Work accidents are a problem for the continuity of the company, unexpected and uncontrolled events not only result in material losses, more than work accidents can cause fatalities. The accident cost iceberg theory shows that the ratio between visible and invisible parts of the iceberg phenomenon is 1 in 5 to 1 in 50, and the cost of property damage is 5 to 50 times greater than the cost of compensation and treatment for work-related injuries.

The direct causes of work accidents are unsafe acts (unsafe acts) and unsafe conditions (unsafe conditions). It was further explained that 85-96% of the causes of work accidents were contributions by worker behavior that led to unsafe actions [1]. Henrich estimates that 85% of accidents are the result of the contribution of unsafe worker behavior. Data from the survey on the causes of accidents in Japan noted that 92% of work accidents were caused by unsafe behavior and 8% due to unsafe conditions [2].

Several approaches were taken to reduce or prevent injury due to accidents, based on the results of the comparison of approaches to behavior that achieved the most effective results for reducing injuries in the workplace, which was 59.6%, followed by the ergonomics approach at 29% and the engineering control approach at 11.4% [3]. The same thing was stated by Geller that efforts to prevent work accidents would be better using a behavioral approach by increasing safety behavior (safety performance) [4].

The concept of safety performance in the safety literature refers to organizational safety performance and worker safety behavior. Organizational safety performance uses quantitative measures that refer to the number of accidents, the number of injuries, and near misses in a certain period [5]. In contrast to organizational safety performance, safety behavior uses qualitative measures to measure safety performance [6]. This study limits the concept of safety performance to worker safety behavior with safety behavior indicators, namely safety compliance and safety participation. Safety Compliance is the behavior of workers to realize and maintain safety, safety participation is the behavior of workers' participation in safety activities and activities [7].

Organizational factors and a supportive work environment can affect safety performance [8]. Support from the social environment can be accepted by workers from colleagues and superiors.8 Social support can be provided in the form of emotional support, appreciation support, instrumental support, information support, and social network support [9]. A person's behavior can be determined by two factors, namely environmental factors and individual factors [10]. Individual factors according to individual characteristics include age, gender, years of service, and education level [11]. In addition to individual characteristics, one of the determinants of behavior that comes from the individual is motivation according to the concept of Self Determination Theory (SDT) [12]. Another determinant of behavior is communication, the effectiveness of communication is very important to maintain a positive culture and involve workers in safety activities [13]. Safety communication does not only affect safety performance but also has an effective impact on organizational safety performance [14].

Various factors can affect safety performance, one of which is commitment. Safety Commitment can improve safety performance. Ghasemi's research states that organizational factors and a supportive work environment can affect the safety performance of workers. Safety commitment is another factor that can affect safety performance, but so far it has received less attention [8]. Azkha's research shows that the commitment factor is related to safety performance [15]. Workers who have a high safety commitment can identify hazards in the workplace and are active in safety activities [16]. Workers' safety commitment can be reflected in compliance with work regulations or procedures and the willingness of workers to be involved in safety activities [17].

The concept of commitment consists of three components, namely affective commitment, normative commitment, and continual commitment [18]. Adjustment of the components of commitment to safety was developed into an affective commitment to safety, normative commitment to safety, and continuance commitment to safety [19]. This study aims to analyze the factors that influence safety performance.

This research is an observational study with quantitative methods and a cross-sectional design. The study was conducted from December 2019 to January 2020. The population in this study was 395 workers in one of the manufacturing industries in East Java, the sample size obtained by simple random sampling technique was 78 workers.

There are two types of variables in this study, namely the independent variable and the dependent variable. The independent variables include worker characteristics (age, gender, years of service, and education level), safety motivation, management safety commitment, safety communication, social support, and safety commitment. The dependent variable is safety performance. The data collection instrument for the dependent variable used an observation sheet, while the independent variable used a questionnaire with a Likert scale modified from previous research and relevant theoretical indicators. Each questionnaire has favorable and unfavorable statement items with a Likert scale range of 1 to 4. Variable categorization uses the interval formula (I = R/K), where R is the range (maximum value - minimum value), and K is a category.

This research has been declared ethically feasible based on the ethical feasibility certificate number 751/HRECC.FODM/XI/2019.

Univariate Analysis

Table 1 shows the results of the cross-tabulation between age and safety performance, where most (64.0%) workers aged 41-51 years have less safety performance. Most of the workers are female, the cross-tabulation results show that 22 workers (55.8%) have less safety performance. Most of the workers (65.6%) have less safety performance. Sufficient experience has not been able to improve safety performance. Most workers with a high school education level (67.8%) have less safety performance.

