Exploring the dimensionality of Safety Climate in the Construction industry

The purpose of this paper is to identity the dimensionality of safety climate in a sample of construction workers who answered the “Valencia PREVACC Battery”. This battery identifies the agent of each safety climate statement and analyses the relationships that flow through the chain of safety responses. It was developed to measure the organizational safety response, the superiors’ safety response, the co-workers’ safety response, the worker’s safety response, and the perceived risk. A factor analysis was conducted in order to explore the dimensionality of the whole battery in this construction sample. Results clearly identify the five factors considered in their theoretical architecture.
Palabras Clave: 
Psychosocial factors; Safety Climate; Construction; Safety Psychology; Accident Prevention
Autor principal: 
José L.
Meliá


Meliá, José L.

Universitat de València / Unitat d’Investigació de Psicometria / Facultat de Psicologia / 46010 València / Spain / Tel: +34 96 386 45 48 Jose.L.Melia@uv.eswww.uv.es/seguridadlaboralSilva, Silvia A.Centro de Investigação e Intervenção Social / Instituto Superior de Ciências do Trabalho e da Empresa/ Lisboa / PortugalMearns, KathrynUniversity of Aberdeen / Aberdeen / Scotland / UKLima, M. LuisaCentro de Investigação e Intervenção Social / Instituto Superior de Ciências do Trabalho e da Empresa/ Lisboa / Portugal

ABSTRACT

The purpose of this paper is to identity the dimensionality of safety climate in a sample of construction workers who answered the “Valencia PREVACC Battery”. This battery identifies the agent of each safety climate statement and analyses the relationships that flow through the chain of safety responses. It was developed to measure the organizational safety response, the superiors’ safety response, the co-workers’ safety response, the worker’s safety response, and the perceived risk. A factor analysis was conducted in order to explore the dimensionality of the whole battery in this construction sample. Results clearly identify the five factors considered in their theoretical architecture.

Key words

Psychosocial factors, Safety Climate, Construction, Safety Psychology, Accident Prevention,

INTRODUCTION

In Spain, construction is a very active and expanding sector. The construction sector is a major engine of the Spanish economy. The growing sub sector of civil construction, for example, moved 40.000 million euros last year, and this was only 24% of the contract activities of the entire construction sector inside Spain. Many big civil investments in highways, ports and airports, such as, the new terminal of the Barajas Madrid airport, (the second in Europe in capacity for travellers), or the tremendous development of the Valencian port, (the most important Mediterranean port for Asian interchanges, and the site of the 2007 international America’s Cup in Sailing), contributes to this civil construction development. Civil construction shares its development with a strong expansive period for the house and building sub sector (for example, last year in Spain more new houses were finished than in France and Germany together). Furthermore, the importance of construction in the Spanish economy goes beyond the Spanish frontiers. The Spanish construction industry has been involved in a strong process of internationalisation since the late eighties. Among the 11 largest groups of construction companies in the world, 7 are Spanish. Inside

Spain, construction companies represent 13.9% of the entire number of companies employing more than 1,913,000 workers, including at least 385,000 self-employed workers, approximately 11.8% of the workforce (this is the highest percentage for the EU-15).

However, all these positive economical figures have a negative side on the accident indexes. In 2004 Spanish construction workers had about 265,850 accidents requiring at least one day off work, which represents an important incidence rate of 13,895.4 injured workers for each 100,000 workers. The construction industry is one of the sectors with the highest accident rates and the most serious accidents in terms of the injuries produced (Karjalainen, 2004). According to rates of occupational injuries in 15 European Union states, construction is the industry that shows the highest levels of work-related accidents with more than three days of absence (Lundholm, 2004). These records of serious injuries and deaths have been considered to be “disproportionally high” (HSC/HSE, 2003) and should require a special preventive effort, in order to reduce the enormous cost of accidents and injuries to workers and companies.

Accidents in construction, as in any other sector, are a complex and probabilistic result of both technical and human causes. The contribution of human psychosocial and organizational factors is of obvious importance. Haslam and colleagues, studying construction accidents, have identified five key causal factors: problems arising from workers or the work team (70% of accidents), workplace issues (49%), shortcomings in equipment (including PPE) (56%), problems with suitability and condition of materials (27%), and deficiencies in risk management (84%). Four of the five factors are clearly associated with the so-called safety climate factors identified in the literature.

