SAFETY LEADERSHIP, SAFETY CLIMATE & SAFETY PERFORMANCE WITHIN TNB'S TECHNICAL WORKFORCE

Occupational health and safety are relevant to all branches of industry, business and commerce companies, and has been continued to be one of the most critical but highly criticized issues within the discipline of human resource management. This study aims to identify the influences of safety leadership towards safety performance mediate by safety climate within TNB's technical workforce. PLS-SEM is use to test relationships among variables, a series of statistical methods including mean, descriptive analysis, and structural model is conducted. This study confirmed that there is a significant influence between each study of variable. Furthermore, the study also shows that safety climate has a mediating effect on the relationship between safety leadership and safety performance. This study may influence and gives an implication towards an organization about the safety practices that apply in the organization. Considering the significance of all three variables, future studies can be carried out to focus on other industry sector.


Introduction
Occupational safety and health management can be seen as one of the most important elements and should be emphasized especially in activities involving high-risk jobs either in manufacturing, construction or service sectors such as electrical utilities. The Occupational Health and Safety Act, [Akta Kesihatan dan Keselamatan Pekerjaan (AKKP)] is responsible for establishing a safety culture within the established organization. AKKP's philosophy is to emphasize the concept of self-regulation, the employer is given a complete mandate to ensure safety at work is guaranteed (Rahmathunisa Beevi, 2013). In a broader context, AKKP promotes the existence of an atmosphere or a working environment that is more in line with the physiology and phycology of the worker. AKKP 1994 acts towards the establishment of systematic occupational safety and health management. However, a comprehensive act will not be successful without support from both employers and employees. The culture of safety practiced by an organization is the main criterion to avoid the occurrence of job disruption as well as the illness caused by the work. If these elements are look without enthusiasm, accidents and occupational injuries is likely will be increased. Previous studies have shown that increased workplace safety performance is greatly influenced by a positive safety culture (Weinberg, 2002). Workplaces health and safety is mandatory for every employer in order to ensure that the health, safety and welfare requirements each of the employees are met. Attention to the health and safety of employees is crucial in order to improve the productivity of employees, as it underlines the efficiency of the organization. Therefore, employers need to be conscious of their responsibilities towards their employees in order to attain world-class efficiency in terms of health and safety. Hence, this study is conducted with the objective to identify the influences of safety leadership towards safety performance mediate by safety climate within TNB's technical workforce. Besides, the results of this study will also lead to a number of suggestions and remedial measures that can be implemented by the management in dealing with or at least reduce the rate of accident and health problems arising from the workplace.

Problem Statement
In today's modern world, manufacturing sector in Malaysia provide a major commitment to the economy of Malaysia. The manufacturing sector experienced moderate improvement consistently and has contributed 7.3 % to the Malaysian economy in the second quarter of 2014 Bank Negara Malaysia (BNM).
However, safety in the workplace is something that affects manufacturing sectors nowadays. By joining a positive safety culture in the workplace, the occupational safety and health management system can be effective. Many organizations have failed to show improved effectiveness when they have introduced new occupational health and safety management strategies because these strategies did not consider the impact of the organizational culture. Throughout the years, a lot of attention has concentrated on the reason of occupational accidents (Kang et al., 2017). Even though manufacturing industry is crucial for Malaysian economic growth, however, given the evidence it can't be denied that working environment in manufacturing industry is more dangerous and hazardous compared to another industries. Therefore, it seems that it is highly necessary to manage workplace safety with a focus on human approach in order to avoid this issue from getting worse. https://doi.org/10.15405/epsbs.2020.12.05.88 Corresponding Author: Zulhafiza Zulkefli Selection and peer-review under responsibility of the Organizing Committee of the conference eISSN:  812 Besides that, although significant research on climate safety problems has been carried out in many nations, however, according to the author's understanding, there are very few safety climate studies have been performed in Malaysia, and to be more specific, there is almost none of the study that have been performed on the manufacturing sector. It can't be denied that the manufacturing industry is one of the most crucial industry where it has enhance the local economic development in Malaysia even though with the highest rate of accidents at the workplace. Therefore, it would appear that further research is required to examine the extent to which long-term effects of safety training could contribute to affect the climate of safety and then serve as a mechanism for facilitating and promoting a good safety culture within an organization.

