Work safety climate. Comparison of selected occupational groups

RESEARCH ARTICLE Work safety climate. Comparison of selected occupational groups Marta Stasiła-SieradzkaID1☯*, Agata Chudzicka-CzupałaID2☯, Marta Znajmiecka-Sikora3☯ 1 Institute of Psychology, Faculty of Social Sciences, University of Silesia in Katowice, Katowice, Poland, 2 Department of Social and Organizational Behavior Psychology, Katowice Faculty of Psychology, SWPS University of Social Sciences and Humanities, Katowice, Poland, 3 Institute of Psychology, University of Lodz, Ło´dz, Poland ☯ These authors contributed equally to this work. * marta.stasila-sieradzka@us.edu.pl Abstract Implementation of effective programs to improve occupational safety should be linked to an understanding of the specific nature of the given job. The aim of the research was to com- pare occupational groups with different job-related specificities: industrial production line workers, retail workers and mine rescuers, in terms of their assessment of the work safety climate. The survey covered 2,995 respondents with diversified demographic characteris- tics. The study used an abridged version of the Safety Climate Questionnaire by Znaj- miecka-Sikora (2019) to assess 10 separate safety climate dimensions. The results of the MANOVA multivariate analysis, Wilks’ multivariate F-tests and univariate F tests prove that there is a statistically significant difference between the respondents representing the three occupational groups collectively in terms of global assessment of all work safety climate dimensions, and also indicate significant differences between workers belonging to the

three occupational groups in terms of their assessment of the individual dimensions of the work safety climate, except the organization’s occupational health and safety management policy as well as technical facilities and ergonomics, which may be due to the universality of the requirements set for organizations with regard to these two aspects of safe behavior. The differences observed in the assessment of the remaining work safety climate dimen- sions induces one to promote more differentiated and individualized activities, taking into account the work specificity and the nature of the threats occurring in the respective working environment of the representatives of the different occupations. The difference in assess- ment of the work safety climate found in the research encourages one to create practical programs for safety, not only in the procedural and technical dimension, but also in the social and psychological one. Introduction Researchers have recently been focusing increasingly often on the safety climate in the organi- zation, including its significance for the way workers behave in terms of health and life protec- tion. The employer’s task is to protect employees against threats. It often boils down to PLOS ONE PLOS ONE | https://doi.org/10.1371/journal.pone.0243056 December 14, 2020 1 / 19 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: Stasiła-Sieradzka M, Chudzicka-Czupała A, Znajmiecka-Sikora M (2020) Work safety climate. Comparison of selected occupational groups. PLoS ONE 15(12): e0243056. https://doi.

org/10.1371/journal.pone.0243056 Editor: Sergio A. Useche, Universitat de Valencia, SPAIN Received: June 18, 2020 Accepted: November 15, 2020 Published: December 14, 2020 Copyright: © 2020 Stasiła-Sieradzka et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the manuscript and its Supporting Information files. Funding: The following institutions provided funding for publication fees: Institute of Psychology, Faculty of Social Sciences, University of Silesia in Katowice, Katowice, Poland, awarded to MSS; and Institute of Psychology, University of Lodz in Ło´dz, Poland, awarded to MZS. The Ministry of Science and Higher Education in Poland provided support for this study to ACC, under the

organizing safe workplaces and preventing the possibility of accidents occurring, e.g. by sys- tematic training of workers, issuing clear instructions, making appropriate personal and col- lective protection equipment available, and using effective shields and safeguards on moving machine parts. However, an analysis of the research reports so far indicates that such a classic approach to understanding the employer’s role in creating a safe working environment, based only on compliance with legal requirements, is incapable of ensuring high standards of that environment [1–3]. Nor is a systemic approach to safety management, consisting in constant monitoring of its status and in implementing intervention measures on the procedural and technical level, a sufficient tool shaping a high level of safety culture within the organization, even though it does yield better results and results in lower accident rates [3–8]. Experts tend to reflect on the role of psychological mechanisms, which often play a key role in accidents [9,10]. Therefore, it is legitimate to take into account, apart from procedural and technical improvements, also the subjective feelings of workers regarding the current level of satisfaction with the safe working conditions created by their organization as well as their own attitude and subjective perception of the existing regulations and practices in this respect. E.S. Geller [11,12] and D. Cooper [13] argue that two dimensions of safety culture need to be taken into account in its analyses. The first dimension consists in visible measures related to

safety management in the organization. Studies on this aspect of safety culture rely on an audit of the procedures applied and of the workplace in terms of its safety, as well as reviews of the causes and effects of accidents that have taken place and the remedial mechanisms introduced. The organization’s security climate is the other significant, hidden dimension, related to psy- chological factors, behavioral patterns, values and attitudes, as well as standards adopted in the organization and their perception by the employees. Importantly, studying this dimension requires the employees to be engaged and it is impossible for researchers to use documentary analysis alone. Similarly, R. Studencki [14], defining work safety culture as part of the overall organiza- tional culture of a company, clearly contrasts it with the safety climate. In this author’s opinion, an organization’s safety culture refers to the state of awareness of threats, the defined standards of conduct in situations when they occur, and the technical and organizational ways to ensure the safety of employees (e.g. readiness of rescue services, restrictive monitoring of the technical efficiency of emergency alarm devices, training of employees). The safety climate is expressed by the individual beliefs of the people working in the organization regarding life and health and the need to protect them in work situations. Other researchers [15–18] define work safety climate as a dimension of the organization’s safety climate, constituting a sum of observations