In recent years, awareness about the fact that psychosocial, worker, team, supervisor, management and organizational factors interact in generating safety and health related outcomes has resulted in a considerable body of literature on safety climate (e.g., Guldenmund, 2000; Larsson, 2005, Mearns, Whitaker & Flin, 2003; Olive, O’Connor & Mannan, 2006; Silva, Lima & Baptista, 2004). Climate can be viewed as an intersubjective construct, in which there are multiple subsystem climates that can be referenced to several organizational issues, such as effectiveness, communication, quality or safety, and can be analysed across organizational levels over time (Falcione, Sussman & Herder, 1987). Safety climate is a subset of organizational climate that involves a subjective perception and evaluation of safety issues related to the organization, its members, structures and processes, based on the experience of the organizational environment and social relationships. Safety climate is one multidimensional construct believed to influence the safety behaviour of employees at the individual, group or organizational level (Smith, Huang, Ho & Chen, 2006).

Safety Climate has been considered as a “snapshot of the state of safety, providing an indicator of the underlying safety culture of a work group, plant or organization” (Flin, Mearns, O’Connor & Bryden, 2000), and it “is generally taken to comprise a summary of employee perceptions of a range of safety issues” (Glendon & Litherland, 2001). Compared to traditional safety indicators, such as accidents or accident indexes, safety climate has some positive properties as a diagnostic indicator. First, it can be measured at any time and assessed periodically, given the opportunity to control the evolution of safety independently of the presence of accidents. Second, it is an authentic preventive measure, and not a post hoc measure like accidents or accident indexes. Post hoc measures, such as accident counts and accident indexes, tell us that the facts we are trying to avoid have actually already happened. Preventive measures, such as safety climate measures, tell us that we have an opportunity to introduce preventive corrections before accidents happen. Third, a safety climate measure can be an analytic measure able to explain not only that something or somebody is working unsafely, but also who or what is working unsafely and where.

Safety climate can be seen as a reference schema of the importance of safety within each organizational context. It defines the frame of acceptable and non- acceptable safety behaviours based on the employees’ perceptions of the work setting and the organizational, supervisor and co-worker attention to safety issues. Safety climate defines the importance of safety expectations and safety behaviours and, therefore, influences safety results. The explanatory powers of safety climate lie in the fact that it provides a link between attributes occurring at the individual level (employees) and at the organizational level (Niskanen, 1994). Safety Climate has been associated with accidents (e.g., Mearns et al., 1998, 2003), safety practices (e.g., Zohar, 1980), safe behaviour (e.g., Coyle, Sleeman & Adams, 1995; Cox & Cox, 1991) and unsafe behaviour (e.g., Brown, Willis & Prussia, 2000; Hofmann & Stetzer, 1996).

Several reviews have identified certain factors as the most representativedimensions of safety climate. Williamson, Feyer, Cairns and Biancotti (1997) stated that there is clearly little real agreement among previous studies on the dimensions that should be incorporated into a safety climate model; however, across all studies, they identify two areas that seem to be reflected consistently, views about management’s attitudes to safety and about workers’ involvement or attitudes to safety. Flin et al. (2000) conclude that the most typically assessed dimensions are related to management, the safety system, risk, work pressure and competence. Seo, Torabi, Blair & Ellis (2004), after a review of the literature, infer that among the emergent themes in safety climate, five constructs “appear to constitute the core of the generic safety climate concept”: (a) management’s commitment to safety, (b) supervisor safety support, (c) co-worker safety support, (d) employee participation in safety-related decision making and activities, and (e) competence level of employees with regard to safety. These dimensions are closely related to the structure of the General Safety Questionnaire (Meliá, 2000) and the subsequent “Valencia PREVACC Battery” (Meliá, 2000; 2002). Both are devoted to measuring the organizational safety response, the supervisors’ safety response, the co-workers’ safety response, the workers’ safety response and the perceived risk.

