½ is also required. Intermediate measures should continue during the follow-up period if we are to have a clear picture of the effects of the program and the reasons for them. Meeting the ultimate goal of reducing novice driversï¿½ serious crashes will also require better management the linkage of driver education with parental and community influences, graduated licensing, and other behavioral influences such as incentives and disincentives. The increasing need, vigorous development in public and private programs, clearer guidelines for evaluation, and support for more systematic, ongoing evaluation research all point toward data-driven development of better driver education programs and ultimately improved safety outcomes.

REFERENCES

Carstensen, G. (2002). The effect on accident risk of a change in driver education in Denmark. Accident Analysis and Prevention 34(1):111-121.

Chipman, M.L. (1982). The role of exposure, experience and demerit point levels in the risk of collision. Accident Analysis and Prevention 14(6):475-483.

Christie, R. (2001). The effectiveness of driver training as a road safety measure: a review of the literature. Melbourne, Australia: Royal Automobile Club of Victoria (RACV) Ltd., Public Policy Group.

Christie, R. (2002). Road safety education and training from a public health perspective. Road Safety Research, Policing and Education Conference Proceedings.

Clinton, K. & Lonero, L. (2006a) Evaluation of Driver Education: Comprehensive Guidelines, Washington, D.C.: AAA Foundation for Traffic Safety.

Clinton, K. & Lonero, L. (2006b) Evaluation of Driver Education: How-to Guide, Washington, D.C.: AAA Foundation for Traffic Safety.

Davis, C.S. (1990). The DeKalb County, Georgia, Driver Education Demonstration Project: Analysis of its long-term effect. Washington, DC: U.S Department of Transportation, National Highway Traffic Safety Administration.

Dreyer, D., and M. Janke. (1979). The effects of range versus nonrange driver training on the accident and conviction frequencies of young drivers. Accident Analysis and Prevention 11:179-198.

Road Safety Educatorsï¿½ Association ï¿½ Driving School Association of the Americas (2005). Process for Curriculum Review. Ontario, Canada: DSAA and RSEA.

Elvik, R., and T. Vaa, eds. (2004). The Handbook of Road Safety Measures. Amsterdam: Elsevier.

Engstrï¿½m, I., N.P. Gregersen, K. Hernetkoski, E. Keskinen, and A. Nyberg. (2003). Young novice drivers, driver education and training: Literature review. Linkï¿½ping, Sweden: Swedish National Road and Transport Research Institute.

Fisher, D.L., N.E. Laurie, R. Glaser, K. Connerney, A. Pollatsek, S.A. Duffy, and J. Brock. (2002). Use of a fixed-base driving simulator to evaluate the effects of experience and PC-based risk awareness training on drivers' decisions. Hum Factors 44(2):287-302.

Forsyth, E.G. Maycock, and B. Sexton. (1995). Cohort study of learner and novice drivers: Part 3, Accidents, offences and driving experience in the first three years of driving. Crowthorne, Berkshire: Transport Research Laboratory.

Fuller, R., and D. Bonney. (2003). Driver education and training in post-primary schools in Ireland. In Behavioural research in road safety: Thirteenth seminar. Crowthorne, Berkshire: Transport Research Laboratory.

Fuller, R., and D. Bonney. (2004). Driver education and training in post-primary schools. Monograph. Dublin: National Council for Curriculum and Assessment.

Glad, A. (1988). Phase 2 Driver Education: Effect on the risk of accident. Oslo, Norway: Norwegian Center for Research, Department of Transportation.

Green, L.W., and M.W. Kreuter. (1999). Health promotion planning: An educational and environmental approach. 3 rd ed. Mountain View, CA: Mayfield Publishing (McGraw-Hill).

Gregersen, N.P. (1994). Systematic cooperation between driving schools and parents in driver education, An experiment. Accident Analysis and Prevention 26(4):453-461.

Howorth, N., N. Kowadlo, and C. Tingvall. (2000). Evaluation of pre-driver education program. Victoria, Australia: Monash University Accident Research Centre.

Highway Safety Center. (2002). Traffic Safety Education Life Long Learning Process. Indiana, PA: Indiana University of Pennsylvania.

Katila, A., E. Keskinen, S. Laapotti, and M. Hatakka. (1995). Changes in slippery road accidents as an effect of renewed driver training in Finland. Turku, Finland: University of Turku.

Keskinen, E., M. Hatakka, and A. Katila. (1998). The Finnish way of driver improvement: Second phase of training. Paper presented at International Driver Improvement Workshop, at Bergisch Gladbach, Germany: Bundesanstalt fur Strassenwesen.

Levy, D.T. (1990). Youth and traffic safety: The effects of driving age, experience, and education. Accident Analysis and Prevention 22(4):327-334.

