Journal Screenshot

International Journal of Academic Research in Business and Social Sciences

Open Access Journal

ISSN: 2222-6990

Evaluating Measurements of Workers’ Demographic and Psychometric Profile: A Prelude to Sustainable Development of Behavioural Safety in Sarawak’s Oil & Gas Construction Industry

Hon-Kuan Lai, Nor Mariah Adam, Mohamad Ibrani Shahrimin Adam Assim, Nurul Hidayu Mat Jusoh, Omar Faruqi Marzuki

http://dx.doi.org/10.6007/IJARBSS/v13-i15/18796

Open access

Significant accidents in Sarawak’s oil and gas construction sites often involve fatality, presumably behavioural related, attracting substantial unwarranted attention from the press. Therefore, it is imperative to develop a measuring instrument to identify and gauge workers’ demographic and psychometric attributes that may significantly influence workers’ safe behaviour potentials to reduce the risk of occupational accident hazards in a sustainable manner. Thus, this paper evaluates constructs’ validity and reliability, and statements’ communality of this new instrument to confirm its objectivity and clarity in measuring constructs as intended. A 5-point Likert-type scaled instrument consists of 93 initial statements was created and utilized to evaluate 51 oil and gas construction workers, randomly selected from local-based oil and gas construction service providers. Demographic results indicate most respondents have more than 5 years work experience, with a high level of experiences in occupational accidents, and almost all respondents possessed some kinds of knowledge and experience in accident prevention to different extents. The analysis also found that most demographic and psychometrics statements of initial research framework yielded a high reliability (Cronbach’s Alpha > 0.7). High validity (KMO <0.5, p = 0.000) in all constructs of personal attributes, perception of safety climate, and safe behaviour potentials are observed. Eventually, a total of 20 statements had been removed, while retaining 73 items mainly due to their high communality (Communality > 0.6) except items BPV3 and LSM5 with valid dispensations. High communality value indicates that the extracted constructs represent the variables well. Through confirming the validity and reliability of initial research framework, and removal of statements with lower communality, a statistically valid and reliable multi-lingual questionnaires is therefore synthesized for subsequent research.