The type of industry has been identified as a potential explanation for differences in the safety climate dimensionality (Guldenmund, 2000; Seo et al., 2004). As different industries have different configurations and levels of hazards, results from different sectors may not be generalizable across industries. Flin and colleagues’ (2000) question about whether there is sufficient evidence for a generic factor structure, or whether the components of safety climate are associated with particular sectors, remains open. Safety construction is a kind of industry characterized by: (1) an extended and defined set of hazards, some of them posing serious danger to life;(2) that changes continuously in work settings under permanent development, where the work setting becomes safe just when the work is finished; (3) a multiplicity of tasks and professions that works together and also sequentially, sharing many different risks, transversally and also longitudinally; and (4) a special structure of tasks characterized by the prevalence of subcontracted activities. The special characteristics of the construction sector may influence the safety climate structure and the relationship among safety climate variables and safety results. Therefore, it is especially interesting to test the safety climate structure under these conditions.

The purpose of this paper is to identity the dimensionality of safety climate in a Spanish sample of construction workers who voluntarily answered the “Valencia PREVACC 2000 Battery”. This paper is being prepared within the HERC Project, which focuses on the development of a safety assessment tool for the psychosocial dimensions related to injuries in the construction sector.

METHOD

Measures

For the employees, the quantitative part of the Valencia PREVACC Battery (Meliá, 2000) is a 45-item Likert-type safety climate questionnaire developed to measure the employees’ perceptions about five issues:

(1) The organizational safety response, that is, the commitment of the organization to safety, expressed through (a) safety structure, (b) safety policies and(c) safety actions, including all the workplace safety and risk management issues that are clearly the responsibility of top management.

The organizational safety response can be captured by means of two kinds of items: general items about the safety structures, policies and actions of the company(e.g. “My company has a safety training program”), and the tangible actions in which the employee has been directly involved (e.g. “I received safety training from mycompany”). In this paper, the first type of items has the acronym OSR followed by the number of the item; the second kind represents the presence of the organizationalsafety response at the worker level and has the acronym OSW followed by the item number. Both the items related to OSR and those related to OSW are concerned withresponsibilities or actions performed by the company and usually decided by the top management. The two differ in the way the content is captured, but not in the contentitself or the social agent who is responsible or involved.(2) The superiors’ safety response, i.e. in the case of employees, the supervisors’ concern, attitudes and behaviours regarding safety. The superiors’ safety response includes (a) the safe or unsafe behaviour of the supervisors in their own work, (b) the supervisors as a source of role safety demands for workers, i.e. the instructions, suggestions, communications and safety training that supervisors give to the workers, and (c) the supervisors as a source of contingencies over their workers, i.e., the reinforcement, punishment or indifference of supervisors to the safe or unsafe behaviour of their workers. The items related to the supervisors’ safety response have been identified with the acronym SSR followed by the number of the item.(3) The co-workers’ safety response, including the safety issues having to do with work team commitment, attitudes and behaviour. The informal social forces coming from the set of co-workers can influence the safety response of each worker and should be taken into account in a complete picture of social agents throughout the whole organization. Co-workers can influence the worker safety response by means of the same three basic mechanisms, i.e., (a) models of safe or unsafe behaviour, (b) demands through communication and (c) contingencies to the worker response. However, there are two main differences between supervisors and co-workers: (a) Supervisors have formal power and can use their corresponding procedures of influence and (b) co-workers usually have more informal power and can use their own methods of influence, i.e., establishing informal rules of safety performance and administrating positive social contingencies for those workers who follow the group norm. The items associated with the co-workers’ safety response have the acronym CSR followed by the number of the item.(4) The worker safety response, i.e. the self-perceived safety response of the respondent, involves the (a) safe or unsafe behaviour of the worker, e.g. the fulfilment of safety rules, (b) the aspects of communication related to safety, e.g. to report on unsafe conditions, and (c) cognitive and affective self-contingencies about safe or unsafe behaviour, e.g. being worried about the rhythm or quantity of production when he or she is working in a safe way. Items related to this matter start with the acronym WSR.

