CHAPTER 4

Conclusions and Suggested Research

Conclusions

The analyses conducted for this project contribute to the field of teen driver safety by providing information on the relationship between the amount of supervised practice driving during the learner’s permit period and safety outcomes when driving independently. Using SPDS NDS data and state-of-the-practice Poisson regression models, the researchers found that teens with more driving exposure during the learner’s permit phase had lower CNC rates than teens with less driving exposure during the learner’s permit phase.

This analysis was conducted by splitting the participant pool using normalized mean hours of supervised practice driving during the learner’s permit phase. Those teens with more driving experience than the normalized mean were grouped into the more-driving-exposure group. Those teens with less driving experience than the normalized mean were grouped into the less-driving-exposure group. A related analysis of total hours of supervised practice driving during the learner’s permit period indicated that approximately half of the participants in this study completed the 45 required hours of learner’s permit driving, as required by the state of Virginia. These combined results suggest that even with a state-regulated number of hours of supervised practice driving during the learner’s permit period, at least half of all teens may not be getting the full number of hours of learner’s permit driving, much less meeting any other requirement, such as a set number of hours of driving at night. Given this result, adding requirements for parents/teens to use electronic-logging cell phone apps to help track the amount of practice teens are getting may help improve compliance and increase the number of hours of practice that teens actually get during the learner’s permit driving phase. The findings of this study suggest that more driving exposure for teen drivers results in fewer crashes on our roadways.

Given these results, the total number of hours that states recommend for supervised practice driving during the learner’s permit phase should probably be no less than the 45 hours required by the state of Virginia (based on this analysis). Perhaps there could be a trade-off where, if a state requires approximately 45 hours of supervised practice driving or less during the learner’s permit phase, then stricter requirements surrounding logging these practice hours would be required for parents. If a much higher number of hours of supervised practice were required, then the electronic-logging restriction could be less strict. Additional research will need to be conducted on electronic logging of supervised practice driving during the learner’s permit phase and how effective and efficient this requirement is at improving the total number of hours of practice for teen drivers.

These analyses also aimed to better understand the impact on safety outcomes of driving diversity during learner’s permit and early independent driving. While analyses using SCE did not result in significant findings, analyses using KRD rates were beneficial in several ways. First, less exposure to night driving during the learner’s permit phase was associated with a significant increase in hard braking events when teens drove independently at night. This strongly

suggests that teens need more practice during nighttime hours during the learner’s permit phase. Additionally, there was a significantly higher number of hard braking events when teens were driving on unfamiliar routes than when they were driving on familiar routes. This also provides evidence that teens should be practicing on unfamiliar routes during the learner’s permit phase to improve safety outcomes once they start driving independently. More exposure to new and unique roadways will result in improved safety outcomes.

The last section of the report focused on specific driver behaviors such as speeding and passenger presence. Teen drivers sped more frequently on roadways with 35, 55, and 60 mph speed limits. Speeding on lower-speed roadways is concerning because the lower-speed roadways tend to have more pedestrians and driveways where severe crashes can occur. Speeding on highways is concerning because higher-speed crashes, in general, are more severe and result in more fatalities. Speed cameras in locations where high rates of speeding may occur may be useful in reducing these speeding behaviors. Additionally, monitoring and feedback on cell phone apps that inform teens (and their parents) when they are speeding may provide useful information and help teens reduce their speeds in these critical locations.

Finally, the percentage of trips where adults (most often parents) rode with teen drivers once the teen was driving independently was low. As compared to teens with more driving practice, teens with less driving practice during the learner’s permit phase drove alone or with teen passengers more often than with adults. Better educating parents on the benefits of continuing to ride with their teen drivers, especially when their teens are driving in new, complex, or unique roadway environments for the first time, is beneficial in improving safety outcomes.

