Investigating the relationship between safety climate and safety performance in the construction of multi-storey buildings in Pakistan

Abstract

The sectoral share of construction in the GDP of Pakistan has consistently increased from 2.42 to 2.58 percent during the last decade; reflecting its momentous contribution in country's economic development. However, construction is continuously suffering from the highest accident rate (14.55% in 2006-2007 to 16.3% in 2014-2015) among all the industries. Most of the construction accidents occur on building projects mainly due to fall from height. Though large construction companies working on multi-storey building projects are following various safety management systems, unsafe behaviors and unsafe conditions are at times noticed, and the employees continue to suffer from fatal accidents mostly due to fall from height. This necessitates exploring the root causes of such accidents as well as identifying the significant safety climate (SC) factors. A focus on these factors would improve the employees' safety behavior which in turn will enhance the overall safety performance (SP) on these projects. The current study, therefore, aims to investigate the relationship between SC and SP in the construction of multi-storey buildings in Pakistan. The specific objectives are to: examine the current safety practices in the construction industry (CI); identify the contributory factors of accident causation on building projects; determine the SC factors that can significantly enhance SP on building projects; examine the causal relationship between SC factors and SP indicators; examine the effects of demographic variables on SC; and finally, recommend strategies to enhance SP in the construction of multi-storey building projects in Pakistan. A mixed-method research strategy comprising concurrent and sequential research methods is adopted in this study. The current safety practices are summarized based on the literature review and interview results. The thirty-two contributory factors of accident causation are shortlisted through a triangulation strategy encompassing eight face-to-face semi-structured interviews with academic and industry experts, coupled with reviewing fifty-eight studies. To prioritize these factors, Delphi survey and inter-rater agreement analysis are conducted among four respondent groups: clients, contractors, safety officials and academic experts. The SC data (N=426), collected through questionnaire survey from forty under-construction multi-storey building projects, are split randomly to achieve two samples. To identify and validate the significant SC factors, exploratory and confirmatory factor analyses are conducted on calibration and validation sub-samples respectively. To investigate the influence of identified SC factors on SP indicators, a SP measurement model is hypothesized and tested. Finally, the effects of various demographic variables on SC are examined. The study has identified three most significant accident causation factors: poor enforcement of safety rules and regulations by the Government agencies; insufficient allocation of safety budget and safety incentives by the client; and insufficient provision of safety training and resources by the contractor. As the developed SC scale of Hong Kong CI could not be cross-validated in Pakistani CI, this study suggested integrating the cultural and regional values while cross-validating an existing scale. Consequently, a 24-item SC scale is developed and validated for building projects in Pakistan. It consists of four factors: management commitment and employees' involvement in health and safety; safety enforcement and promotion (SE&P); applicability of safety rules and safe work practices (SR&WP); and safety consciousness and responsibility. The study discovers SE&P as one of the most significant SC factors influencing the SP, whereas SR&WP is identified as the most overlooked SC factor. In addition, it develops a SP measurement model and verifies the significant positive impact of SC on safety compliance and safety participation, and negative impact on number of self-reported accidents/injuries. Contrary to some of the previous studies, safety compliance has established the strongest positive impact on SP as compared with safety participation and number of self-reported accidents/injuries. While examining the influence of ten demographic variables on employees' safety behavior, the study highlights various employees groups that can be targeted for achieving an enhanced SP level. Lastly, based on the analysis of causes of accidents, demographic variables and SP measurement model, strategies are recommended for enhancing the SP on building projects in Pakistan. These strategies may help the construction stakeholders to prioritize their efforts towards achieving a zero-accident rate. The study adds to the body of knowledge by: ranking the underlying causes of construction accidents as per their significance; highlighting the implications of adopting an existing SC scale across the region and culture; revealing the deviations in SC dimensions such as discovering SE&P as a new SC factor; and developing a SP measurement model for building projects in Pakistan that can be generalized in developing countries sharing similar work environment. In addition, the study introduces the applicability of inter-rater agreement analysis technique for the analysis and validation of Delphi survey data. The study advocates conducting more cross-validation studies to test the applicability of designed SC scale in other developing regions and cultures. As the scope of this study is limited to multi-storey building projects in Pakistan, caution may be exercised while generalizing the research findings to other sectors of CI. Nonetheless, the study's robust methodology can be replicated in other industries and regions for element prioritization and SC investigation.