The "regression-to-mean" effect : some empirical examples concerning accidents at road junctions.

Author(s)
Brüde, U. & Larsson, J.
Year
Abstract

A randomly large number of accidents during a "before-period" is normally followed by a reduced number of accidents during a corresponding "after-period" even if no countermeasures have been implemented. This statistical phenomenon is termed the "regression-to-mean" effect (or briefly the regression effect). Road junctions constitute points in the road network with particularly high accident rates although the average number of accidents per junction is low. The latter means that the regression effect can be expected to appear even in very modest accident numbers. The examples described in this report are based on accidents at unaltered rural junctions in the national major road network. The years 1972-1975 have been regarded as the before-period and 1976-1978 as the after-period. The examples show that the regression effect (accident reduction) in accidents reported to the police often can be about 30-40%. For accidents involving personal injury the regression effect is often about 50-60%. In the case of junctions with a significantly large number of accidents (in relation to the amount of traffic) during the before-period the regression effect is usually even greater. (a) for the covering abstract of the conference see IRRD abstract no 264967.

Request publication

2 + 3 =
Solve this simple math problem and enter the result. E.g. for 1+3, enter 4.

Publication

Library number
B 20979 (In: B 20971) /81 / IRRD 264975
Source

In: Seminar on short-term and area-wide evaluation of safety measures, Amsterdam, April 19-21, 1982, p. 47-54, 7 tab., 4 ref.

Our collection

This publication is one of our other publications, and part of our extensive collection of road safety literature, that also includes the SWOV publications.