Suggested Research

While this work found that more practice driving during the learner’s permit phase does improve safety outcomes, the analyses were less definitive regarding the diversity of practice driving. Additional data collection and analyses to better understand the relationship between diversity in supervised practice driving during the learner’s permit phase and safety outcomes would greatly enhance the findings of these analyses. Additionally, gaining insight into the types of diversity that may be most important for safety outcomes could greatly enhance supervised practice driving during the learner’s permit phase. For example, while more experience with driving at night is likely important, it would be more beneficial to gain experience on higher-speed roadways, in heavy traffic conditions, or in navigating complex road configurations such as roundabouts, multilane intersections, and/or other unique road configurations. A better understanding of teen driving behavior differences in diverse driving conditions could be informative for driver’s education instructors and help licensing agencies improve driver testing procedures.

Additional research that better assesses the optimal number of hours of supervised practice driving during the learner’s permit phase would be helpful. While the “optimal” number of hours of practice is likely dependent upon the individual, there may be an “optimal” number of hours of practice that would be beneficial for 80% of the population, 90% of the population, etc., whereby the number of teen driver crashes would be significantly reduced. To conduct this research, additional NDSs may need to be conducted, though these may be able to be conducted using more streamlined, commonly available data acquisition systems as opposed to the complex data acquisition system used for the NDSs used in these analyses.

Research that determines better ways to motivate teen drivers to drive within the speed limit and not engage in hard braking/hard cornering maneuvers would be informative. While research has shown that monitoring and feedback systems have promise for improving safety outcomes (Carney et al. 2010; Klauer et al. 2017), additional research is needed to assess whether there are ways to motivate teen drivers without also relying on their parents.

Finally, research on messaging for and education of parents of teen drivers would also be helpful. Several states have incorporated a 90-minute parent session into the driver’s education curriculum. This is an opportunity to educate parents not only about the risks facing teen drivers but also about the steps parents can take to reduce the risk of their teen being involved in a crash. This report highlights multiple messages for parents, including making sure they provide as much supervised driving practice during the learner’s permit phase as possible, continuing to ride with their teen after independent licensure, and ensuring broad exposure to different roadway/traffic environments, especially nighttime driving, during the learner’s permit period. Creating effective messaging for parents can greatly improve teen driving safety and warrants more research.

Limitations

There are several limitations to these analyses, some due to the constraints of NDSs in general and some in terms of the analyses themselves.

With all NDSs, sample size is a concern, as is the type of participants who agree to participate in NDSs. The SPDS NDS was a sample of 90 teenagers in a similar geographic location (Southwest Virginia). Because all the participants were from a homogeneous geographic location, it was difficult for the analyses to parse out different types of roadways in ways that also maintained statistical power. While the SHRP 2 NDS dataset had 254 teen drivers, only a small amount of data was collected during the first 6 months of driving (55 participants with an average of 2.5 months of data collected for each driver during the first 6 months). This limited data for the first 6 months of driving (i.e., early independent driving) made it difficult to use the SHRP 2 NDS in assessing behavior by initial driving experience.

Another limitation of many NDSs is that participants must agree to sign up for the research study. While there is research to suggest that crash risk is not uniform across race and ethnicity (Glassbrenner et al. 2022), it can be difficult and costly to recruit across racial and ethnically diverse populations, as different populations might need different recruitment approaches and different messaging. The current NDS populations are fairly homogeneous. The SHRP 2 NDS population is less than 1% Latino, and race and ethnicity were not a variable found in the demographic questionnaire.

For every research project, there is a limit to the resources available. This is true from a financial perspective as well as a time perspective. The research team would have liked to include additional analyses of speeding behavior for the SHRP 2 NDS participants and to have examined the relationship of distracted driving behaviors to the amount of practice driving. The team ran out of both financial and time resources; however, the analyses that were performed were those that the research team believed were the most critical.

This research utilized powerful NDS datasets coupled with innovative statistical models that allowed the research team to expand upon previous knowledge and gain critical new results regarding the positive relationship between the number of hours of supervised practice driving during the learner’s permit phase and a reduction in SCEs. These analyses also further corroborate research suggesting that more diverse practice driving during the learner’s permit phase leads to reductions in KRD rates. Future research could continue to investigate the role of practice driving hour requirements and practice driving in diverse roadway environments in reducing the heightened crash risk facing teen drivers during their first months of independent driving.