E-scooters have often been identified as more dangerous than e-bikes, but that picture changes when they are compared on equal terms. A recently published study from Chalmers University of Technology, Sweden, shows in fact that the crash risk is eight times higher for e-bikes than for e-scooters, calculated based on the trip distance with rental vehicles in cities. This surprising result provides a better basis for cities to make decisions on how much to facilitate different types of micromobility.
E-scooters have become an increasingly common, and often hotly debated, means of transport in cities. The media's reporting on these crashes has been intense at times, and a number of previous studies have identified e-scooters as far more dangerous than e-bicycles and ordinary bicycles.
But according to a new study from Chalmers University of Technology in Sweden, it is not as simple as that.
"Previous studies have often compared apples with oranges," says Marco Dozza, Full Professor in Active Safety and Road-User Behaviour at Chalmers. "They have lumped together e-bicycles with ordinary bicycles, and haven't taken into account where, how and how much these vehicles are used – or whether they are rented or privately owned. When we took all these factors into account, we found that e-scooterists actually have a lower rate of crashes than e-cyclists."
GPS data contributed to equitable comparison
The study, which is published in the Journal of Safety Research, is based on a unique data set from trips using rented e-bicycles and e-scooters in seven European cities: Gävle in Sweden, Berlin and Düsseldorf in Germany, and the UK cities of Cambridge, Kettering, Liverpool and Northampton.
The researchers analysed 686 crashes involving e-scooterists and 35 involving e-cyclists. The high number of crashes involving e-scooters reflects that they were used much more frequently than e-bicycles. But their crash risk was actually much lower – regardless of whether the risk was calculated on the basis of the number, duration, or distance of the trips.
"When we calculated using trip distance, it turned out that e-cyclists were eight times more likely to have a crash than e-scooterists. It's a result that surprised us," says Marco Dozza.
This is the first time that a study of this kind has been able to compare micromobility in such a detailed and equitable way, and from so many countries and cities. A key to being able to do the study in this way was the use of GPS data. This made it possible to measure what is termed 'exposure' – which refers to how much a vehicle is actually used – with greater precision than previously.
All vehicles in the study were rented and used in city centres, which makes the comparison more equitable than previous studies that have often mixed together urban and rural settings, or mixed rented vehicles with privately owned vehicles.
Safety of e-scooters grossly underestimated
Despite their results, the researchers stress that they should not be seen as definitive proof that e-scooters are safer than e-bicycles. Uncertainties remain, such as under-reporting of crashes and differences in the way these vehicles are used.
"But what we can say is that previous studies have grossly underestimated the safety of e-scooters in relation to e-bicycles," says Marco Dozza. "This in turn could have consequences for how cities regulate and plan micromobility. In some cities, attempts are being made to steer micromobility towards e-bicycles, which are considered to be better because previous research has created the idea that all types of cycling are safer than all types of e-scootering," he adds.
"Now that it turns out that isn't correct, decision-makers may need to think again. The results might also affect consumers' decisions if they have rented e-bicycles instead of e-scooters because they believed it's safer," he says.
According to the researchers, future analyses of crash risk should always include GPS data and precise information about how the vehicles are used. They would also like to see additional comparable data sets from other parts of the world; in particular, data sets that include more e-bicycle journeys in order to improve statistical reliability.
"With more detailed data, we can make better decisions about transport for the future. And to achieve that, it's important that we compare apples with apples," says Marco Dozza.
More about the research
The study only compares e-scooters with e-bicycles, unlike previous studies where e-bicycles and ordinary bicycles were lumped together in the same group. It is also the first study to also include several other important factors in the comparison: ownership, geographical location, usage, and exposure.
- Only rented vehicles were included in the study.
- The locations were limited to highly urbanised city centres using geofencing.
- Usage type was further controlled by comparing e-scooters and e-bicycles from the same rental company.
- Exposure was investigated using three different measures: number, duration, and distance of the trips.
The difference in crash risk between these vehicle types was greatest when trip distance was used as the measure for exposure, when the crash risk was 8.3 times higher for e-bicycles than for e-scooters. But even when using the other two measures for exposure, the crash risk was considerably higher for e-bicycles.
The data in the study comes from GPS data from trips with rented e-scooters and e-bicycles in seven European cities in the years 2022–2023 and includes a total of 686 reported crashes with e-scooters and 35 with e-bicycles. Despite the low number of crashes with e-bicycles, the results of the study are statistically significant when the data from all the cities was weighed together.
The study – Is e-cycling safer than e-scootering? Comparing injury risk across Europe when vehicle-type, location, exposure, usage, and ownership are controlled – was funded by Trafikverket, the Swedish Transport Administration.
Previous press releases about Marco Dozza's research on micromobility:
E-scooter crashes mainly caused by reckless driving
Why your e-scooter may pose higher risks than a rental
How e-scooters can safely operate in a city