shared by the employees about their working environment and the organizational manage- ment system, including policy, practices and procedures aimed at ensuring safety. The work safety atmosphere reflects in a way the perceived atmosphere related to occupational health and safety. It influences workers’ attitudes and behavior in terms of health and safety at work, determining safe or risky behaviors [10,15,17,19]. It is emphasized in the literature that safety climate surveys may be considered as activities making it possible to develop practical recommendations for improving occupational safety in the organization [3,8,18,20,21]. They concern mainly spheres such as safety education, shaping standards and values, building the community and co-responsibility for others, responsible leadership, building authority, and motivating safe behaviors. Implementation of programs of modifying hazardous behaviors, based on the results of safety climate surveys, contributes to an improvement of the safety culture and of the quality of life of employees, and is also an effective tool for accident prevention. As empirical studies so far [5,15,17,20–25] have shown, studying the safety climate may constitute an important element of an appropriate and com- plete diagnosis of an organization’s safety culture. It is worth noting that the climate survey PLOS ONE Work safety climate PLOS ONE | https://doi.org/10.1371/journal.pone.0243056 December 14, 2020 2 / 19 2019-2022 program "Regional Initiative of Excellence", project number 012 / RID / 2018/19.

Competing interests: The authors have declared that no competing interests exist.

itself, which becomes a starting point for the implementation of practical solutions for shaping a positive safety culture of a particular organization, may be crucial for improving the safety of employees. Safety climate in selected occupational groups. Risk factors Research reports [26] show that safety climate is perceived differently by staff depending on the size of the company and on the way in which work safety is managed, as well as on the gen- der of the employees, their seniority, their position in the organizational structure, their ethnic background, and the industry represented. Research so far points to significant differences in the evaluation of threats and in the perception of the safety climate presented by occupational groups such as underground, open-pit and borehole mines, airport ground staff, sailors, people working in polar regions, medical staff, lighting industry and construction industry workers [26–32]. Building a positive culture of work safety is therefore not marked by organizational or sectoral universality. Its appropriate creation requires identification of the factors of the work safety climate, which may constitute its important components, taking into account both the size of the organization and the way in which it is managed (including work safety manage- ment), as well as the different nature of the threats and the specificity of the tasks in different areas of occupational activity. Planning and implementation of effective programs to improve the work safety climate

should be linked to an understanding of the specific nature of the occupation and of the work- ing conditions in the relevant organization. It is important to make sure that they take into account the assessment of the safety level, performed by the occupational group representing it operationally. The specificity of work is defined by a number of variables, including the way tasks are performed, the type and scope of hierarchical relations, customer relations, continu- ous contact with technology (work in a human-machine system), the level of training required for the employee and the types of threats present in the working environment. All activities aimed at developing solutions contributing to the shaping of a positive safety culture of the organization should therefore take into account the specific nature of the latter. They must not rely on the adoption of solutions implemented in other organizations without reflection. The aim of the research undertaken by the author was to continue the empirical analyses related to the comparison of the overall assessment of the work safety climate and its components, car- ried out by different occupational groups. The comparative groups were industrial manufacturing line workers, retail workers and mining rescuers. The choice of these occupa- tional groups was dictated by the different nature of their work and of the threats occurring in the working environment. Work on an industrial production line involves many potential threats whose nature is

physical (e.g. moving parts of machines, noise, vibrations, electromagnetic field, high or low temperatures), chemical (toxic substances, irritants and allergens), and biological (fungi, bacte- ria, pathogenic viruses). Physical strain on the body may result from the need to perform work in a forced body position, repetitive activities, and excessive energy expenditure caused by strain on selected body parts. In addition, an automated working environment characterized by a routine scope of tasks and limited interactions staff members, for instance due to noise, can significantly affect the workers’ sense of threat. The work is usually carried out under in strict subordination within a hierarchical structure, paid for on an hourly or piecework basis, it is sometimes carried out in shifts, and often does not require specialized training. Retail workers are also exposed to numerous physical risks. Their working environment often involves excessively high or low temperatures related to the storage conditions of goods (meat, dairy products), humidity (plant trade), biological hazards (in the case of trade in PLOS ONE Work safety climate PLOS ONE | https://doi.org/10.1371/journal.pone.0243056 December 14, 2020 3 / 19

