Most EHS teams measure training effectiveness in completion rates and quiz scores. Three years of incident data from 41 sites shows neither metric correlates meaningfully with what comes next.
The correlation that's actually there
What does correlate with reduced incidents: time-to-first-training for new hires, refresher cadence for high-risk roles, and the gap between near-miss reporting and the next training delivery.
What doesn't correlate
Quiz scores. Completion rates. Training hours per employee. We expected at least one of those to track with TRIR, and none did at statistically meaningful levels across the sample.
What this means for your program
If you're justifying your training budget on completion rates, you're solving for the wrong number. Tie training velocity to incident velocity. Train fast after a near-miss. Refresh on cadence for the roles where exposure is highest.