Research Questions
The research question for this study:

Safety Climate
Safety climate refers to employees' shared perceptions of safety, policies, procedures, practices, as well as the overall importance and true priority of safety within the organization (Jiang et al., 2010;Larsson et al., 2008;Zohar & Luria, 2005). The importance of safety climate and its relationship with occupational safety has been established across a range of industry settings (Clarke, 2006;Eid et al., 2012;Zohar, 2010).
Safety climate can be seen as an aspect of the safety culture or as an indication of the real implementation of the safety culture (Lee & Harrison, 2000).

Safety Leadership
Safety leadership is defined as a process of interaction between leaders and followers, through which leaders can exert their influence on followers to achieve organizational safety goals (Jimmieson et al., 2016). According to Hoffmeister et al. (2014), such leadership refers to the way in which supervisor's influence and promotes safety to their followers at the workplace. Meanwhile, safety leader refers to leaders who engage in safety-related matter with enthusiasm and inspiration and fully concentrate on supervising their followers (Conchie, 2013). In recent years, the important role of safety leadership in the field of occupational health and safety is increasingly gaining acceptance.

Safety Performance
Generally, safety performance means the outcomes of safe working records over a period of time, so that if there are many reported incidences in a period of time, therefore the safety performance can be considered as poor (Atak & Kingma, 2011;Fogarty & Shaw, 2010). There are many factors of safety performance, and one of them including safety culture. Safety performance is a result that comes from summarisation of accidents and incidence that occur from organization that operates in a certain industry (Bellamy et al., 2008). Safety performance can be described as a self-reported rate of accident and occupational injuries. Smith et al. (2006) have done their researched safety in many workplaces, including the manufacturing, construction, service and transportation industries. They defined safety performance as employee safety control and self-reported occupational injury (Shang & Lu, 2009).
Based on the literature above, the following hypothesis is proposed: H1: There is a significant effect of safety leadership on safety performance H2: There is a significant effect of safety leadership on safety climate H3: There is a significant effect of safety climate on safety leadership H4: Safety climate mediate the relationship between safety leadership and safety performance

Research Methods
This quantitative study conducted with a cross sectional approach. A stratified random sampling method is employed in order to collect the sample for this study. For this study, the population involves

Demographic
The participants of this research had different background with regard to gender, age, race, marital status, education, working hours and distribution of all these variables were deliberate using descriptive statistics such as frequency and percentage. Results from the analysis of respondent's gender, majority is male with (73%). As for age the respondents who aged between 31 to 40 years had the highest frequency (57.1%) followed by respondents aged between 20 to 30 years (25.4%). For race, it is divided to four categories which is Malay, Chinese, Indian and other, where the highest number of respondents were Malay with frequency of (88.9%). For marital status, the highest percentage belonged to respondents who were married (65.1%) and 34.9% were single. Regarding their education, 54% had a degree followed by https://doi.org/10.15405/epsbs.2020.12.05.88 Corresponding Author: Zulhafiza Zulkefli Selection and peer-review under responsibility of the Organizing Committee of the conference eISSN: 2357-1330 814 respondents with diploma (28.6%). The majority of respondents work in normal working hours (69.8%).
Lastly, from total of 61 respondents, (98.4%) does not involve in accident for the past 12 months

Measurement Model Analysis
To evaluate the measuring model, the reliability and validity of the latent variables must be tested to complete the structural model examination. The reliability of the measures in this study is assessed using internal consistency analysis and indicators. Convergent validity and discriminatory validity analyzes of the measurement model for this study are evaluated using the procedures recommended by (Chin, 2010).
For reliability, internal consistency reliability was the first criteria to be assessed. The measurement model has sufficient internal consistency reliability when each variable 's composite reliability (CR) surpasses the threshold value of 0.6. Table 01 shows that each variable 's CR for this study is above the recommended threshold value of 0.6. The results show, therefore, that the elements used to represent the variables have satisfactory internal consistency reliability.
Next is validity, it is done by analyzing both convergent and discriminant validity. Convergent validity can be evaluated via the average variance extracted (AVE) (Hair et al., 2010). The result in Table   01 shows that the outer loadings of all items for all variables in initial and modified measurement model according to these results all outer loadings, except three items related to safety leadership (SCH6, SCR5 and SCO7), seven items related to safety climate (MP6, MP14, SP1, SP3, SP4, SP5, and SP6) and two item related to safety performance (SR6 and SM4) were deleted from initial measurement model due to low loading factor which were less than 0.5 which confirmed their low contribution to related constructs. From Table 2 the result shows that all variable AVE is range from 0.391 to 0.558, some of the AVE result are less than 0.5, however, all the value of AVE is accepted as according to Fornell and Larcker (1981) if AVE is less than 0.5, but composite reliability is higher than 0.6, the convergent validity of the construct is still acceptable. Thus, the results proved that convergent validity exist for the constructs of this study.
Discriminant validity in this study was assessed through Heterotrait-Monotrait (HTMT) ratio of correlations (Henseler et al., 2015). Henseler et al. (2015) have suggested the assessment of the HTMT correlations to examine the discriminant validity. This recent approach shows the estimation of the true correlation between two latent variables. A threshold value of 0.90 has been suggested for HTMT (Henseler et al., 2015). Above 0.90 shows a lack of discriminant validity (refer Table 02). This study therefore concludes that the measurement model has established its discriminant validity.