agricultural products and animals), and chemical hazards (trading in pesticides). The nuisance related to work in retail is caused primarily by the physical effort, related to dynamic loads (when walking) and static loads (in a standing position). In addition, retail workers are exposed to substantial psychological strain. Working in constant contact with the customer, the feeling of threat related to financial liability, the need to work during holidays, as well as intensified workloads during the so-called high season (e.g. before Christmas, during the sales period) may deepen emotional tensions, lead to greater stress and disturb the work-life bal- ance. Work is carried out both in teams and individually, and the qualifications required from the employees tend to vary and depend on the specific range offered by the given retail outlet. The job of a mine rescuer is a difficult and dangerous profession in which helping others in an emergency involves risking one’s own most valuable resources, life and health. This is due to the fact that rescuers have to participate in situations involving high physical strain, as the pro- fession involves a high likelihood of injury, and mental strain, due to the frequently experienced traumatic stress. This is a highly injury-prone occupation. This is because rescuers have to par- ticipate in events causing strong physical and mental strain due to the frequent experience of traumatic stress [33–36]. The aim of the actions undertaken by rescuers is to rescue people and

property at risk and to eliminate the threat posed by mining operations, caused by natural or technical factors related to mining. Rescue operations last from a dozen hours or so to several dozen days [37]. The threats associated with this occupation involve exposure to work in harm- ful, unpredictable conditions and rapidly changing circumstances (caving, fire, gas, and water hazards), with the possibility of various types of risk factors being activated in an unpredictable manner, with high physical effort. Workers are exposed to active participation in events during which operations are carried out under constant threat of worsening conditions, which gener- ates and intensifies the feeling of stress. Mine rescuers always work in teams. The smallest orga- nizational unit in mine rescue services is composed of five members in total, including four rescuers and one leader. It is a fixed group of people, whose goal is to perform activities jointly as part of the rescue operation. In mine rescue units, care is taken to make sure that the group’s composition stays permanent, its members participate together in training and rescue opera- tions, and remain on standby duty for many days, quartered together over that time. During performance of rescue operations, the unit works in one place, and all its members must remain within sight of the others. Rescuers are required to watch one another constantly and to be attentive to the external environment, which guarantees safety on the individual and on the

group level. The members of the unit operate in accordance with the “one for all, all for one” principle. Recruitment for this job involves a restrictive selection procedure. In addition to the formal requirements, training and experience related to work as a miner, the candidates are also expected to be fully mentally and physically fit and ready to participate in specialized training courses, involving improvement of the competencies required for mining rescue operations. Staying in the profession involves the requirement to undergo systematic training with regard to technical skills as well as to maintain a high level of psychophysical fitness. Research objectives The study was of an exploratory nature. The aim was to compare selected occupational groups, industrial production line workers, retail workers and mine rescuers, in terms of their assess- ment of the work safety climate. The criterion for selection of the occupational groups was the different specificity of the work performed by their respective representatives and the different types of threats related to the environment where it is carried out. Due to the lack of research reports referring to comparative analyses with regard to the assessment of the work safety cli- mate between the occupational groups identified in the presented studies, it was decided not PLOS ONE Work safety climate PLOS ONE | https://doi.org/10.1371/journal.pone.0243056 December 14, 2020 4 / 19

to put forward hypotheses about the nature of the potential differences and similarities. The aim of the study was to fill the gap existing in this area. The research questions put forward take into account both the differences in the assessment of the safety climate at work, considered globally, and the differences with regard to its individ- ual components, each of its dimensions individually. Question 1. Are there any statistically significant differences between workers belonging to the three occupational groups studied in terms of their global assessment of the safety cli- mate at work safety (i.e. superposition of the distinguished dimensions in the research tool)? Question 2. Are there any statistically significant differences between workers belonging to the three occupational groups studied in terms of their assessment of the individual dimen- sions of the work safety climate: a. employee participation in safety-related matters, b. safe behaviors, c. management engagement, d. modeling and enhancing safe behaviors in the organization, e. risk management in the workplace, f. technical facilities and ergonomics, g. pace of work and fatigue level, h. occupational health and safety training process, i. atmosphere in the workplace, j. organizational policy with regard to occupational health and safety management? Material and methods The research was conducted in the territory of Poland in the period from 2017 to 2019. It was a quantitative survey carried out using questionnaires, filled out by the respondents using the

paper-and-pencil method, with the written consent of the authorized representatives of the working establishments employing the respondents. All study participants gave their consent to participate in the study and received relevant information to give informed consent to par- ticipate. The research was conducted in compliance with the ethical standards in line with the provisions of the Declaration of Helsinki. The Institutional Review Board and the Ethics Com- mittee of the Silesian University (Poland) (Reference Number KEUS 31/04/2020/ Human par- ticipants, project title: Work safety climate. Comparison of selected occupational groups) approved the research proposal and the consent procedure. The study involved 2,995 respondents (see S1 Data. Work safety climate.xlsx). Purposive sampling was used, based on criteria related to the respondents being members of one of the three occupational groups distinguished earlier. There were 1,331 respondents working directly on production lines manufacturing various products (ceramic, lighting and food industries), i.e. 44.4% of the total sample, 1,431 respondents (47.8% of the sample) were retail workers (cosmetics, household cleaning products, food products, household appliances), while 233 respondents (7.8% of the sample) were mine rescuers (from the Central Mine Rescue Sta- tion–Centralna Stacja Ratownictwa Go´rniczego S.A. in Bytom, the Mine-Smelter Emergency Rescue Division of KGHM Polska Miedź S.A. in Lubań, the Mine Rescue Unit of PGNiG SA–