Lonero, L., and K. Clinton. (1998). Changing road user behavior: What works and what doesnï¿½t. PDE Publications. Available from www.drivers.com.

Lonero, L., K. Clinton, J. Brock, G. Wilde, I. Laurie, and D. Black. (1995). Novice driver education model curriculum outline. Washington, DC: AAA Foundation for Traffic Safety.

Lonero, L.P., K.M. Clinton, G.J.S. Wilde, K. Roach, A.J. McKnight, H. Maclean, S.J. Guastello, and R. Lamble. (1994). The roles of legislation, education, and reinforcement in changing road user behaviour. Toronto, Ontario: Ontario Ministry of Transportation.

Lonero, L.P., K.M. Clinton, B.N. Persaud, M.L. Chipman, and A.M. Smiley. (2005). A longitudinal analysis of Manitoba Public Insurance Driver Education Program. Winnipeg, Manitoba: Manitoba Public Insurance.

Lonero, L. & Clinton, K. (2006) Evaluation of Driver Education: Management Overview, Washington, D.C.: AAA Foundation for Traffic Safety.

Lonero, L., & Clinton, K. (2006) Situation Analysis of Driver Education in Canadian Provinces and Territories. Ottawa, Ontario: Natural Resources Canada.

Lonero, L., Clinton, K., Sleet, D. (2006). Behavior Change Interventions in Road Safety.(in) Geilen, A.C., Sleet, D.A., & DiClemente, R.J. Injury and Violence Prevention: Behavioral Science Theories, Methods, and Applications, San Francisco, California: Jossey-Bass.

Lund, A.K., A.F. Williams, and P. Zador. (1986). High school driver education: Further evaluation of the DeKalb County study. Accident Analysis and Prevention 18 (4):349-357.

Masten, S.V., and E. A. Chapman. (2003). The effectiveness of home-study driver education compared to classroom instruction: The impact on student knowledge, skills, and attitudes. Report to the Legislature of the State of California. Report no. CAL-DMV-RSS-03-203. Sacramento, CA: California Department of Motor Vehicles.

Masten, S.V., and E.A. Chapman. (2004). The effectiveness of home-study driver education compared to classroom instruction: The impact on student knowledge and attitudes. Traffic Injury Prevention 5(2):117-121.

Maycock, G. (1995). Accidents in the first three years of driving. TRL Annual Review: 71-77.

Maycock, G., and E. Forsyth. (1997). Cohort study of learner and novice drivers. Part 4: Novice driver accidents in relation to methods of learning to drive, performance in the driving test and self assessed driving ability and behaviour. Crowthorne, Berkshire: Transport Research Laboratory.

Mayhew, D.R., and H.M. Simpson. (1995). The role of driving experience: Implications for the training and licensing of new drivers. Toronto, ON: Insurance Bureau of Canada.

Mayhew, D.R., and H.M. Simpson. (1997). Effectiveness and role of driver education and training in a graduated licensing system. Ottawa, Ontario: Traffic Injury Research Foundation.

Mayhew, D.R., H.M. Simpson, and A. Robinson. (2002). The safety value of driver education and training. Injury Prevention 8 (Supplement II):3ii-8.

Mayhew, D.R., H.M. Simpson, and A. Pak. (2003). Changes in collision rates among novice drivers during the first months of driving. Accident Analysis and Prevention 35(5):683-691.

McKenna, C.K., B. Yost, R.F. Munzenrider, and M.L. Young. (2000). An evaluation of driver education in Pennsylvania. Harrisburg, PA: Pennsylvania Department of Transportation.

McKnight, A.J. (1984). Driver education ï¿½ when? Journal of Traffic Safety Education (Jan):13-15.

McKnight, A.J., and A.S. McKnight. (2003). Young novice drivers: Careless or clueless? Accident Analysis and Prevention 35(6):921-925.

MacNeil, S. (in press). Driver education, training and licensing. In R.E Dewar and P. L. Olson. Human Factors in Traffic Safety. Second edition. Tucson, AZ: Lawyers and Judges Publishing Co., 2007.

Nichols, J. L. (2003). Driver education review. Paper presented at the NTSB Forum on Driver Education and Training, October 2003 . Washington, DC: National Transportation Safety Board.

National Institute of Mental Health. (2004). Teenage brain: A work in progress. NIMH, 05/13/04 2001 (accessed November 16, 2004). Available from http://www.nimh.nih.gov/publicat/teenbrain.cfm.

National Research Council, Institute of Medicine, and Transportation Research Board. (2007) Preventing Teen Motor Crashes: Contributions from the Behavioral and Social Sciences, Workshop Report. Program Committee for a Workshop on Contributions from the Behavioral and Social Sciences in Reducing and Preventing Teen Motor Crashes. Washington, DC: The Natinal Academies Press.

