Mopeds at Night

An analysis of Virginia’s moped safety and registration law on operator collision rates and injury outcomes

2017 Terry Lathrop Award, best Master’s project in transportation

Joseph Seymour

Vulnerable road users (VRU), which includes pedestrians, cyclists, and motorcyclists, are found blocked in the midst of transportation policies and roadway infrastructures that are heavily designed for standard automobiles. Mopeds, a subgroup of motorcycles, are frequently neglected due to its vehicle classification as a gray area between motorcycles and bicycles. As many states do not require moped registration, estimates for moped sales in the state and national levels are difficult if not impossible to procure, and due to its being overlooked, no tools or requirements exist to record and track automobile safety reporting. Literature on moped safety is limited to areas whose moped presence is denser, such as Australia and Europe, and most research based in the US is largely limited to the effects of alcohol on crash severity, rather than on traffic safety policies.

In 2013 Virginia enacted SB 1038, a two-phase measure designed to improve moped user safety through vehicle registration, identification, and safety requirements. Phase I, effective July 1, 2013, requires moped operators to wear helmets and carry government issued identification. Phase II, effective July 1, 2014, requires users to register their mopeds at the Virginia Department of Motor Vehicles. In order to test the effectiveness of both phases, the policy’s pre- and post-policy implementation periods were assessed using a three-part evaluation framework consisting of an Interrupted Time Series Analysis (ITSA), a Comparative Interrupted Time Series Analysis (CITSA), and logistic regression. The ITSA compared levels of collisions, injuries, and fatalities during the pre- and post-policy months (January 2010 ~ July 2013 and July 2013 ~ July 2016, respectively). The CITSA, on the other hand, sought to assess SB 1038’s Phase II effectiveness by comparing Virginia’s collisions, injuries, and deaths with those of North Carolina (July 2013 ~ July 2015), whose moped safety laws have included a helmet policy since the 1970s, and which enacted a measure similar to SB 1038’s Phase II in 2015. A logistic regression using the ITSA’s same time frame provided an analysis on the risk of serious injury and fatality during SB 1038’s pre- and post-implementation.

The ITSA and CITSA produced conflicting results: according to the ITSA, Phase I saw reduced collision rates in the months following its implementation, but the time period after Phase II showed no reduction in serious injury or fatality rates. The CITSA, however, found an increase in collision overall, including time after Phase II. An ITSA performed on Virginia data alone reflecting the original CITSA’s time frame found no significant outcomes for rates of injuries and collision after Phase II implementation, suggesting an issue with the use of North Carolina as a control. A logistic regression, while finding several significant variables that affected risk of injury (e.g. urban/rural designation, alcohol consumption, helmet use) found no support for Phase II’s part in reducing any risk of serious injury or outcomes. This result supports the ITSA’s findings.

Combining results from the ITSA, CITSA, and logistic regression suggest that while a decrease in moped collision rates was found between the implementation of Phase I and Phase II, the frequency and risk of serious injury/fatality saw no reduction in the same time frame. Sample attenuation from a third variable may be at play, in which moderate risk-taking individuals are reduced from the pool of moped riders following SB 1038’s new regulation, leaving high risk-taking individuals to continue their operating (leading to lower collision rates, but persisting injury/fatality rates). Another possibility can be found in the measurement of moped collision rates, due to the fact that data regarding registered moped populations prior to SB 1038 are limited, and substitution methods may not be entirely accurate. Future studies evaluating SB 1038 must consider ways to account for this. Nevertheless, this analysis, which sought to clarify SB 1038’s influence on moped safety, shows the policy’s partial effectiveness in reducing moped collisions, and the implementation of similar policies in other states (SB 1038’s Phase I in particular) may have a positive effect on reducing moped collision rates overall.

Comments are closed.