Cet article porte sur les approches utilisées dans les organisations pour amener les superviseurs à adopter un mode de gestion plus participatif de la sécurité au travail avec leurs employés. Les données présentées sont basées sur cinq études de cas réalisées selon un devis commun et analysées de façon comparative. Utilisant l'approche systémique du changement organisationnel, les auteurs développent le concept de stratégie de changement pour analyser les approches organisationnelles observées, en faisant l'hypothèse que ce sont les stratégies plus systématiques qui auront comme effets d'accroître l'utilisation de la gestion participative de la prévention par les superviseurs, et de réduire le taux de fréquence des accidents. Les résultats confirment largement les hypothèses, mais l'analyse montre aussi que les stratégies non efficaces, au plan des effets escomptés, ont néanmoins une fonction d'utilité qui est discutée.
Over the last few decades, various countries have adopted laws and regulations fostering joint regulation and labour participation approaches in occupational heath and safety through joint health and safety committees, safety representatives, and workers' rights to be informed about work hazards and to refuse dangerous work. In order to succeed, these mechanisms should not remain isolated, but should be accompanied by larger organizational changes in labour relations, union and management practices. This article is about one of these organizational changes, namely, the transformation of supervisory management practices in occupational safety towards a more participative approach.
Concepts and Hypotheses
Getting supervisors to adopt a participative management approach in occupational safety means changing their tasks and their way of doing them. From the perspective of a Systems theory of organizational change, it can be hypothesized that such a change will succeed only if consistent changes are brought about in other major components of the organizational System, namely : the individuals and groups involved; the formal organizational structures supporting and controlling individuals in their tasks; and the informal political and cultural dynamics of the organization. Each of these other components may present problems during the change process: resistance to change from individuals and groups; loss of control from formal structures; or loss of support from power groups and fallure of adjustment of cultural corporate values.
The concept of organizational change strategy is used to designate all actions taken in the organization to address the aforementioned problems. Consequently, change strategies, which are the independent variable in this study, may be more or less developed and systematic. It is hypothesized that when the change strategy is more developed and systematic, (1) supervisors will use a more participative approach in managing prevention activities (inspection, task analysis, safety meetings, design of corrective measures, etc.) that is, getting their employees to participate in these activities, and (2) the lost-time accident frequency rate will be reduced. Variations in participative supervisory management of occupational safety and lost-time accident frequency rate are, therefore, the dependent variables in the study. A third hypothesis is that a greater use of safety participative management by supervisors should be related to a lower accident frequency rate. Figure 1 in the article illustrates this conceptual model and the hypotheses.
The study was conducted in five industrial firms located in the province of Quebec, Canada. These firms were selected from a larger sample, using a quota method, in order to include (1) different industrial sectors, (2) different firm sizes, (3) unionized and non-unionized firms, and (4)various geographical locations. In each firm, data were collected through (l)semi-structured interviews (n = 55) with supervisors, workers' representatives, top and middle managers, safety managers and members of the health and safety committee, and (2) questionnaires filled out by all supervisors (n = 63) and various other managers (n = 29).
The development level of each change strategy (the independent variable) was measured using a scale (1-10) theoretically constructed by defining a total of ten dimensions referring to the three major change problems (resistance, control, and power and culture) that should be addressed and can be solved by consistent action steps during the change process. Data on actions taken, means and measures used by firms between 1988 and 1992 to encourage supervisors to use safety participative management, were collected and analyzed according to these ten dimensions.
Variations in the use of participative safety management by supervisors (the dependent variable) were estimated (+/-) using data collected from various sources, including supervisors, about the evolution of the latter's management practices between 1988 and 1992. The last dependent variable, variation in the lost-time accident frequency rate, was measured by comparing the rates between the first and last year of a given strategy during the period under study.
Table 2 presents the main results of the study. On the left side of the table, each of the ten strategies observed is identified, its level of development is indicated, and the number of dimensions for which action steps were taken is mentioned for each of the three major problems of the change process. On the right side, data is provided about dependent variables. Generally speaking, results largely support the hypotheses. More developed and systematic strategies (A-2, D-2, D-4, E, B, C-2) are all followed by an increase in the supervisors' use of participative safety management and a decrease in the accident frequency rate, while three out of the four less developed and systematic strategies (A-l, C-l, D-3) have inverse results.
Discussion and Conclusion
Although the results provide strong support for the hypotheses and confirm the results of previous studies about the positive impact of participative supervisory management on accident rates, the case studies also shows that non-effective strategies have had an important function in the social construction of effective strategies of organizational change. In brief, all non-effective strategies are characterized by an important unsolved problem of supervisors' and workers' resistance to the change in the safety practices expected by the organization. It appeared that this resistance is not only psychosocial, as generally conceptualized in the Systems theory of organizational change, but is also sociological in terms of a strategy used by supervisors and work teams to negotiate their place in the new "organizational order" which is constructed in the change process. Actually, this social process of "implicit bargaining" appears to be the main mechanism for the social construction of more effective strategies, that is, strategies addressing the resistance problem by developing change ownership by supervisors and workers, the power and culture problem by raising the top management commitment in the change process, and the structural support problem by adapting formal Systems of management in a manner consistent with the desired change.