Citation

Barnett A. Transp. Res. Rec. 1993; 1423: 52-61.

Copyright

(Copyright © 1993, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

unavailable

PMID

unavailable

Abstract

Because fatal air crashes are rare, it is often asserted that data about them cannot yield reliable inferences about patterns in air safety. That assertion may overstate, however, both the limitations of small data samples and the sensitivity of analytic outcomes to changes in starting assumptions. The issue is explored by discussing criticisms by a TRB panel and other commentators of two recent air-safety papers cowritten by this author. The goal is not to suggest that monitoring aviation safety can be reduced to studying fatal crashes, but to avoid an unnatural deemphasis on such crashes because of statistical misunderstandings. Beyond aviation, this may interest the broader group of transportation researchers who work with small data samples.

Record URL:
http://onlinepubs.trb.org/Onlinepubs/trr/1993/1423/1423-008.pdf

Language: en

Keywords

Accident prevention; Accidents; Monitoring; Sensitivity analysis; Crashworthiness; Sampling; Protection; Airports; Statistical methods; Aviation; Air transportation