The concept of safety response goes beyond the traditional concept of safe or unsafe behaviour. Given that the self-reporting questionnaires allow us to explore some psychosocial issues hardly measurable by behavioural observation, i.e.

communication and social aspects of work or cognitive and affective self-responses, the measures of safety response consider the safety behaviour as a part of the broader concept of safety response.

The important issue of the absolute or relative importance (i.e., its balance with productivity) of safety can be considered across the various safety responses. As with other safety issues, it can be understood as critical at the organizational –top management- level and transmitted to employees step by step, throughout the organizational levels.

(5) The perceived risk, that is, the perceived probability of accidents. The way the employees perceive risk can be considered both an antecedent and a consequence of their own safety response, and it can be understood as the result of the chain of psychosocial influences that flow through the organization. At the same time, given a certain set of objective risks, it is a clear indicator of the degree of consciousness that guides the safety response. The items related to the perceived risk of accidents begin with the acronym PRA, followed by the number of the item.

The Valencia PREVACC Battery is the result of a long development process at the Psychometrics Research Unit of the University of Valencia since the eighties; it has been previously tested in several general samples (e.g. Meliá, 2004a), and it follows the rationale of the psychosocial model of work-related accidents (Meliá, 1998; 2004b). This rationale involves (1) identifying the agent (who performs what) of each safety climate statement and (2) analysing the chain of relationships that flows among the safety responses of the different organizational agents.

Sample

Responses were obtained from 374 construction employees belonging to a set of 64 small Spanish construction companies working on 182 construction and building sites. A qualified safety professional from their safety insurance company, who guaranteed anonymity, invited all the companies and workgroups personally.

Usable questionnaires represent a 42% overall response rate. All respondents were male. The average age was 36.47 years (SD = 12.11). They worked on 13 characteristic construction jobs, but most of them were building workers (31.8%), bricklayer’s mates (27.8), plumbers (9.4%) and cement structure builders (6.1%). 97.4% had a full-time contract, and only 3.6% had a part time contract. 58.5 % had a permanent job; the rest had different types of temporary job contracts, 6.1% between1 and 3 years. In this construction sample, only 9.8% worked for subcontracted companies. Average tenure in the organization was 6.32 years (SD = 8.68); 24.6 % had worked for their company for 9 years or more.

RESULTS

A principal components analysis with varimax rotation was conducted, in order to explore the dimensionality of the whole 45-item questionnaire in this construction sample.

The scree plot suggests a 5-factor structure (Figure 1), which is coherent with the number of factors expected following the rational of the battery.

Figure 1. Scree plot of the principal component analysis.

Scr e e Plo t10

86

4

Text Box: Eige n va lue

201 7 13

4 10

19 16 22

25 31 28

37 43 34 40

Co m pone nt Nu m b e r

The five main factors together explained 40.9 % of the total variance (Table 1). In the rotated solution, each factor explains from 8.95% of the variance to 6.69% of the variance.

Table 1. Percentage of variance explained by the 5-factors solution in the component analysis.

Extraction Sums of Squared Loadings

Rotation Sums of Squared Loadings

Total

% of Variance

Cumulative %

Total

% of Variance

Cumulative %

7.985

17.745

17.745

3.998

8.884

8.884

3.443

7.652

25.396

3.973

8.829

17.713

3.079

6.843

32.239

3.762

8.360

26.073

2.110

4.688

36.928

3.717

8.261

34.334

1.893

4.207

41.134

3.060

6.800

41.134

The factorial structure clearly identifies the five main issues considered in the development of the Battery. Table 2 presents the factorial saturations of the 45 items.

Table 2. Rotated solution. Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization.