Table 2 shows that most workers have high safety motivation, whereas most of the workers with high safety motivation (56.8%) have less safety performance. This means that the encouragement to behave safely is not enough to improve the safety performance of workers.

Table 3 shows that most (75.6%) workers who assess management safety commitment are low and have less safety performance. The management's low commitment to the implementation of OHS is seen as a form of the company's lack of seriousness in ensuring the safety and health rights of workers so that it can give a bad perception and ultimately reduce the safety performance of workers.

Table 1. Distribution of Safety Performance Based on the Characteristics of Individual Workers in the Manufacturing Industry

Table 2. Distribution of Safety Performance Based on the Safety Motivation of Workers in the Manufacturing Industry

Table 3. Distribution of Safety Performance Based on Management Safety Commitment in Manufacturing Industry

Table 4. Distribution of Safety Performance Based on the Safety Communication in the Manufacturing Industry

Table 5. Distribution of Safety Performance Based on Social Support of Workers in the Manufacturing Industry

Table 6. Distribution of Safety Performance Based on the Safety Commitment of Workers in the Manufacturing Industry

Table 7. Candidate Variables for Binary Logistics Regression Test

Table 8. Binary Logistics Regression Test Results Effect 

Table 4 shows that the majority (67.4%) of workers who assess the company's safety communication as lacking have less safety performance. Safety communication carried out by the company has not been able to improve the safety performance of workers.

Table 5 shows that most (75.7%) workers who get low social support have low safety performance. This can be an indication that social support related to occupational health and safety need to be managed properly to improve safety performance.

Table 6 shows that most (59.3%) workers who have an effective commitment to work safety have good safety performance. This means that workers who are committed to safety based on an emotional bond with the company's leadership, have a congruence of safety goals with the company, and feel involved in the development of OHS are more likely to have good safety performance.

Bivariate Analysis

In this study, bivariate analysis was conducted to determine the candidate variables for the binary logistic regression test, namely variables with a p-value <0.25. Bivariate analysis using the chi-square test with the following results:

Based on table 7, it can be seen that the candidate variables for the binary logistic regression test are education level, management safety commitment, social support, and safety commitment. The candidate variables were then jointly tested using binary logistic regression analysis to see which variables had the most influence on safety performance.

Multivariate Analysis

Table 8 shows that social support and safety commitment has a significant effect on safety performance. The value of B shows a positive sign, meaning that workers who get high social support can improve safety performance. However, the B value on safety commitment shows a negative value, meaning that the more workers have normative and continuance commitment to safety, the lower safety performance will be.

The Influence of Individual Characteristics on Safety Performance

Age: The results showed that statistically, age had no significant effect on safety performance. In this study, it was found that most workers in the 41–51-year age category had less safety performance. This is in line with the results of Mahardhika's research that there are workers in the old age category who behave disobediently and do not obey safety rules [20]. This means that the category of old age and young age both have the potential to have less safety performance. The influence of the age factor on safety performance is of course influenced by other things, such as knowledge. Based on the results of the study at the research site, researchers found that there were still many workers who did not understand the purpose and benefits of work safety. This condition can cause age to have no significant effect on safety performance because both young workers and old workers both have less knowledge of safety behavior (safety performance).

Gender

According to Saleh, Gender affects safety performance because men and women are gifted with different strengths and physical strengths [21]. The influence of gender on safety performance is also explained by Gyekye and Salminen that female workers have more positive perceptions of safety, are more obedient to safety rules, and have a low accident rate [22]. The results of this study also show that gender has no significant effect on safety performance. One of the factors that cause gender does not have a significant effect on safety performance, namely, most the male workers have less safety performance, and most the female workers have less safety performance. This means that gender does not guarantee good safety performance. Based on the results of the study at the research site, there is no influence of gender on safety performance, it can be caused by the absence of differences in work safety regulations for male and female workers so that both male workers and female workers have the same opportunity to obey or violate the rules. The results of this study are in line with Mujiadi's research which concluded that gender does not affect workers' unsafe behavior [23].

Work Period 

The working period has a role in changing the safety behavior of workers. This was conveyed by Saleh who revealed that the length of work or the length of time a person pursues the work he does is often associated with a high level of safety performance because he has known experience and adaptation for a long time [21]. The results of this study also show that the working period has no significant effect on safety performance. This study is in line with research by Suryanto and Widajati, who concluded that tenure is not associated with the unsafe behavior of workers [24]. However, this is not in line with Apriluana's research which concluded that tenure has a significant effect on the behavior of using PPE. It was further explained that workers with long tenure often received OSH socialization so that they could increase workers knowledge [25]. Based on the results of studies at the company, OHS socialization is rarely carried out, so there is no difference in OHS knowledge among workers with old and new tenures.

