Safety-in-numbers: An updated meta-analysis of estimates. (August 2019)

• Record Type: Journal Article
• Authors: Elvik, Rune; Goel, Rahul
• Is Part Of: Accident analysis and prevention. Volume 129(2019)
• Page Start: 136
• Page End: 147
• Publication Date: 2019-08
• Subjects: Safety-in-numbers -- Meta-analysis -- Meta-regression -- Cyclists -- Pedestrians; Accidents -- Prevention -- Periodicals; Accident Prevention -- Periodicals; Accidents -- Prévention -- Périodiques; 363.106
• DOI: 10.1016/j.aap.2019.05.019
• Languages: English
• ISSNs: 0001-4575

Abstract

Highlights: Studies of safety-in-numbers are reviewed. Results are summarised by means of meta-analysis. Summary coefficients are close to 0.40 for all groups of road users. Results of primary studies are very heterogeneous. Abstract: Safety-in-numbers denotes the tendency for the number of accidents to increase less than in proportion to traffic volume. This paper updates a meta-analysis of estimates of safety-in-numbers published in 2017 (Elvik and Bjørnskau, Safety Science, 92, 274–282). Nearly all studies find safety-in-numbers, but the numerical estimates vary considerably. As virtually all studies are cross-sectional, it is not possible to determine if safety-in-numbers represents a causal relationship. Meta-regression analysis was performed to identify factors which may explain the large heterogeneity of estimates of safety-in-numbers. It was found that safety-in-numbers tends to be stronger for pedestrians than for cyclists, and stronger at the macro-level (e.g. citywide) than at the micro-level (e.g. in junctions). Recent studies find a stronger tendency towards safety-in-numbers than older studies.