Division of the Borehole Mining Rescue Station in Krakow). All the research results presented PLOS ONE Work safety climate PLOS ONE | https://doi.org/10.1371/journal.pone.0243056 December 14, 2020 5 / 19

and the conclusions formulated on their basis refer, therefore, only to the three occupational groups presented here. Table 1 shows the sociodemographic data of the respondents in total as a group and the data of individuals belonging to the individual occupational groups. An abridged version of the Safety Climate Questionnaire by Znajmiecka-Sikora [38], con- sisting of 50 questions (see S1 Appendix. Items.pdf), was used to measure the safety climate. The tool used a Likert scale from 1 to 5, with 1 meaning “strongly disagree”, 2 –“disagree”, 3 –“hard to say”, 4 –“agree”, and 5 –“strongly agree”. Some questions require inversion. The tool consists of 10 separate safety climate dimensions, distinguished on the basis of the- oretical analysis [6,12]. The contents of the individual items in the scales was evaluated by competent judges to assess the accuracy of individual questionnaire items, and the attribution of the specific items to the individual scales and dimensions was confirmed in the course of exploratory factorial analysis. Each scale is composed of 5 questions the minimum score on each scale is 5 points and the maximum score is 25 points. Table 2 contains a brief description of each of the scales and examples of questions which they include. The individual results show how specific workers perceive the work safety climate in the organization employing them in the individual dimensions. The overall result, obtained by all the employees, reflects the work safety climate in the organization, and makes it possible to

conclude on the level of safety culture in the company. It also allows comparisons to be made this respect with results obtained in other organizations and occupational sectors. Reliability of the entire Safety Climate Questionnaire, calculated using Cronbach’s alpha, is 0.93, ranging for the individual scales of the questionnaire between 0.68 and 0.89 [38], and for the presented study from 0.61 to 0.88. Table 3 indicates the Cronbach’s alpha values for the individual scales in the tested samples. Statistical data analysis methods In the first stage of the analysis, descriptive statistics of the variables and correlation coeffi- cients between the results of the individual Safety Climate Questionnaire scales were Table 1. Study sample description. Pro´ba Średnia wieku SD Płeć Wykształcenie Średni staż pracy w zakładzie i zawodzie SD All respondents (N = 2995) 33.31 10.43 (F)1881(62.8) (E)61(2.0%) 5.98 7.59 (V)403(13.5)% (M)1114(37.2) (S)1908(60.7%) (H)623(20.8%) Grupa A (Mnfc) (N = 1331) 36.44 12.05 (K)678(51%) (P)47(3.5%) 8.15 9.45 (Z)191(22%) (M)652(49%) (Ś)864(65%) (W)129(9.5%) Grupa B (Retl) (N = 1431) 29.31 7.24 (K)1201(84%) (P)14(1.0%) 2.95 2.85 (Z)87(6.0%) (M)230(16%) (Ś)929(65%) (W)401(28%) Grupa C (MnRsc) (N = 233) 39.83 7.19 (K)0(0%) (Z)25(10.5%) 11.61 7.59 (M)233(100%) (Ś)115(49.5%) (W)93(40%) SD–standard deviation; Gender: F-female, M-male; Education:E-elementary, V-basic vocational, S-secondary, H-higher Group: A(Mnfc)-production workers; B(Retl)-retail workers; (MnRsc)C-mine rescuers

https://doi.org/10.1371/journal.pone.0243056.t001 PLOS ONE Work safety climate PLOS ONE | https://doi.org/10.1371/journal.pone.0243056 December 14, 2020 6 / 19

Table 2. Description of the Safety Climate Questionnaire and contents of the individual items. No. Safety Climate Questionnaire scales Description of the scale corresponding to the specific safety climate dimension A high score on the scale means that: Example statements from the scale 1 Employee participation in safety- related matters workers feel engaged by their superiors in safety-related matters and actions implemented in this field are consulted with them I have been engaged in the process of assessment of occupational risk related to their job 2 Safe behavior workers comply with the applicable safety procedures, use personal protective equipment and keep their working station orderly and tidy Sometimes I engage in risky behavior at work, e.g. I remove machine guards or shields, I perform minor repairs when the machine is running, I exceed speed limits, I take shortcuts to reach my destination faster.� 3 Management engagement in health and safety workers have a high opinion of the attitude and behavior of their superiors and their engagement in work safety aspects My superiors “turn a blind eye” to the way in which work is done–it does not necessarily have to be done safely, the important thing is to meet the deadlines and that the quantity is appropriate.� 4 Modeling and enhancing safe behaviors in the organization safe way of performing tasks, compliance with safety rules is noticed and appreciated by superiors, and their attitude towards safety is taken into account in the employee performance appraisal process