National Transportation Safety Board. (2005). National Transportation Safety Board Public Forum on Driver Education and Training, October 29-30, 2003. Report of Proceedings. Washington, DC: NTSB.

OECD. (1986). Effectiveness of road safety education programmes. Paris: Road Transport Research, Organisation for Economic Co-operation and Development.

OECD Scientific Expert Group. (1990). Behavioural adaptations to changes in the road transport system. Paris: Road Transport Research, Organisation for Economic Co-operation and Development.

Patton, M.Q. (1997). Utilization-focused evaluation: The new century text. 3rd ed. Thousand Oaks, CA: Sage Publications.

Plato, K.C., and A. Rasp, Jr. (1983). Parental views on issues related to traffic safety education and the licensing of teenage drivers. Olympia, Washington: Washington Office of the State Superintendent of Public Instruction.

Potvin, L., F. Champagne, and C. Laberge Nadeau. (1988). Mandatory driver training and road safety: The Quebec experience. American Journal of Public Health 78:1206-1209.

Preusser, D.F. (1996). Licensing practices and crash risk in the United States. In New to the road: Reducing risks for young motorists. Proceedings of the First Annual Symposium of the Youth Enhancement Service, June 8-11, 1995, edited by H. Simpson. Los Angeles, CA: Youth Enhancement Services, University of California.

Preusser, D.F., S.A. Ferguson, and A.F. Williams. (1997). Effect of teenage passengers on the fatal crash risk of teenage drivers. Arlington, VA: Insurance Institute for Highway Safety.

Quadrel, M.J., B. Fischhoff, and W. Davis. (1993). Adolescent (in)vulnerability. American Psychologist 48(2):102-116.

Ray, H.W., M. Sadof, J. Weaver, J.R. Brink, and J.R. Stock. (1980). Safe Performance Secondary School Driver Education Curriculum Demonstration Project. Washington, DC: U.S. Department of Transportation, National Highway Traffic Safety Administration.

Roberts, I., and I. Kwan. (2004). School based driver education for the prevention of traffic crashes. The Cochrane Library (1).

Robertson, L.S. (1980). Crash involvement of teenaged drivers when driver education is eliminated from high school. American Journal of Public Health 70 (6):599-603.

Robertson, L.S., and P.L. Zador. (1978). Driver education and fatal crash involvement of teenage drivers. American Journal of Public Health 68(10):959-965.

Road Safety Educatorsï¿½ Association. (2005). Process for curriculum review. Ontario, Canada: DSAA.

Shope, J.T., and L.J. Molnar. (2003). Graduated driver licensing in the United States: Evaluation results from the early programs. Journal of Safety Research, 34(1S):63-69.

Siegrist, S. (2003). Driver training and licensing ï¿½ The European perspective . Paper presented at the NTSB Forum on Driver Education and Training, October, 2003. Washington, DC: National Transportation Safety Board.

Simpson, H., ed. (1996). New to the road: Reducing the risks for young motorists. First Annual International Symposium of the Youth Enhancement Service, June 8-11, 1995: Youth Enhancement Service.

Sleet, D.A. & Lonero, L. (2002). Behavioral approaches to preventing motor vehicle crashes. In L. Breslow, J. Last, L. Green, & M. McGinnis (Eds.), Encyclopedia of Public Health. Mosby.

Smiley, A., L. Lonero, and M. Chipman. (2004). Final Report. A review of the effectiveness of driver training and licensing programs in reducing road crashes. Paris, France: MAIF Foundation.

Smith, D.L. (1983). The DeKalb Driver Education Project ï¿½ The same mistakes: Improper criteria. Journal of Traffic Safety Education 30(January):14 & 26.

Smith, M.F., and J. Blatt. (1987). Summary of preliminary results: Follow-up evaluation. Safe Performance Curriculum Driver Education Project. Spokane, WA: American Driver & Traffic Safety Education Association.

Smith, M. F. Research agenda for an improved novice driver education program: A report to congress. U.S. Department of Transportation, National Highway Traffic Safety Administration, Washington, D.C. 1994.

Stock, J.R., J.K. Weaver, H.W. Ray, J.R. Brink, and M.G. Sadof. (1983). Evaluation of Safe Performance Secondary School Driver Education Curriculum Demonstration Project. Washington, DC: U.S. Department of Transportation, National Highway Traffic Safety Administration.

Strang, P.M., K.B. Deutsch, R.S. James, and S.M. Manders. (1982). A comparison of on-road and off-road driver training. Canberra City, Australia: Department of Transport and Construction Office of Road Safety.

Stufflebeam, D.L. (2001). Evaluation Models. New Directions for Evaluation (89):1-106.