Item / Item Content

FI SSR

FII PRA

FIII OSR

FIV CSR

FV WSR

SSR3 Training

.770

SSR1 Safe behaviour

.678

SSR5 Instruction

.635

.330

SSR4 Contingencies in case of unsafe behaviour

.627

.205

SSR8 Safety rules

.543

.267

.270

OSR6 Setting and machines safe

.499

.324

SSR7 Absence of supervision in case of unsafe acts (D)

.427

.394

OSR2 Contingencies in case of unsafe behaviour

.420

.237

OSW7 Safety instructions

.408

.315

.214

PRA5 Protection

.348

.279

.304

PRA4 Risk hands or arms (D)

.741

PRA7 Risk legs, foots, trunk (D)

.718

PRA3 Risk eyes, face, head (D)

.688

PRA1 Risk of accidents (D)

.665

PRA2 Risk fatal or very serious accidents (D)

.654

PRA6 Risk slight accidents (D)

.620

.235

PRA8 Risk occupational illness (D)

.591

OSR9 Priority of safety (D)

.353

.420

SSR2 Instructions in case of safe behaviour (D)

.222

.379

OSR3 Safety training

.733

OSW4 Safety training

.656

OSR8 Safety promotion

.649

OSR1 Safety inspections

.591

OSR5 Safety meetings

.566

OSW5 Safety inspections

.554

OSW1 Safety written instructions

.521

OSR7 Safety contingencies

.488

.360

-.210

OSW2 Safety contingencies

.433

.272

OSR10 Safety representatives

.429

CSR7 Safety rules

.234

.689

.242

CSR6 Contingencies in case of unsafe behaviour

.224

.686

.258

CSR8 Safety support

.361

.599

CSR2 Safety information

.592

CSR4 Priority of safety

.576

CSR3 Contingencies in case of safe behaviour

.569

CSR5 Safety behaviour

.546

CSR1 No contingencies in case of unsafe behaviour (D)

.267

WSR3 Personal Protective Equipment

.665

WSR6 Self-contingencies in case of unsafe behaviour

.250

.612

WSR5 Safety communication

.215

.569

WSR2 Safety behaviour

.213

.322

.539

WSR7 Safety rules

.495

WSR4 Safety behaviour

.232

.484

OSR4 Personal Protective Equipment

.349

.473

OSW3 Knowledge about safety

.247

.335

* Items have been rearranged following their factorial saturations. Item-factor correlations less than 0.25 have been suppressed for clarity. The symbol (D) means that the item has been reversed before its consideration in the principal components analysis.

The first factor explains 8.88% of the variance in the rotated structure. Taking into account in the interpretation the saturations greater than .400, the first factor involves items SSR3, SSR1, SSR5, SSR4, SSR8, OSR6, SSR7, OSR2, and OSW7. Itincludes six items referred to the Superiors’ Safety Response and two items related tothe Organizational Safety Response. Consequently it is reasonable to consider that this first factor represents the Supervisors’ Safety Response. The coefficient alpha for these nine items was 0.80.

The second factor explains 8.83% of the variance and encompasses the items about the Perceived Risk PRA4, PRA7, PRA3, PRA1, PRA2, PRA6, PRA8 and one item about Organizational safety Response, OSR9; therefore, this factor will be called Perceived Risk of Accidents. These 8 items had a coefficient alpha of 0.80.

The third factor explains 8.36% of the variance and includes 10 items, all about the Organizational Safety Response: OSR3, OSW4, OSR8, OSR1, OSR5, OSW5, OSW1, OSR7, OSW2, and OSR10. The coefficient alpha for these items was 0.78.

The seven items in the fourth factor (CSR7, CSR6, CSR8, CSR2, CSR4, CSR3, and CSR5) were about Co-workers’ Safety Response and explained 8.26% of the variance. Their coefficient alpha was 0.82.

The fifth factor includes seven items, six of them (WSR3, WSR6, WSR5, WSR2, WSR7, and WSR4) about the self-perceived Worker’s Safety Response and one of them about Organizational Safety Response (OSR4); therefore, the factor was called Workers’ Safety Response. This last factor explains 6.80% of the total variance. The coefficient alpha for these seven items was 0.72.

Using this rather strict criterion (saturations > .400), no item saturates in more than one factor, and only three items are not included in the five factors: PRA5, SSR2, and OSW3, although each of them presents saturations greater than .300 in one or more factors. The coefficient alpha for the entire 45 items was 0.88.