My superiors actively participate in campaigns promoting safety organized by the company. 5 Workplace accident risk management workers rate positively the accident prevention and response measures undertaken by the organization Each accident that has occurred at the working establishment is discussed by the superior during information meetings. 6 Technical facilities and ergonomics workers have a high opinion of both the technical condition of the plant and machinery and the way in which work is organized Particularly dangerous places are suitably marked. 7 Pace of work and fatigue level workers rate positively the way work is organized and do not feel overtired The number of tasks I have to perform every day makes me work at a very fast pace.� 8 Occupational health and safety training process workers feel well informed about aspects related to safety, occupational risk, job-related threats, and methods of providing first aid I feel well informed about the methods of protecting myself from the threats related to my work. 9 Atmosphere in the workplace workers rate positively the communication method and the level of support provided by the superior, and trust one another When dangerous work is performed, I know I can trust my co-workers with whom I perform the dangerous task. 10 Organizational policy with regard to occupational health and safety management workers recognize and rate positively the actions undertaken by the employer to improve safety in the organization Safety is a priority in our organization, and a continuous

improvement process with regard to safety is in course. �Question requiring inversion Source: compilation based on Znajmiecka-Sikora [38]. https://doi.org/10.1371/journal.pone.0243056.t002 Table 3. Scale reliability: Cronbach’s alpha for the individual scales of the Safety Climate Questionnaire in the individual respondent groups. Safety Climate Questionnaire scales Group A (Mnfc) Group B (Retl) Group C (MnRsc) Employee participation in safety-related matters α = 0.67 α = 0.61 α = 0.77 Safe behavior α = 0.70 α = 0.70 α = 0.75 Management engagement in health and safety α = 0.71 α = 0.75 α = 0.83 Modeling and enhancing safe behaviors in the organization α = 0.77 α = 0.74 α = 0.81 Workplace accident risk management α = 0.76 α = 0.67 α = 0.81 Technical facilities and ergonomics α = 0.72 α = 0.73 α = 0.84 Pace of work and fatigue level α = 0.77 α = 0.81 α = 0.77 Occupational health and safety training process α = 0.69 α = 0.74 α = 0.88 Atmosphere in the workplace α = 0.73 α = 0.78 α = 0.79 Organizational policy with regard to occupational health and safety management α = 0.78 α = 0.77 α = 0.85 https://doi.org/10.1371/journal.pone.0243056.t003 PLOS ONE Work safety climate PLOS ONE | https://doi.org/10.1371/journal.pone.0243056 December 14, 2020 7 / 19

calculated, separately for the individual respondent groups (representatives of the three occu- pational groups distinguished). A MANOVA multivariate analysis was conducted to find answers to the research questions put forward (cf. Table 6). Multivariate Wilks’ F tests were used to find an answer to question 1, concerning the existence of a statistically significant difference between the representatives of the three occupational groups examined together and the three groups of workers compared in pairs (production workers to retail workers, production workers to mine rescuers, and retail workers to mine rescuers) in terms of the global assessment of the safety climate at work (superposition of the assessment of all safety climate dimensions). Univariate F tests were used to find an answer to question 2, concerning the existence of differences in the assessment of the individual work safety climate dimensions, made by representatives of all three occupa- tional groups together, and then for the occupational groups analyzed in pairs. The adoption of such an analysis model is “an optimal choice when the researcher treats the explained measurable variables as interrelated and thus as ones creating a relational structure of properties which can be meaningfully and substantively interpreted, in whole or in part” [39], and this is the case with respect to the assessment of the individual dimensions of the work safety climate. All statistical analyses were carried out using the STATISTICA 13 statisti- cal package. Results

Descriptive statistics concerning all the studied variables (the mean values obtained and the standard deviations) for each group of respondents, representing the three occupational groups, are presented in Table 4. Table 5 contains the intercorrelations between the studied variables separately for the three occupational groups studied. All variables are significantly correlated with a significance level of p = 0.01. The correlation in the vast majority of the cases is weak or average, and only single coefficients reach 0.7, indicating a high correlation. The results shown in Table 6 were analyzed to answer question 1: they prove the existence of a statistically significant difference between the studied representatives of the three selected occupational groups together in terms of the global assessment of all work safety climate dimensions. The multivariate Wilks’ F test value (F = 69.18, p = 0.00), indicates that all the Table 4. Descriptive statistics of the studied variables. Variables Group A (Mnfc) (N = 1,331) Group B (Retl) (N = 1,431) Group C (MnRsc) (N = 233) M SD M SD M SD Employee participation in safety-related matters 18.59 3.08 18.07 2.97 18.61 3.83 Safe behavior 19.93 3.51 20.75 2.94 18.34 3.81 Management engagement in health and safety 20.12 3.43 21.32 2.90 18.64 4.13 Modeling and enhancing safe behaviors in the organization 18.06 3.05 18.59 2.85 17.64 3.43 Workplace accident risk management 18.99 3.17 16.94 3.16 19.56 3.20 Technical facilities and ergonomics 20.56 2.38 20.35 2.53 20.51 2.82

