Assessing safety culture : a special focus on the Ghana oil industry

• Main Author: Tetteh, Lovelace
• Published: University of Nottingham 2012
• Subjects: 363.119622338
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.587833

Incidents that have occurred in some parts of the world have alerted researchers and industries of the need to focus on safety culture. Assessing safety culture provides an "early warning" signal of potential safety system failure. The objective of the research was to examine how the industry manages safety, how safety is perceived by employees working in the oil industry, employees' action in preventing accidents, and a model explaining overall safety culture. Safety culture was assessed, collecting both qualitative and quantitative data, in the Ghana oil industry. The assessment was based on a systems approach to corporate culture. This approach involves method triangulation (distribution of questionnaires, interviews, and observation). Questionnaires were distributed to oil workers at the beginning of a behavioural safety initiative. Prior to the distribution, interviews were conducted with managers in the oil industry and results formed the basis for further studies. Instruments were developed based on literature review and the first study, reflecting the norms of the industry. The results presented findings on the components of safety culture: safety climate, safety management system, safety performance. Principal component analysis revealed a seven factor safety climate structure and a two factor safety management system structure. Analyses of variance showed some differences based on age, years of experience, and occupational group. Finally, to test the relationship between these components and safety performance, a model was proposed and tested. An extended model was developed showing relationship between the three dimensions of safety culture (safety management systems, safety climate, and safety behaviour). Regression analyses demonstrated that employees' attitude, employees' involvement, and safety management system were ii predictive of actual safety behaviour. Also, safety priorities and need for safety and safety climate (attitude to safety and behaviour) predicted accident involvement. In addition, safety management moderated the safety climate and safety performance (observable behaviour and accident involvement) relationship confirming and contradicting findings with the extant safety culture literature. The study supports the use of a safety culture measure as a useful diagnostic tool in ascertaining the way safety is operationalised and to improve the state of safety in the industry.