Improving Safety at Work: the Role of Climates

Main organizational and psychological factors considered by literature about safety concerned the concept of safety climate - defined as individual perceptions about the policies, practices, and procedures related to safety issues affecting well-being at work - and that one of safety behaviors, represented in a dichotomy that comprises both safety compliance and safety participation behaviors. Literature review suggested that the main trend of research in that field has been the focus on the ability of safety climate to predict safety behaviors, not considering some important challenges for research development in this area. The main challenge concerns the influence of other variables related to the social context on safety issues; for instance, the existence of multiple simultaneous organizational climates with different effects on the same safety outcomes. Therefore, the first aim of this PhD thesis was to consider a broader approach when studying organizational dynamics pertaining to safety and to test whether other dimensions of organizational climate can be strictly relevant for safety even if they are not specific to safety. In particular, the increasing presence of diverse workforce in the organizations requires to be taken into account when studying the relationship between safety climate and safety performance; indeed, if companies want to take the most from today’s ever-diversifying markets, they will have to hire, coordinate, and retain a broad base of human resources, with consequences for organizational and group performance, even specifically that one pertaining to safety. For that reason, the first study of the present dissertation has been focused on developing an Italian validation of a useful instrument to measure Diversity Climate, by adapting the Mor Barak, Cherin and Berkman (1998) Diversity Perceptions Scale, the most widely used questionnaire to assess diversity climate, considering the lack of any validated measure of diversity climate in the European context. Results interestingly showed a different factorial solution for the Italian sample using a shorter version than the original scale; moreover the scale seemed to be invariant for both men and women and to show good discriminant and criterion validities. Therefore, this first study allowed us to make this instrument suitable for the Italian context, in order to use it for evaluating whether diversity climate could specifically affect some safety outcomes in the subsequent studies. In the second study the diversity climate measure was used together with a measure of climate for inclusion (Mor Barak, 2005) to test their impact on safety participative behaviors at the individual level and for the first time, by hypothesizing a mediating role performed by the motivation to actively promote safety at work. The choice to consider also the construct of inclusion climate as linked to safety is due to the particular kind of social empowerment and involvement – regardless of workers’ diversities - which form this kind of climate; moreover it has been already demonstrated to influence work quality and behaviors, so we hypothesized that it could also have positive consequences for workers’ involvement and commitment to safety and their safety performance. Results showed that diversity climate and climate for inclusion are different constructs which have different and important impacts on safety; specifically safety participation motivation revealed to fully mediate the relationship between diversity climate and safety participation behaviors, whereas it only partially mediated the relationship between inclusion climate and safety participation behaviors. The third aim of the present dissertation was to enlarge the effort of investigating the impact of distinct facet-specific climates on specific safety behaviors, by exploring the possible influence of multiple existing climates on injury rates at the unit-level and with a qualitative statistical technique - multi-correspondence analysis. In addition to diversity climate and inclusion climate, safety climate (e.g., Zohar, 1980; 2000) and communication climate (Smidts, Pruyn & van Riel, 2001) measures were also used for this purpose, as previous research stressed how safety performance is strictly dependent on the supervisors’ and coworkers’ communication and quality of social exchange. Results showed the presence of four differentiated patterns of climates; however, injury’s distribution did not seem to be coherent with the climates’ patterns, and references to the halo effect (Wells, 1907; Thorndike, 1920) and to the injury “reporting climate” (Probst & Estrada, 2010, p. 1443) have been made to explain those findings. Summarizing the results of the three studies, it is possible to conclude that they represents very innovative contributions to the organizational climate literature, and with regard to the safety area, they explore research avenues not addressed before.