CONCLUSIONS

Although considerable theoretical and methodological progress has been made in safety climate research, little attention has been paid to the consideration of safety climate results as an important resource for the improvement of safety (Niskanen, 1994). Safety climate measures have demonstrated their relevance in providing a proactive and useful indicator of safety behaviour and safety results. In order to transform safety climate measures into a useful diagnostic and intervention tool, safety climate measures should indicate operative objectives for preventive actions, and these objectives should be located inside the organization, identifying where and what should be changed. The Valencia PREVACC approach to the measure of safety climate is based on the separate identification of the agents that perform or are responsible for each safety climate statement, and a separate identification of the different safety processes involved in each safety response.

Construction is an important economic sector involved in unacceptable safety results and characterised by some special features that can affect the psychosocial and safety climate variables. The application of this diagnostic safety climate approach to the construction industry will help to develop a safer industry.

In this paper, the factorial structure of the workers’ questionnaires from the

Valencia PREVACC Battery has been analysed using an exploratory principal components approach in a Spanish construction sample. The results clearly identify the main five variables theoretically proposed: Organizational Safety Response, Supervisors’ Safety Response, Co-workers’ Safety Response, Workers’ Safety Response and Perceived Risk. These results visibly follow the theoretical structure of the questionnaire and provide support for many of the main dimensions identified in the review of the literature (Flin et al 2000; Guldenmund, 2000; Larsson, 2005; Meliá & Becerril, 2006; Seo et al. 2004). Moreover, this pattern of results matches the one described in previous studies using this instrument in other samples (Meliá, 2004b), reinforcing thus the stability of the dimensions identified. The dimensions of safety climate are important because they are involved in work-related accidents in several ways, and, therefore, they can contribute as useful safety diagnostic indicators once the agent of each safety climate and the processes involved are clearly identified. For these reasons, further analyses on these data will be performed, in order to confirm the stability of the structure using structural equations methods.

Safety climate has become a leading indicator of safety performance (Flin, et al., 2000); therefore, this paper proposes that the main safety climate dimensions should be considered in the regular assessment of psychosocial risks in order to check on the state of safety and develop effective safety programs.

ACKNOWLEDGEMENTS

This paper was developed within the HERC Project, [BIA2004-05475], which focuses on the development of a safety assessment tool for the psychosocial dimensions related to injuries in the construction sector. Financial support was provided by the Ministerio de Educación y Ciencia (España) and The European Regional Development Fund (ERDF - FEDER). The website of the HERC Project is located within the Universidad of Valencia site: www.uv.es/seguridadlaboral

REFERENCES

  • 1. Brown, K. A., Willis, P. G., & Prussia, G. E. (2000). Predicting safe employee behaviour in the steel industry: development and test of a sociotechnical model. Journal of Operations management, 18(4), 445465.
  • 2. Cox, S., & Cox, T. (1991). The structure of employee attitudes to safety: a European example. Work and Stress, 5(2), 93106.
  • 3. Coyle, I. R., Sleeman, S. D., & Adams, N. (1995). Safety Climate. Journal of Safety Research, 26(4), 247254.
  • 4. Falcione, R. L., Sussman, L. & Herden, R. P. (1987). Communication climate in organizations. In F. Jablin, L. Putnam, K. Roberts & L. Porter (Eds.), Handbook of Organizational Communication, London: Sage Publications, pp. 195225.
  • 5. Flin, R., Mearns, K., O’Connor, P., & Bryden, R. (2000). Measuring safety climate: identifying the common features. Safety Science, 34, 177192.
  • 6. Glendon, A. I., & Litherland, D. K. (2001). Safety climate factors, group differences and safety behaviour in road construction. Safety Science, 39(3), 157188.
  • 7. Guldenmund, F. W. (2000). The nature of safety culture: a review of theory and research. Safety Science, 34(13), 215257.
  • 8. Hofmann, D. A., & Stetzer, A. (1996). A crosslevel investigation of factors

influencing unsafe behaviors and accidents. Personnel Psychology, 49(2), 307-339.

