Citation

Lee T, Harrison K. Safety Sci. 2000; 34(1-3): 61-97.

Copyright

(Copyright © 2000, Elsevier Publishing)

DOI

unavailable

PMID

unavailable

Abstract

Definitions of safety culture abound, but they variously refer to the safety-related values, attitudes, beliefs, risk perceptions and behaviours of all employees. This assembly may seem too inclusive to be meaningful, but each represents a different level of processing and the choice for measurement (or intervention) is more pragmatic than theoretical. The present study addresses mainly attitudes, but also reported behaviours. This is done using a 120-item questionnaire covering eight domains of safety in three nuclear power stations. Principal components analysis yields 28 factors -- all but four of which are correlated with one or more of nine criteria of accident history. Differences by gender, age, shifts/days and work areas are revealed, but these are confounded by type of job and ANOVAS are applied to clarify the main sources of variation. The effects on safety culture of a number of organisational components are also explored. For example, the role of safety in team briefings, management style, work pressure versus safety, etc. It is concluded that personnel safety surveys can usefully be applied to deliver a multi-perspective, comprehensive and economical assessment of the current state of a safety culture and also to explore the dynamic inter-relationships of its 'working partsandrsquo.