Education Level 

The results showed that statistically, the level of education had no significant effect on safety performance. This is following the results of Feng's research which found there was no strong correlation between competence and education with safety performance [26]. The results of the study in the company where the research was conducted, there was no significant effect of the level of education on safety performance, it could be caused by the involvement of other factors, namely safety training. Based on information from supervisors and workers, safety training is still rarely carried out. So far, training is still limited to fire protection and safety riding, not covering things that affect safety behavior such as safety awareness training, which aims to increase worker awareness of safety, and so that workers can recognize potential hazards in the work environment. Inadequate and unscheduled training implementation affects workers' awareness of safety behavior.

The Effect of Safety Motivation on Safety Performance

The results of this study indicate that most workers have high safety motivation. However, statistically, safety motivation has no significant effect on safety performance. The results of this study are in line with research conducted by Retnani and Ardyanto who concluded that employee motivation has no significant effect on unsafe behavior [27]. The results of this study indicate that most workers have high safety motivation. However, statistically, safety motivation has no significant effect on safety performance. The results of this study are in line with research conducted by Retnani and Ardyanto who concluded that employee motivation has no significant effect on unsafe behavior [28]. Supervision carried out by supervisors and shift heads so far has focused more on manufactured goods and the production process. Supervision has not been fully carried out on aspects of work safety, for example, the use of PPE and the level of safety of workers in doing work.

The Effect of Management Safety Commitment on Safety Performance

In this study, statistically, management safety commitment had no significant effect on worker safety behavior. The results of this study are in line with Saridewi's research which concluded that management's commitment to work safety had no significant effect on the safety behavior (safety performance) of nurses in the Emergency Room (IGD) RSUD Dr. Soetomo Surabaya [29]. 

Based on the results of the study at the research site, there is no influence of management safety commitment on safety performance because the company's management commitment to the implementation of OHS is still low. This is supported by the supervisor's statement at the research site who explained that based on his assessment, the company's management had not given serious attention to OHS. It was further explained that the absence of a special OHS department in the company is proof that the company has not shown a commitment to the implementation of OHS.

This condition must be immediately controlled by the company's management because the lack of management control can lead to substandard practices and conditions which are the cause of work accidents. Companies need to show real commitment to the implementation of OHS. The steps that must be taken by the company's management are to place top management as the head of P2K3 so that they have full authority in making decisions. The position of chairman of P2K3 must be occupied by the person who has the fullest authority in the company, in this case, the president and director of the company. Companies also need to implement OHSM so that all activities are integrated with OHS aspects.

The Effect of Safety Communication on Safety Performance

Based on statistical tests, shows that safety communication has no significant effect on the safety performance of workers. The results of this study are in line with research conducted by Mashi which concluded that safety communication had no significant effect on worker safety compliance [30]. Based on the results of studies at the research site, safety communication does not have a significant effect on safety performance, it can be caused by the implementation of safety communication from management is still lacking. Several things that indicate that safety communication is still lacking include, OHS signs that have not been installed in every work unit, requests for PPE from workers have not been responded to properly, and workers are rarely involved in discussions related to work safety. In addition, the form of safety communication that can cause a direct influence on safety behavior has not been implemented at the research site.

The company has made every effort to conduct a briefing before starting work on a shift change, however, the matters presented in the briefing are still limited to production purposes. In addition, management also needs to conduct safety talks as a routine effort to continue to provide information and remind workers regarding potential hazards that can occur in every work process. Companies to implement OHS will not be optimal if they have not implemented OHSM, the key to integrating all company systems with the OHS system is to implement OHSM [31]. 

The Effect of Social Support on Safety Performance

In this study, statistically, social support had a significant effect on the safety performance of workers. The results of this study are following research conducted by Anditya who concluded that all social support factors have a significant effect on safety behavior [32]. Based on the results of studies at the research site, social support has a significant effect on safety performance, which can be caused by the establishment of good and harmonious relationships between fellow workers. This good relationship can be a medium between workers to remind each other, foster empathy, and care for each other's safety for co-workers. If you see a coworker not complying with safety rules, other workers can easily remind and advise them to comply with safety rules. This condition must be utilized properly by the company's management, to participate in facilitating good social support between fellow workers. For example, implementing the “you see you act” program, this program is in the form of a quick response by workers when they see a potential hazard that has not been identified or when a worker sees a co-worker violating safety rules by taking immediate action. Actions were taken, such as reminding, reprimanding, handling dangerous conditions, and reporting those dangerous conditions if they cannot be handled by themselves.

