Title

Correcting for Selectivity Bias in the Estimation of Road Crash Costs

Authors

Margaret Giles

Document Type

Journal Article

Faculty

Faculty of Business and Law

School

School of Accounting, Finance and Business Economics

RAS ID

9041

Comments

Giles, M. J. (2003). Correcting for selectivity bias in the estimation of road crash costs. Applied Economics, 35(11), 1291-1301.

Abstract

Police road crash data comprise a non-random sample of the true population of road crashes, the bias being due to the existence of crashes that are not notified to the Police. Heckman viewed similar problems as ‘omitted variables’ problems in that the exclusion of some observations in a systematic manner (so-called selectivity bias) has inadvertently introduced the need for an additional regressor in least squares procedures. In the case of Police road crash data, selectivity bias arises from factors affecting the notification of crashes to the Police, such as the number of vehicles in the crash and the type and location of the crash. Using Heckman's methodology for correcting for this selectivity bias, Police road crash data for Western Australia are reconciled with total road crash data in the estimation of the property damage costs of road crashes.

DOI

10.1080/0003684032000090717

 
COinS
 

Link to publisher version (DOI)

10.1080/0003684032000090717