Larsson, Susanna ; Pousette, Anders; Törner, Marianne National Institute of Working Life West/ P.O. Box 8850/ SE-402 72 Gothenburg, Sweden
A construction project in Sweden, was studied, exploring relations between safety climate, psychological climate, safety determinants and psychological determinants on one hand, and safety behaviour on the other. Results revealed a high number of significant relations between the dimensions, implicating the importance of not only having the specific safety climate in focus, but also the individual perceptions of the psychological work environment, the psychological climate.
Differences were found between white collar and blue collar workers. Differentiating between the blue collar and white collar workers is of importance when designing and evaluating studies.
Safety climate, psychological climate, safety determinants, psychological determinants, safety behaviour, construction industry, white collar workers, blue collar workers
Building and construction industry is an area of working life that has a high rate of serious accidents and injuries. Preventive work has mainly been directed towards passive prevention. The importance of integrating psychological, social and organisational factors in the prevention of accidents has been emphasized by e.g. Cox, Griffiths and Houdmont  in their research in rail safety.
The awareness of organisational, managerial and human factors as interacting causes of accidents has resulted in extensive but straggling research on culture and climate related to safety, as shown in reviews by e.g Guldenmund . There iscontroversy over whether principal dimensions of a good safety culture may at allbe identified, and under any circumstances, such dimensions are difficult to access. Viewing safety climate as the groups shared perceptions of conditions within the organisation and with importance to safety [3,6,7], identifying main dimensions of safety climate may be more accessible. The views on what constitute principal dimensions of safety climate are starting to converge into a core of a limited number of dimensions. Flin, Mearns, O’Connor and Bryden  in a review of 18 studies, identified five main themes of safety climate, namely management/supervision, safety systems, risk, work pressure and competence, where the first three were most common. Other dimensions, also found to be of importance, have been communication, individual safety responsibility/motivation,
teamwork/group involvement in safety  and incentives for safety, such as positive feed-back and effects of safety involvement on the possibility for promotion and social status 
The concept of safety climate has its basis in the theory on organisational climate  and climate has long been regarded as important in studies of organisationalbehaviour [4,13]. Glick [6,7] stated that organisational climate had been used torefer to a broad class of organisational and perceptual variables that reflect individual-organisational interactions, affecting individual behaviour. Regarding the multilevel issues and differences of conceptual meaning in the use of these variables, the same author suggested making a distinction between organisational climate and psychological climate. Concerning theory, measurement and analysis, the organisational climate label should refer to the organisation as the unit, while the psychological climate label should refer to the individual as the unit. Glick, W.
H.  stated that organisational and subunit climates should be seen as providing the context in which psychological climate may be understood. Regarding research on safety, psychological climate has not widely been taken in to consideration. Psychological climate, referring to how individuals in the organisation perceive their personal psychosocial conditions (psychosocial work environment), has been shown to have reliable relationships with employee attitudes to work, psychological well- being, motivation and performance . In their review of the literature concerning psychological climate perceptions, Parker, Baltes, Young, Huff, Altmann, Lacost and Roberts  adhered to five psychological climate dimensions; job characteristics , role characteristics, workgroup characteristics, leadership characteristics and organizational attributes, originally provided by Jones and James . The dimension of organisational attributes were, however, not included in later work bye. g. James and McIntyre . Jones and James  described job characteristics by degree of autonomy, variety, feedback, challenge, importance and pressure, while their description of role characteristics referred to degree of ambiguity/clarityand conflict. Leadership characteristics were defined by degree of support, goalemphasis, planning, interaction and confidence while degree of effectiveness, friendliness and co-worker support constituted the workgroup characteristics. To distinguish psychological climate from the psychological determinants, commitment to workplace and job satisfaction, the same authors pointed at the descriptive and cognitive nature of climate, differing from the evaluative and affective aspects of the determinants.
The construct of psychological climate contains dimensions, i. e. role clarity, role conflict and job demands, are in stress related research referred to as occupational stressors [16,26] Lawton and Parker  stated that the effects of stress on cognitive performance requires further research, to clarify the role of subjective perceptions of stress versus reactions to the experience of stress, and the effects of these conditions on accidents. Goldenhar, Williams and Swanson , in their research in construction safety, found job demands to be significantly related to near-misses, with physical symptoms as mediating mechanism.
Safety behaviour, as an outcome parameter of safety climate, has been dealt with by e. g. Cheyne, Cox, Oliver and Tomás  and Neal, Griffin and Hart . Pousette, Törner and Larsson  identified three different aspects of self-rated safety behaviour among construction workers. Structural safety behaviour (SSB), concerning participation in organized safety activities (e.g. taking part in risk assessment); Interactional safety behaviour (ISB), concerning safety activities in the daily work in interaction with co-workers and management (e.g. discussed a safety issue with a fellow worker) and Personal safety behaviour (PSB) comprising behaviour aimed at personal protection (e.g. use of all prescribed protective equipment, following safety rules).
The aim of the present study was to explore the relations between certain dimensions and the three types of self rated safety behaviour (SSB, ISB, PSB)
among blue collar workers and white collar workers in Swedish construction industry.
Six dimensions of Safety climate, eleven dimensions of Psychological climate (of which three in literature are considered as important stressors), two dimensions of
Safety determinants and three dimensions of Psychological determinants were studied.
The study was based on cross-sectional data from the second wave of a longitudinal research project studying safety issues within a major construction project of building a road tunnel under central parts of Gothenburg, Sweden.
A questionnaire distributed to all personnel involved in the construction project at the time of measurement (N=291) resulted in 276 valid responses, representing 164 blue collar construction workers and 112 white collar workers, giving a total response rate of 95%.
The white collar workers were managers, supervisors, specialists and administrative personnel and 63% had a supervisory position. The blue collar worker groupconsisted of concrete workers, miners, electricians, painters, carpenters, welders,road construction workers, pipe layers, plate workers, truck drivers and machine operators.
The blue collar worker mean age was 44,1 years (SD= 11.2), mean job tenure 19.8 years (SD= 11.9) and all were males.
The mean age of the white collar workers was 42.1 years (SD=12.7), mean jobtenure 14.9 years (SD=13.0) and 91% were males.
The respondents background were categorised into one of three levels of education: elementary/compulsory school, upper secondary school/apprentice orcollege/university. The blue collar worker group had, as would be expected, asignificantly lower level of education than the white collar worker group(P