Evidence Base
Evidence Summary
Safety TrainingModerate

The Science of Safety Training: What Works and What Doesn't

3
Meta-Analyses
123
Studies
21K
Participants
The Question
Safety training is one of the largest recurring investments in workplace safety. It is also one of the least scrutinised. Most organisations default to classroom lectures, e-learning modules, or toolbox talks — methods chosen for scalability and cost rather than effectiveness. If you are responsible for safety training spend, you need to know: which training methods actually reduce injuries? Is there a point at which more training stops helping? And why does so much safety training fail to change behaviour beyond the training room?
Key Findings
Overall training effect (d)
SMLd = 0.60
Cohen's d (d)
Standardised effect size expressing the difference between groups (.20 small, .50 medium, .80 large)
The Bottom Line
The evidence on safety training effectiveness reveals an uncomfortable gap between common practice and what actually works. Across 123 studies and over 20,000 participants, the research is clear: as training methods become more engaging and participatory, safety outcomes improve. Behavioural modelling and hands-on practice produce the largest effects. Lectures, pamphlets, and passive e-learning — the methods most organisations rely on — are the least effective. And mandatory training consistently underperforms voluntary training.
What You Can Do
1
O
Shift training budget from passive to active methods
2
O
Build behavioural modelling into your core safety training
3
O
Redesign mandatory training for engagement
4
O
Use shorter, more frequent training modules
Intervention Level:
I
Individual
G
Group
L
Leader
O
Organisation
April 11, 20266 min read · Full article at evidencebase.app

Tip: Save the PNG first, then attach it to your LinkedIn or X post for maximum visual impact.