  • 9. HSC/HSE. (2003). Strategic research outlook. Sheffield: HSE.
  • 10. Karjalainen, A. (2004). A statistical portrait of health and safety at work in the construction industry (Nº 16084144). Luxembourg: European Agency for Safety  and Health at Work.
  • 11. Larsson, S. (2005). Constructing Safety: Influence of Safety Climate and Psychological Climate on Safety Behaviour in Construction Industry. Göteborg: Department of Product and Production Development. Chalmers University of Technology.
  • 12. Lundholm, L. (2004). European statistics on accidents at work; The number of work accidents in the member states has it decreased in the last 10 years? with particular focus on the construction area. Swedish Work Environment Authority
  • 13. Mearns, K., Flin, R., Gordon, R., & Fleming, M. (1998). Measuring safety climate on offshore installations. Work and Stress, 12(3), 238254.
  • 14. Mearns, K., Whitaker, S. M., & Flin, R. (2003). Safety climate, safety management practice and safety performance in offshore environments. Safety Science, 41, 641 680.
  • 15. Meliá, J.L. (1998). A psychosocial causal model of work accidents. Anuario de Psicologia, 29(3): 2543.
  • 16. Meliá, J. L. (2000). Valencia PREVACC Battery (version B1 Spanish): Diagnostic Instruments for the prevention of work related accidents. Valencia: Cristobal Serrano. ISBN: 846992317X.
  • 17. Meliá, J. L. (2000). The general safety questionnaire (GSQ). In J. Maltby, C. A. Lewis, & A. Hill (Eds.), Commissioned reviews on 250 psychological test (Volume 2). Lampeter, Wales: Edwin Mellen Press. Pp 845850.
  • 18. Meliá, J.L. (2002). Battery Valencia PREVACC 2002 Diagnostic Instruments for the Prevention of Work Accidents (English, Spanish, Portuguese, and Chinese). Valencia: Cristobal Serrano.
  • 19. Meliá, J. L. (2004a). The Psychosocial causal Model of workrelated Accidents of the University of Valencia: Perspectives and new developments. In R. Mondelo, P.; Mattila, M.; Karwowski, W.;  & Hale, A. (eds.), Proceedings of the Third International Conference on Occupational Risk Prevention ORP2004. ISBN: 84 93332828.
  • 20. Meliá, J. L. (2004b). The Valencia PREVACC Battery of the University of Valencia: The assessment of the behavioural, group and organizational dimensions that affect workrelated accidents. In R. Mondelo, P.; Mattila, M.; Karwowski, W.;  & Hale, A. (eds.), Proceedings of the Third International Conference on Occupational Risk Prevention ORP2004. ISBN: 8493332828.
  • 21. Meliá, J.L. & Becerril, M. (2006). Safety climate dimensions from the “agent” point of view. In Mondelo, P; Mattila, M.; Karwowski, W. & Hale, A. (Eds.) Proceedings of the Fourth International Conference on Occupational Risk Prevention. Sevilla.
  • 22. Niskanen, T. (1994). Assessing the safety environment in work organization of road maintenance jobs. Accident Analysis and Prevention, 26(1), 2739.
  • 23. Olive, C., O'Connor, T. M., & Mannan, M. S. (2006). Relationship of safety culture and process safety. Journal of Hazardous Materials, 130, 133140.
  • 24. Seo, D.C., Torabi, M.R., Blair, E.H., & Ellis, N.T. (2004). A cross validation of safety climate scale using confirmatory factor analytic approach. Journal of Safety Research, 35(4), 427445.
  • 25. Silva, S., Lima, M. L., & Baptista, C. (2004). OSCI: an Organisational and Safety Climate Inventory. Safety Science, 42(3), 205220.
  • 26. Smith, G. S., Huang, Y. H., Ho, M., & Chen, P. Y. (2006). The relationship between safety climate and injury rates across industries: The need to adjust for injury hazards. Accident Analysis and Prevention, (in press).
  • 27. Williamson, A. M., Feyer, A.M., Cairns, D., & Biancotti, D. (1997). The development

of a measure of safety climate: The role of safety perceptions and attitudes. Safety

Science, 25(1-3), 15-27.

28. Zohar, D. (1980). Safety Climate in industrial organizations: theoretical and applied implications. Journal of Applied Psychology, 65(1), 96-102.