Pace of work and fatigue level 13.18 3.82 14.10 4.24 16.52 3.59 Occupational health and safety training process 19.49 2.74 20.36 2.38 20.81 2.76 Atmosphere in the workplace 19.27 2.68 20.62 2.83 20.17 2.84 Organizational policy with regard to occupational health and safety management 19.23 2.66 19.21 2.67 19.28 3.19 M–mean, SD–standard deviation Sample:A(Mnfc)–production workers; B(Retl)–retail workers; C(MnRsc–mine rescuers https://doi.org/10.1371/journal.pone.0243056.t004 PLOS ONE Work safety climate PLOS ONE | https://doi.org/10.1371/journal.pone.0243056 December 14, 2020 8 / 19

three groups differ significantly in this respect. The results (cf. Table 6) also make it possible to conclude that in the global assessment of the work safety climate, there are significant differ- ences between representatives of all the occupational groups analyzed in pairs: industrial pro- duction line workers with retail workers (F = 82.14, p = 0.00), industrial production line workers with mine rescuers (F = 42.46, p = 0.00) and retail workers with mine rescuers (F = 72.00, p = 0.00). Univariate F tests were performed to find an answer to question 2. Calculations were per- formed for the assessments of the individual work safety climate dimensions for the three occupational groups together and for all the groups compared in pairs. The results (cf. Table 6) suggest that all the studied worker groups together differ significantly in terms of the following dimensions: employee participation in safety matters, safe behaviors, management’s Table 5. Correlations between the studied variables. Variable 1 2 3 4 5 6 7 8 9 Group A (Mnfc)–Production workers (N = 1,331) 1. Employee participation in safety-related matters 2. Safe behavior 0.17 3. Management engagement in health and safety 0.30 0.55 4. Modeling and enhancing safe behaviors in the organization 0.48 0.36 0.50 5. Workplace accident risk management 0.49 0.28 0.37 0.51 6. Technical facilities and ergonomics 0.43 0.37 0.50 0.49 0.47 7. Pace of work and fatigue level 0.17 0.19 0.30 0.28 0.09 0.22 8. Occupational health and safety training process 0.41 0.29 0.40 0.48 0.44 0.52 0.19

9. Atmosphere in the workplace 0.29 0.30 0.42 0.45 0.31 0.45 0.30 0.48 10. Organizational policy with regard to occupational health and safety management 0.56 0.35 0.52 0.59 0.61 0.59 0.26 0.53 0.48 Group B(Retl)–Retail workers (N = 1,431) 1. Employee participation in safety-related matters 2. Safe behavior 0.28 3. Management engagement in health and safety 0.28 0.60 4. Modeling and enhancing safe behaviors in the organization 0.49 0.40 0.52 5. Workplace accident risk management 0.41 0.21 0.25 0.47 6. Technical facilities and ergonomics 0.36 0.34 0.39 0.41 0.49 7. Pace of work and fatigue level 0.28 0.40 0.39 0.39 0.20 0.19 8. Occupational health and safety training process 0.43 0.36 0.42 0.48 0.36 0.52 0.22 9. Atmosphere in the workplace 0.32 0.41 0.51 0.47 0.19 0.34 0.37 0.53 10. Organizational policy with regard to occupational health and safety management 0.44 0.40 0.50 0.58 0.52 0.56 0.31 0.55 0.45 Group C(MnRsc)–Mine rescuers (N = 233) 1. Employee participation in safety-related matters 2. Safe behavior 0.44 3. Management engagement in health and safety 0.48 0.74 4. Modeling and enhancing safe behaviors in the organization 0.64 0.52 0.64 5. Workplace accident risk management 0.53 0.30 0.42 0.61 6. Technical facilities and ergonomics 0.62 0.53 0.61 0.68 0.58 7. Pace of work and fatigue level 0.19 0.54 0.55 0.41 0.19 0.40 8. Occupational health and safety training process 0.56 0.48 0.54 0.60 0.67 0.73 0.28

9. Atmosphere in the workplace 0.48 0.47 0.58 0.65 0.61 0.65 0.39 0.76 10. Organizational policy with regard to occupational health and safety management 0.62 0.57 0.67 0.71 0.68 0.75 0.42 0.72 0.68 p = 0.01 https://doi.org/10.1371/journal.pone.0243056.t005 PLOS ONE Work safety climate PLOS ONE | https://doi.org/10.1371/journal.pone.0243056 December 14, 2020 9 / 19

commitment to health and safety, modeling and strengthening safe behaviors in the organiza- tion, workplace accident risk management, pace of work and level of fatigue, health and safety training process, and atmosphere in the workplace. For the areas mentioned above, the values of the univariate F test are statistically significant. However, no statistically significant differ- ences were found between the three occupational groups studied together with regard to the assessment of the other work safety climate dimensions, such as technical facilities and ergo- nomics, as well as the organizational policy with regard to occupational health and safety management. The analysis of the differences between the assessment made by representatives of the indi- vidual occupational groups analyzed in pairs proves the existence of statistically significant dif- ferences between all the groups studied. Significant differences occur between industrial production line workers and retail workers in all the work safety climate dimensions studied, except the organizational occupational health and safety management policy dimension. Table 6. Results of MANOVA variance analysis (multivariate and univariate tests). All effects (Wilks multivariate test): F = 69.18p = 0.00 partial eta-squared = 0.19 Variables Univariate test for 3 groups A, B, C Multivariate test (for groups A and B) F = 82.14 p = 0.00 eta-squared, partial = 0.23 Multivariate test (for groups A-C) F = 42.46 p = 0.00 eta-squared, partial = 0.21 Multivariate test (for groups B-C) F = 72.00p = 0.00 eta-squared, partial = 0.30

