Development and evaluation of an industry safety leadership toolkit

Abstract

Despite decades of improvement workplace safety still has a long way to go (Dekker, 2019). Many thousands of workers continue to be injured across the globe (ILO, 2019). Even in developed countries like Australia, injury rates are stalling, and even increasing in some industries (Safe Work Australia, 2019), and workers’ compensation claims are on the rise (Work Cover Qld, 2019). Consequently, research on ways to improve workplace safety is still very much in demand. Since the late 1970s, organisational culture’s role in contributing, firstly, to large-scale disasters (Turner, 1979), and later, to positive capabilities for successful and safe performance (Weick & Roberts, 1993), prompted the coining of the term ‘safety culture’ (INSAG, 1991).A safety culture is widely considered to be a subset of the broader organisational culture of relevance to safety (Griffin & Curcuruto, 2016), or alternatively, an analytical lens from which to examine and evaluate the broader organisational culture for its effects on safety properties (Guldenmund, 2018). Yet, empirical intervention research on safety culture is still in its infancy (Aburumman et al., 2019). My programme of research aims to contribute to this gap in academic and practical understanding.In this research, I develop an integrative and situationally-based model of safety leadership that contributes to a positive safety climate and culture through the implementation of various ‘control strategies’ (Leverage, Energise, Adapt, and Defend; LEAD). The research programme consists of theoretical development, followed by a series of studies designed to develop and validate a new measure of safety leadership. The final study tests an intervention toolkit designed around the new model, drawing on a multi-organisational setting.The theoretical paper describes how the LEAD model can be used to exert an influence over workers’ safety performance by creating a shared social context regarding the applicability and utility of various ‘bundles’ of safety practices (i.e., control strategies). This model is linked to a number of theories in safety science, including high reliability organising, Rasmussen’s (1997) control model, as well as the extant literatures around safety culture, climate, and leadership.In the second section of the research programme, four studies are undertaken to create, test, and evaluate the psychometric performance of a LEAD scale. Qualitative interviews were undertaken with 25 subject matter experts and an item pool built from coded and thematically-analysed comments. Next, an item pool was subjected to testing and refinement. The LEAD factor structure was established through exploratory factor analysis. Finally, two separate applied studies confirmed the LEAD model’s factor structure and provided evidence of its predictive validity insofar as accounting for variance in safety performance and motivation.The final study involved the development and evaluation of a LEAD toolkit that was designed for industry. The toolkit consisted of separate worker and leader training programs, a senior manager workshop, and a survey tool with results debrief. Six organisational units within a much larger academic institution participated in the intervention research. Training feedback was positive overall, and increased in positivity as the sessions progressed. Comparisons before and after the training, and between training participants and non-participants, showed that the toolkit did influence self-reported safety performance and perceptions of two LEAD dimensions: Adapt and Defend.The thesis concludes with a general discussion, including an extended treatment of directions for future research, and a personal reflection.