Evaluative Research on a Union Training Program About Health and Safety at Work

The overall objective of the evaluative research project reported here was to determine the extent to which a health and safety training program offered by a Quebec labour federation produced expected outcomes or equivalently achieved its objectives. The ultimate objective of the fifteen-hour training program, delivered over three days, was to improve the health and safety interventions of the federation's members. This report discusses the results concerning occupational accidents, one of the themes of the training program.The study population was composed of members of the federation's unions who were eligible for the training program. A pre-test/post-test study design with a non-equivalent control group was used, and subjects were recruited using a clustered sampling strategy. The exposed group was composed of 65 union representatives who participated in one of four three-day health and safety training programs between February and November 1999. The control group was composed of 60 union representatives who had participated in three-day training programs on noise, work organization (2 groups) or the role of union representatives (3 groups) between November 1997 and February 2000. In all cases, the pre-test was administered at the beginning of the first day of the training program and the post-test administered at the end of the last day. The objectives of the accident module of the health and safety training program were to improve participants' understanding of the multiple causes of occupational accidents, improve their ability to identify these causes and accident-prevention measures, and enable trainees to attribute occupational accidents to environmental hazards for which employers are legally responsible. Data was collected from subjects using a self-administered questionnaire that included questions related to the expected outcomes of the training program. The questionnaire also provided data on the subjects and this was used to control for selection biases that could compromise the internal validity of the study.The dependent and control variables were subjected to descriptive analyses, and the groups were compared using bivariate analyses. The effect of exposure to the training program on each of the six expected outcomes of the program was estimated using multivariate analysis. The independent variable in these analyses was exposure to the training program, measured as a dichotomous variable (exposed/control), while the six dependent variables were the subjects' post-test scores for each of the expected program outcomes. In order to estimate the net training effect, i.e. the difference between post-test and pre-test knowledge and skills, the pre-test results were used as control variables for each of the six program outcomes. In addition, the three variables related to health and safety competence, acquired before program exposure, were also used as control variables, in order to control for biases due to interaction between selection and test habituation.The response rates for the exposed and control groups were 65% and 70%, yielding exposed and control groups of 42 subjects each. The only variable exhibiting a statistically significant association with the attribution of accidents to worker-related causes during the post-test was exposure to the training program. In fact, participation in the training program reduced the probability of such attribution by 85.9% (odds ratio : 0.141). Both pre-test results and participation in the training program affect the probability that subjects attribute accidents to environmental causes. After controlling for pre-test results, this probability was four times greater in the group taking the program than in the control group. The training program therefore appears to have attained one of its objectives, namely to increase participants' understanding of the role that environmental factors play in occupational accidents.No association was observed between any variables and the identification of individual measures that would prevent accidents during the post-test. However, a statistically significant association was observed between the identification of organizational measures that would prevent accidents and both pre-test scores and participation in the training program. After controlling for pre-test results, subjectshaving participated in the training program were fourteen times more likely to identify organizational measures.A statistically significant association was observed between three variables and the percentage of responsibility attributed to employers for accidents. These three variables account for 30.4% of the total variation in the percentage of responsibility attributed to employers. Participation in the training program is the variable that appears to exert the greatest effect, followed by pre-test scores and exposure to other occupational health and safety training. Furthermore, a statistically significant association was observed between two variables and the percentage of responsibility attributed to workers for accidents. These two variables account for 17.8% of the total variation in the percentage of responsibility attributed to workers. Responsibility of workers was negatively associated with participation in the training program and positively associated with pre-test scores. These results appear to indicate that the union representatives who participate in the training program are more likely to attribute accidents to external causes.The content of the health and safety training program is meant to be directly relevant to union activities recommended by the labour federation. Rather than technically proficient health and safety specialists, the program aims at producing individuals who can mobilize workers to establish a balance of power with their employers. This then allows them to participate in activities designed to eliminate hazards at their source. To achieve this, union federation officers believe it essential to modify the attribution of occupational accidents by their members, since internal attribution is likely to result in a scapegoating of accident victims, which in turn hinders the implementation of prevention measures designed to eliminate occupational accidents at their source.These results indicate that the accident module of the training program achieves most of its expected outcomes, and no program modification is therefore recommended. The study allowed the development of a measurement technique that the labour federation can use to evaluate the expected outcomes of its program and to identify union members who would benefit from the program.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.