Univariate test for groups A-B Univariate test for groups A-C Univariate test for groups B-C F Adjusted R2 F Adj. R2 F Adj. R2 F Adj. R2 1. EPSRM 10.97�� 0.01 20.96�� 0.01 0.01 -0.00 6.21� 0.00 2. SB 62.68�� 0.04 45.45�� 0.02 39.42�� 0.02 123.39�� 0.07 3. MEH&S 91.69�� 0.06 99.00�� 0.03 34.47�� 0.02 149.11�� 0.08 4. M&ESBO 16.68�� 0.01 22.55�� 0.01 3.49 0.00 20.77�� 0.01 5. WARM 173.12�� 0.10 289.70�� 0.09 6.35� 0.00 137.52�� 0.08 6. TF&E 2.46 0.00 4.93� 0.00 0.09 -0.00 0.74 -0.00 7. PWFL 72.74�� 0.05 35.80�� 0.01 153.39�� 0.09 67.69�� 0.04 8. OH&STP 52.07�� 0.03 80.88�� 0.03 45.97�� 0.03 6.61�� 0.00 9. WA 81.90�� 0.05 163.46�� 0.06 21.66�� 0.01 5.04� 0.00 10. OOH&SMP 0.09 -0.00 0.04 -0.00 0.08 -0.00 0.16 -0.00 significance of differences between mean values for study samples A, B and C �� p = 0.01 � p = 0.05 Group: A(Mnfc) (N = 1,331)–production workers; B(Retl) (N = 1,431)–retail workers; C (MnRsc) (N = 233)–mine rescuers Variables: EPSRM–Employee participation in safety-related matters SB–Safe behavior MEH&S–Management engagement in health and safety M&ESBO–Modeling and enhancing safe behaviors in the organization WARM–Workplace accident risk management TF&E–Technical facilities and ergonomics PWFL–Pace of work and fatigue level OH&STP–Occupational health and safety training process WA–Atmosphere in the workplace OOH&SMP–Organizational policy with regard to occupational health and safety management https://doi.org/10.1371/journal.pone.0243056.t006 PLOS ONE Work safety climate

PLOS ONE | https://doi.org/10.1371/journal.pone.0243056 December 14, 2020 10 / 19

three occupational groups together concern the global assessment of all work safety climate dimensions. The effect size (cf. Table 6) is relatively high here, as 19% of the variance of the global safety climate assessment is explained by the determining factor, i.e. the fact of belong- ing to the specific occupational group. The results also confirm the existence of significant differences in the global assessment of all work safety climate dimensions between industrial production lines workers and retail workers (23% of the variance of the global work safety climate assessment is explained here by the factor of belonging to the occupational group), between industrial production lines work- ers and mine rescuers (21% of the variance of the global work safety climate assessment is explained by the factor of belonging to the occupational group) and between retail workers and mine rescuers (where belonging to the occupational group explains as much as 30% of the variance of the global work safety climate assessment). Question 1 can be answered affirmatively. The univariate F test values and effect sizes measured using adjusted R2 (cf. Table 6) indi- cate that all the studied worker groups together differ significantly between one another in terms of the assessment of most work safety climate dimensions, and particularly strongly in their assessment of the following dimensions: workplace accident risk management (the lowest rating was given to this dimension by retail workers, and the highest by mine rescuers), man-

agement engagement in occupational health and safety (rated lowest by mine rescuers, and highest by retail workers), workplace atmosphere (rated lowest by industrial production line workers, and highest by retail workers), pace of work and fatigue level (rated lowest by indus- trial production line workers, slightly higher by retail workers, and highest by mine rescuers), safe behavior (this safety culture aspect was rated lowest by mine rescuers, and highest by retail workers) and the occupational health and safety training process (rated lowest by industrial production line workers, and highest by mine rescuers). All three occupational groups together also differed significantly between one another in terms of the assessment of the following dimensions: modeling and enhancing safe behaviors in the organization (this safety culture aspect was rated lowest by mine rescuers, and highest by retail workers) and employee participation in safety-related matters (rated lowest by retail workers and highest by mine rescuers). There are no significant differences between the PLOS ONE Work safety climate PLOS ONE | https://doi.org/10.1371/journal.pone.0243056 December 14, 2020 11 / 19

studied representatives of the three occupational groups together in terms of their assessment of the organizational occupational health and safety management policy and the technical facilities and ergonomics dimensions. It can be concluded on the basis of the results that ques- tions 2a, 2b, 2c, 2d, 2e, 2g, 2h and 2i can be answered affirmatively, while questions 2f and 2j should be answered negatively. The results also indicate the existence of a significant difference between the assessments given by the respondents when the analyses involved comparisons of two occupational groups each time (univariate F test values and effect sizes measured with adjusted R2, see Table 6). Sig- nificant differences were found between production workers and retail workers in most of the studied work safety climate dimensions, except the organizational occupational health and safety management policy dimension. These differences concern mainly the following dimen- sions: workplace accident risk management (rated much higher by production workers), workplace atmosphere, management engagement in health and safety and occupational health and safety training process (the last three dimensions rated significantly higher by retail workers). Similarly significant differences between industrial production line workers and mine res- cuers in terms of their judgments concern six dimensions of work safety climate. The main dif- ferences between these two groups concern their assessments of work safety climate aspects

