Safety-in-numbers: Estimates based on a sample of pedestrian crossings in Norway

Abstract

Safety-in-numbers denotes the tendency for the risk of accident for each road user to decline as the number of road users increases. Safety-in-numbers implies that a doubling of the number of road users will be associated with less than a doubling of the number of accidents. This paper investigates safety-in-numbers in 239 pedestrian crossings in Oslo and its suburbs. Accident prediction models were fitted by means of negative binomial regression. The models indicate a very strong safety-in-numbers effect. In the final model, the coefficients for traffic volume were 0.05 for motor vehicles, 0.07 for pedestrians and 0.12 for cyclists. The coefficient for motor vehicles implies that the number of accidents is almost independent of the number of motor vehicles. The safety-in-numbers effect found in this paper is stronger than reported in any other study dealing with safety-in-numbers. It should be noted that the model explained only 21% of the systematic variation in the number of accidents. It therefore cannot be ruled out that the results are influenced by omitted variable bias. Any such bias would, however, have to be very large to eliminate the safety-in-numbers effect.

Exploring the safety in numbers effect for vulnerable road users on a macroscopic scale

“…That study and many later studies have confirmed the existence of a safety in numbers effect, but one needs to know the mechanisms producing the effect if one aims to exploit it in planning infrastructure to encourage walking or cycling (Bhatia and Wier, 2011). The early studies on Safety in Numbers focused on pedestrians (Geyer, 2006;Bhatia and Wier, 2011)., There are fewer studies focusing on bicyclists (Johnson et al, 2014;Fyhri et al, 2016), and even fewer studies that include both pedestrians and cyclists (Elvik, 2016). More recent studies attempt to demonstrate the safety in numbers effect on a macroscopic scale for potential use in planning and predictions on the zonal level, but again only focus on a single group of road users (Wang and Kockelman, 2014).…”

“…However, as cities grow and develop, robust multimodal systems are required to enable adequate integration of land use and transportation, and provide viable travel options for nondriving populations. This is where the concept of Safety in Numbers emerges, supported by the assumption that more multimodal travel options would lead to more walking and cycling, which would be associated with an increase in crashes, but less than proportional with the increase in walking and cycling (Elvik, 2016).…”

Design Criteria for Improving Safety Performance of Rural Intersections

“…A vertical jump re ects a lack of exibility in the functional form within the model. e Cumulated Squared Residuals plot should not have long increasing or decreasing runs because they correspond to areas of substantial over-and underestimation [28,29]. (iv) e Performance Diagram is a means of visually identifying how close the sample of predicted values is to the observed values.…”

Modelling of road traffic fatalities in India

“…For instance, models with pedestrian injuries as outcome and pedestrian and car volume as explanatory variables. No model in the literature has accounted for more than two road users, except Elvik (2016) who modelled pedestrian injuries using volume of cars, cyclists and pedestrians.…”