Stufflebeam, D.L. (2003). The CIPP model for evaluation. Paper presented at Annual Conference of the Oregon Program Evaluators' Network, October 3, at Portland, Oregon.

Taylor-Powell, E., L. Jones, and E. Henert. (2002). Enhancing Program Performance with Logic Models. [Accessed November, 2004]. Available from the University of Wisconsin-Extension web site: http://www1.uwex.edu/ces/lmcourse/

The Health Communication Unit. (2006). Evaluating health promotion programs. Toronto, ON: Centre for Health Promotion, University of Toronto.

Trochim, W. (2001). The research methods knowledge base. Cincinnati, OH: Atomic Dog Publishing.

Vernick, J.S., G. Li, S. Ogaitis, E.J. MacKenzie, S.P. Baker, and A.C. Gielen. (1999). Effects of high school driver education on motor vehicle crashes, violations, and licensure. American Journal of Preventive Medicine 16(1S):40-46.

Weaver, J.K. (1987). Summary of preliminary results. Follow-up evaluation. Safe Performance Curriculum Driver Education Project. Washington, DC: U.S. Department of Transportation, National Highway Traffic Safety Administration.

Wiggins, S. (2005). Graduated Licensing Program interim evaluation report - Year 3. Vancouver, BC: Insurance Corporation of British Columbia.

Wilde, G.J.S. (1993). Improve risk perception and reduce risk acceptance: Two proposals for driver education. In Novice Driver Education Symposium, University of Alberta, Edmonton, Alberta, April 22-33. Edmonton, AB: Alberta Motor Association Driver Education.

Wilde, G.J.S. (1994). Risk homeostasis theory and its promise for improved safety. In Challenges to accident prevention: The issue of risk compensation behaviour, edited by R. M. Trimpop and G. J. S. Wilde. Groningen, The Netherlands: Styx Publications.

Williams, A.F. (2003). Teenage drivers: Patterns of risk. Journal of Safety Research, 34(1S), 5-15.

Woolley, J. (2000). In-car driver training at high schools: A literature review. Adelaide, Australia: Transport SA.

Wynne-Jones, J.D., and P.M. Hurst. (1984). The AA driver training evaluation. In Wellington, New Zealand: Ministry of Transport. Traffic Research Report No. 33. Wellington, New Zealand: Ministry of Transport.

Zhao, J., R.E. Mann, M. Chipman, E. Adlaf, G. Stoduto, and R.G. Smart. (2006). The impact of driver education on self-reported collisions among young drivers with a graduated license. Accident Analysis and Prevention 38:35-42.

Further comments to this article have been disabled.

All Comments (5)

Showing 1 - 5 comments

nathan,

avg atey viris saner

Drivers Advocate,

Please have a look at driversadvocate.org and let me know if any of the information provided there about driving schools and driver education fromt the experience of a driving instructor helps or can assist in this discussion at all. Thank you.

Terri Harwell,

Many driving schools that I have worked for in Texas are all about $$$ and little about what it takes to teach a novice to drive. Two friends and I are hoping to make a difference with the opportunity awarded to us to be able to train the trainers.

Clayton M Collier,

Larry,
Could you expand on what you said regarding DE and the government road tests?
Clayton

Pierro Hirsch,

Hello Larry,

As usual, your perspective is wide and your approach is comprehensive.

You mention the need to address the "Logical links between curricula and young drivers’ needs" and you also refer to Carstensen's evaluation of the changes in the Danish driver education.

In Denmark, the curriculum was altered to focus on the development of crash prevention skills from the start of the training (vs car control skills )and these same crash prevention skills were tested at the licensing stage. The evaluation showed decreases in multiple vehicle crashes but not single vehicle crashes.

My appreciation of learner drivers is that their focus is on passing their primary license exam. While most candidates also have some spare mental capacity for safety issues, the license exam criteria dominate. On a population level in a free market, we can only assume that what is on the license exam will be consistently taught to all paying clients of driving schools.

Any evaluation of DE effectiveness that does not also consider the effect of the government license exam criteria is missing a huge piece of the puzzle. The Danes seem to be the only ones I have found who addressed this issue.

The fact that the improvements in the Danish program did not influence single car crashes, the usual proxy for intentional risk taking crashes, indicates to me that other measures are needed to identify and rehabilitate certain subgroups, those endowed with more skill than judgment, prior to allowing them to drive without supervision.

So, what do you think the chances are that someone in North America will be able to consider the influence of the government license exam in relation to driver education programs?

Larry, if you ever pass through Montreal, come and visit me at Virage Simulation (www.viragesimulation.com).
We are working with the SAAQ to validate the effectiveness of driver simulator training for novice drivers and I would love to have your feedback on our project.

Best regards,

Pierro