such as the pace of work and fatigue levels and the occupational health and safety training pro- cess (these dimensions are rated much higher by mine rescuers). Significant differences also concern the assessment of the following work safety climate dimensions: safe behavior and management engagement in health and safety (dimensions rated much higher by production workers) as well as workplace atmosphere and workplace accident risk management (rated higher by mine rescuers). Industrial production line workers and mine rescuers do not differ in terms of their assessment of employee participation in safety-related matters, modeling and enhancing safe behaviors in the organization, technical facilities and ergonomics, and organi- zational occupational health and safety management policy. The results also prove the existence of statistically significant differences in the assessment of most dimensions of the work safety climate between retail workers and mine rescuers. They concern mainly aspects such as management engagement in health and safety and safe behav- ior (rated significantly higher by retail workers), workplace accident risk management, and pace of work and fatigue levels (rated significantly higher by mine rescuers). Significant differ- ences between these occupational groups were also observed in the assessment of the following dimensions: modeling and enhancing safe behaviors in the organization and atmosphere in the workplace (rated significantly higher by retail workers), as well as the occupational health

and safety training process and employee participation in safety-related matters (perceived more positively by mine rescuers). No significant differences were found between retail work- ers and mine rescuers in terms of their assessments of work safety climate aspects such as tech- nical facilities and ergonomics, and the organizational policy with regard to occupational health and safety management. Discussion The lack of differences observed in the presented results between the studied groups in their assessment of such work safety climate dimensions as organizational occupational health and safety management policy and technical facilities and ergonomics may result from the fact that in practice, they constitute the basic and often the only area of activities in many organizations. Initiatives of this nature are highly visible to the employees, as they are related to the reward and penalty systems in place in the respective companies, as well as systematic training and PLOS ONE Work safety climate PLOS ONE | https://doi.org/10.1371/journal.pone.0243056 December 14, 2020 12 / 19

instructions related to the safe use of tools and behavior in the workplace, whose nature is uni- versal [5,40]. With regard to the differences observed between the groups, it was noted that industrial production line workers did not rate any of the studied work safety climate dimensions signifi- cantly higher than representatives of the other groups. On the other hand, they rated signifi- cantly lower dimensions such as pace of work and fatigue level, occupational health and safety training process, and atmosphere in the workplace. The differences observed are justified by the specific nature of the work performed by mem- bers of this occupational group. Their work involves physical strain, which is certainly linked to the level of fatigue. Limited contact with other colleagues, due to the specificity of the place of work (single-workstation job) and the competitive reward system, may have a significant impact on the lower rating attributed to social relations. The process of training or informing employees about safety procedures is different among production workers compared to the other occupational groups analyzed. Information meetings are organized for the whole staff (shift), and training is usually held for large groups, which may affect both the quality of the communication and the effectiveness of the training as such [41]. Compared to the other occupational groups, retail workers declare the safest behaviors, giv- ing the highest rating to the management’s engagement in health and safety, modeling and

enhancing safe behaviors, and the workplace atmosphere. They also give the lowest rating, compared to the other groups, to the process of participation in safety-related matters and accident risk management. Compared to production workers, retail workers declare a lower level of fatigue and a slower pace of work, and they give a better rating to the occupational health and safety training process. The situation is totally different when retail workers are compared to mine rescuers. Retail workers declare a higher level of fatigue than mine rescuers, as well as a faster pace of work, and they give a lower rating to the training process. Also in this case, the results obtained are justified by the specific nature of the work per- formed by the study respondents. Work in retail usually does not require one to use personal protective equipment, with a lower level of psychophysical strain than in the case of mine res- cuers and production workers. Their work is carried out mostly in small teams, which makes it undoubtedly easier to build interpersonal relationships and an appropriate atmosphere. Train- ing with regard to safety rules is held most often in small groups, which improves the effective- ness of such processes [41] The possibility of having direct contact with one’s superior makes it possible to model appropriate attitudes towards safety rules in this case. Mine rescuers, compared to representatives of the other two occupational groups, rate sig- nificantly higher such work safety climate dimensions as pace of work and fatigue levels, occu-

pational health and safety training process, and organizational policy with regard to health and safety management. In addition, representatives of this occupation group rated significantly higher the atmosphere in the workplace compared to production line workers, and employee engagement in safety-related matters compared to retail workers. They gave a significantly lower rating, on the other hand, to work safety climate dimensions such as safe behavior and management engagement. Compared to retail workers, they also rated significantly lower modeling and enhancement of safe behaviors in the organization as well as atmosphere in the workplace. The results obtained are justified by the specific nature of work in mine rescue services. Referring to the differences noted between the groups in their assessment of the pace of work and fatigue level, it is worth pointing out that more positive assessment in the rescuer group may be explained by the fact that, although the conditions in which rescue operations are car- ried out involve a huge physical and mental strain, the circumstances of extreme strain are cur- rently treated as extraordinary. Representatives of the other two groups, in turn, are often PLOS ONE Work safety climate PLOS ONE | https://doi.org/10.1371/journal.pone.0243056 December 14, 2020 13 / 19