The Effect of Safety Commitment on Safety Performance

In this study, statistically, safety commitment has a significant effect on safety performance. The results showed that workers were more likely to have a continuance commitment to safety. However, most workers who have a continuance commitment to safety have less safety performance. This means that theoretically, workers can understand the potential losses experienced when they do not comply and do not participate in workplace safety. However, this is not enough to improve the safety performance of workers, management's role is needed to involve workers in discussions related to safety issues and convey safety objectives intensely to workers. It aims to make workers feel they have the same OHS goals as the company.

Based on the results of the study at the research site, safety commitment has a significant effect on safety performance because workers who have participated in firefighting socialization and Job Safety Analysis (JSA) socialization feel that OHS is an important part of the worker safety and production safety. This makes workers more likely to strive for safe behavior based on the compatibility of their safety goals with the organization. Based on the research results, workers who have good safety performance are workers who have an affective commitment to safety.

Based on the results of the study at the research site, safety commitment has a significant effect on safety performance because workers who have participated in firefighting socialization and Job Safety Analysis (JSA) socialization feel that OHS is an important part of the worker safety and production safety. This makes workers more likely to strive for safe behavior based on the compatibility of their safety goals with the organization. Based on the research results, workers who have good safety performance are workers who have an affective commitment to safety.

1. Bird, F.E.J. and G.L. Germain. Practical Loss Control Leadership. International Loss Control Institute, Inc., 1992.

2. Alvan, A. Evaluasi Penerapan ABG (Aktif Berbagi) Sistem Sebagai Upaya Mengurangi Angka Kecelakaan Kerja di PT. Coca Cola Amatil Indonesia Central Java. Skripsi, Ilmu Kesehatan Masyarakat, Universitas Negeri Semarang, 2015.

3. Guastello, S. "Do we really know how well our occupational accident prevention programs work?" Safety Science, vol. 16, no. 3–4, 1993, pp. 445–463.

4. Galler, E. The Psychology of Safety Handbook. Lewis Publishers, 2001.

5. Nadim, E.A. et al. "Investigating the relationships between safety climate and safety performance indicators in retrofitting works." Construction Economics and Building, vol. 18, no. 2, 2018, pp. 110–129.

6. Hon, C.K.H. et al. "Relationships between safety climate and safety performance of building repair, maintenance, minor alteration, and addition (rmaa) works." Safety Science, vol. 65, 2014, pp. 10–19.

7. Neal, A. and M.A. Griffin. "A study of the lagged relationships among safety climate, safety motivation, safety behavior, and accidents at the individual and group levels." Journal of Applied Psychology, vol. 91, no. 4, 2006, pp. 946–953.

8. Ghasemi, F. et al. "The Role of personal commitment to safety in shaping safety performance of front-line employees: a case study in small manufacturing industries." Journal of Ergonomics, vol. 6, no. 2, 2018, pp. 16–23.

9. Sarafino, E. and T.W. Smith. Health Psychology: Biopsychosocial Interactions. 9th ed., John Wiley & Sons, Inc., 2017.

10. Suhariadi, F. "Productivity as a form of behavior: an alternative effort toward psychological measurement." Australian Journal of Psychology, vol. 57, 2005, pp. 148–158.

11. Christian, M.S. et al. "Workplace safety: a meta-analysis of the roles of person and situation factors." Journal of Applied Psychology, vol. 94, no. 5, 2009, pp. 1103–1127. doi:10.1037/a0016172.

12. Ryan, R.M. and E.L. Deci. "Overview of self-determination theory: an organismic-dialectical perspective." Handbook of Self-Determination Research, edited by E.L. Deci and R.M. Ryan, University of Rochester Press, 2002, pp. 3–33.

13. Rashid, R.A. et al. "Impact of safety communication on safety commitment with leader-member exchange quality as a moderating factor: a conceptual framework." Global Business and Management Research: An International Journal, vol. 6, no. 4, 2014, pp. 277–282.

14. Chen, W.T. et al. "The Impact of safety culture on safety performance: a case study of taiwan’s construction industry." International Journal of Organizational Innovation, vol. 11, no. 1, 2018, pp. 1–15.

15. Azkha, L.H.A.N. "Faktor-faktor yang berhubungan dengan kecelakaan kerja pada karyawan pt. kunanggo jantan kota padang tahun 2016." Proceedings of the National Seminar IKAKESMADA “Peran Tenaga Kesehat dalam Pelaks SDGs”, 2017, pp. 235–241.