This study was a meta-analysis that systematically evaluated evidence on the effects of workplace safety training interventions on: safety performance antecedents (safety motivation, knowledge, climate) and safety performance (safety compliance and participation).
This study was a meta-analysis that systematically evaluated evidence on the effects of workplace safety training interventions on: safety performance antecedents (safety motivation, knowledge, climate) and safety performance (safety compliance and participation).
This study was a meta-analysis that systematically evaluated evidence on the effects of workplace safety training interventions on: safety performance antecedents (safety motivation, knowledge, climate) and safety performance (safety compliance and participation).
This study was a meta-analysis that systematically evaluated evidence on the effects of workplace safety training interventions on: safety performance antecedents (safety motivation, knowledge, climate) and safety performance (safety compliance and participation).
90 published studies were included. In conjunction with the above, they also studied the effect that the industry risk level has on the interventions (i.e. high-risk vs low-risk industries), and, curiously enough, also risk homeostasis theory.
Risk homeostasis theory (RHT) is where “risk-taking behavior is formed homeostatically by comparing a subjectively accepted degree of risk (target risk value) and an objective risk that is perceived in the environment … [and] people compare the actual risk they perceive in a situation with the degree of risk that they are willing to accept. The degree of risk that respondents are willing to accept is called the target risk value” (p2-3). Based on the target risk, people will adapt their behaviour accordingly to align with their acceptable risk.
The link between RHT and safety training is hypothesised to be such that in high-risk environments, people will naturally be more sensitive to the inherent dangers and thus, training interventions will be less effective. [Note: RHT is highly contested and evidence isn’t that great supporting the concept]
Also, the authors highlight why they didn’t include an analysis of accidents and incidents. Although incidents analysis is often the first aspect of safety that people look at “it is generally one of our most distal outcomes and is difficult to directly assess as a result of a generally low base rate” (p2). Consequently, according to the authors there are few studies which have studied the relationship. [I think the former is an important point: we often look to incidents as the gold standard of measurement in safety but they are actually quite rare and distal outcomes, whereas there are a range of upstream inputs which influence performance which are more common and more readily evaluated.]
Furthermore, quoting the paper “a lack of accidents is not a sufficient indicator of a safe work environment; instead, a high prevalence of accidents in the workplace is better understood as an indicator of a lack of safety within an organization” (p2).
Results
Across the meta-analytical results, safety training was found to statistically improve all of the studied outcomes, although there was a wide dispersion among the effect sizes, making it difficult to narrow down specific factors (e.g. the context of the study, its participants, the training methodology etc. impacted the outcomes).
Training interventions had the strongest impact on safety knowledge, then followed by safety motivation and then safety compliance (safety compliance being following rules and other established means) activities. The impact on climate and safety participation (this being behaviours more focused on proactive initiatives and taking the lead with safety rather than compliance-focused) were weaker than the other factors.
In explaining why training interventions had the weakest impact on improving safety participation, the authors suggest that it takes more volitional effort to go above and beyond in proactive initiatives compared to simply following company mandated rules.
Mixed results were found with RHT. Consistent with predictions, the effect of training on safety performance in low-risk industries is stronger than in high-risk industries. But in contrast, the impact on safety antecedents (motivation and safety climate) is stronger in high-risk industries than low-risk. The authors suggest that this indicates there’s other theoretical explanations for how interventions affect attitudinal outcomes and that these weren’t modelled in their study. Not surprisingly it’s concluded that “Homeostasis Theory may not be suited to capture the emergence of attitudes as a result of safety interventions” (p8).
The authors suggest that sensemaking theory, such as from Weick, may be better placed to help explain some of these relationships, such as why training had a bigger impact on attitudinal variables in high-risk industries. It’s said that since the operating environment in high-risk environments are more hazardous, it’s easier for participants to align the training to their own sensemaking experience of the environment. It was found that although training in the high-risk industries was “more intensive in terms of hours of actual training”, the intensity didn’t significantly explain variance of the safety interventions.
Thus, the authors state that “risk, but not training intensity, results in greater gains to motivation and safety climate in high-risk industries” (p9).
For practical application, it’s said that training should be developed cognisant of the industry risk-level and develop interventions in alignment with that (e.g. if it’s a low-risk industry then people will have trouble aligning with the content if you’re teaching them about fatal risks, so you need to put more thought into the sensemaking process). According to the authors, “in low-risk industries interventions researchers and practitioners should increase sense-giving efforts of explaining the gap between the actual risk the employee experience and the need to invest time and energy in prevention of accidents” (p9).
Authors: Derek Hutchinson, Gil Luria, Shani Pindek, Paul Spector, 2021, Safety Science
Study link: https://doi.org/10.1016/j.ssci.2021.105594
Link to the LinkedIn article: https://www.linkedin.com/pulse/effects-industry-risk-level-safety-training-outcomes-ben-hutchinson
SafetyInsights.org 