Structural Measurement Model Analysis
The structural model has been conducted as to validate the proposed hypothesis. The bootstrapping procedure "estimates the standard errors of the parameter estimates, calculates the ratio of a parameter estimate to its standard error, and compares this statistic to the t distribution to obtain the p-value" (Rönkkö & Evermann, 2013, p. 15). The results of boot strapping method have been shown in Table 04, where it demonstrates a p-value for each path. All structural model relationships were significant considering a pvalue <0.05.
In the model, the first hypotheses (H1) had a significant a positive coefficient. According to the results the effect of safety leadership on safety performance is significant (β=-0.349, t=2.661**, p<0.001).
These results receive support from several authors (Barling et al., 2002;Griffin & Hu, 2013;Neal & Griffin, 2006;O'Dea & Flin, 2001;Zohar, 2002b). It is known that safety performance in this study is adopted with the purpose to measure workplace safety. Thus, it can be concluded that the higher level of safety leadership will ensure the lower level of workplace accidents.
As for hypotheses 2 (H2), according to the results the effect of safety leadership on safety climate is also significant (β =0.603, t=5.744**, p<0.001). It is supported by previous studies, which indicates there is a significant relationship between both variable (Barling et al., 2002;Hofmann et al., 2003;Komaki et al., 1982;Zohar, 2002a).
The last hypotheses in this model (H3) which is the effect of safety climate on safety performance had a significant effect (β =0.479, t=3.104**, p <0.001). It is similar with the findings by (Brewer, 2006;Clarke, Sloane, & Aiken, 2002;Flin, Mearns, O'Connor, & Bryden, 2000;Hofmann & Mark, 2006;Savkur et al., 2004;Stone & Gershon, 2006;Zohar, 1980). A positive safety climate has direct and indirect relationship with safety performance of the workers by giving knowledge and by motivating the workers to perform their work in safe manner.
R 2 represents the amount of variance in the dependent variable, in this thesis strategic decision making, that is explained by the model. According to Hair et al. (2017), 2 values of 0.75, 0.50, or 0.25 for dependent constructs are considered substantial, moderate, and weak, respectively. As shown in Table 03 below 2 value for safety performance is 0.539, which can be considered moderate; it indicates that 54% of the variance in the safety performance is explained by safety leadership. The results of this study confirm that there is a mediating influence of safety climate on safety leadership and safety performance (Neal et al., 2000;Probst & Estrada, 2010;Wu et al., 2008). Hypothesis 4 is supported by a positive coefficient path (β = 0.289), which was significant at 0.01 (t = 2.864**). Past studies have also studied safety climate as a mediating variable effects (Cooper & Phillips, 2004;Meliá et al., 2008;Nielsen et al., 2008). A positive safety climate will make contribution to the individual safety awareness and behaviour and then strengthen the relationship between them. As a result, it can be concluded that the safety management should not only concentrate on the individual behaviour and awareness but also take safety climate into consideration.

Conclusion
In conclusion, this research study has succeeded in answering all of the research objectives, which aimed at analysing the influences of safety leadership on safety performance and how safety climates mediate safety leadership and safety performance. This study is focused on investigation of the influences between safety leadership and safety performance that are focusing in TNB based manufacturing companies chosen from Federation of Malaysian Manufacturers (FMM). The finding enhances researcher's understanding of the variables which are safety leadership, safety performance and safety climate. Result from this study may lead to understanding of the relationship all the independent variables and a dependent variable. This study also may influence and gives an implication towards an organization about the safety