Safety Culture Assessment: How to Measure OHS Maturity in Your Organization
The Company With Perfect Records and a Fatal Incident
A UK petrochemical company maintained exemplary safety documentation. Incident reports were filed promptly. Training records were complete. Management attended every quarterly safety review. The formal safety management system was genuinely well-implemented.
Then a serious incident occurred — a fatality during a maintenance operation that should have been low-risk. The investigation found that the maintenance crew had observed the abnormal condition that preceded the incident on two previous occasions and had not reported it, because "reporting things that hadn't caused a problem yet" was not how things worked in that facility.
The documentation was excellent. The culture was not. Workers had learned that safety reporting was welcomed in meetings and tolerated on forms, but not rewarded — and sometimes subtly penalized — in the daily reality of the workplace.
Safety culture is the set of values, beliefs, attitudes, and behaviors that determine how safety is actually practiced in an organization — not how it is documented. Assessing it requires different methods than auditing a management system.
What Safety Culture Is — and What It Isn't
Safety culture is often confused with safety climate, safety management systems, and safety performance. These are related but distinct:
| Concept | Definition | How It's Measured |
|---|---|---|
| Safety culture | Shared values, assumptions, and beliefs about safety that shape behavior | Qualitative methods: interviews, observation, focus groups |
| Safety climate | Workers' perceptions of safety at a point in time | Quantitative surveys (safety climate scales) |
| Safety management system | Documented processes, procedures, and controls | Audits, documentation review |
| Safety performance | Outcomes — incident rates, injury severity | TRIR, LTIR, near-miss rates |
A safety management system describes what should happen. Safety culture determines what actually happens when no one is watching.
Safety Culture Maturity Models
Several maturity models provide a framework for assessing where an organization sits on the safety culture continuum. The most widely used is the Parker-Hudson Safety Culture Ladder (also known as the Hearts and Minds model):
| Level | Name | Characteristics |
|---|---|---|
| 1 | Pathological | "We don't have a safety problem" — safety is a nuisance; compliance driven only by external pressure |
| 2 | Reactive | Safety is addressed after incidents happen; rules are enforced after accidents |
| 3 | Calculative | Systems are in place; compliance is monitored; "we manage safety by the numbers" |
| 4 | Proactive | Safety is anticipated; workforce is actively engaged; leading indicators are used |
| 5 | Generative | Safety is integrated into how everything is done; workers at all levels take ownership |
Most industrial organizations sit at Level 3 (Calculative) — compliance is managed, but the deeper engagement that characterizes Levels 4 and 5 is absent. The gap between Level 3 and Level 4 is the gap between managing safety documentation and actually preventing incidents.
Dimensions of Safety Culture
Safety culture is not a single dimension — it has multiple components that can develop at different rates:
| Dimension | What It Reflects | Indicators |
|---|---|---|
| Leadership commitment | Whether leaders demonstrably prioritize safety over production when they conflict | Management behavior during incidents; resource allocation decisions |
| Worker involvement | Whether workers are genuinely involved in safety decisions, not just informed | Near-miss reporting rates; participation in hazard identification |
| Reporting culture | Whether workers feel safe reporting errors, near misses, and concerns without fear of blame | Near-miss report frequency; ratio of near misses to injuries |
| Learning culture | Whether the organization systematically learns from incidents and near misses | CAPA closure rates; recurrence of identified issues |
| Just culture | Whether the response to error distinguishes between honest mistakes and reckless behavior | Whether disciplinary action is applied fairly and consistently |
| Communication | Whether safety information flows freely up and down the organization | Whether workers know about hazards; whether management knows about concerns |
How to Assess Safety Culture
Method 1: Safety Climate Surveys
Safety climate surveys are standardized questionnaires that measure workers' perceptions across the dimensions above. Validated instruments include the Nordic Safety Climate Questionnaire (NOSACQ-50) and the Safety Climate Tool (IOSH/HSE).
What they measure well: Point-in-time perceptions; comparison over time; differences between departments or sites
Limitations: Self-report bias; "say safe" pressure; may reflect desired rather than actual behavior
Best practice: Administer anonymously; report results transparently (including negative findings); close the feedback loop — show workers what changed based on their responses
Method 2: Structured Interviews and Focus Groups
Qualitative methods explore the underlying assumptions and beliefs that drive behavior. Questions explore:
- "When you see something unsafe, what do you do? What stops people from reporting?"
- "How do you see your supervisor responding when safety and production conflict?"
- "What happened the last time someone reported a near miss or error?"
- "What would you do if you thought a task wasn't safe?"
What they reveal: The stories workers tell about how safety actually works; the unwritten rules; the gap between formal and informal systems
Limitations: Time-intensive; requires skilled facilitation; can be influenced by group dynamics
Method 3: Behavioral Observation Programs
Structured observation programs (STOP, BBS — Behavior Based Safety) involve systematic observation of work behaviors in the field:
- Percentage of observations where safe behaviors are exhibited
- Types and frequency of at-risk behaviors
- Differences between observed and self-reported behavior
Limitations: Observer effect (workers behave differently when being watched); requires significant training; prone to "gaming" if tied to targets
Method 4: Document and Record Analysis
Indicators embedded in existing records:
- Ratio of near misses to recordable injuries (high ratio = healthy reporting culture)
- CAPA closure rates and recurrence rates
- Supervisor involvement in incident investigations
- Training completion rates and follow-through
- Inspection finding rates and trends
Method 5: Triangulation
No single method reliably assesses culture. The most valid assessments combine:
- Quantitative survey data (what workers say)
- Qualitative interview data (what workers believe)
- Behavioral observation (what workers do)
- Record analysis (what actually happens)
Where these data sources diverge, the divergence itself is informative.
Key Safety Culture Metrics
| Metric | What a Healthy Culture Looks Like |
|---|---|
| Near-miss to injury ratio | 300:1 or higher (Heinrich's triangle); low ratio suggests under-reporting |
| Near-miss reporting rate | Increasing or stable; declining rates warrant investigation |
| CAPA closure rate | ≥90% within defined timeframes |
| Management safety walk frequency | Regular, documented, with visible follow-up |
| Safety training completion | ≥98% with genuine understanding (not just completion) |
| Safety climate survey score | Trending positive over assessment periods |
| Repeat findings in inspections | Declining — indicates learning and systemic correction |
Common Safety Culture Failure Patterns
The paper tiger: Management systems are sophisticated; actual behavior is not. Procedures exist for everything; no one follows them in practice.
The blame culture: When incidents occur, the response focuses on who did something wrong rather than what system allowed the error to occur. Workers learn to hide errors rather than report them.
The normalization of deviance: Small deviations from procedure become routine because nothing bad happens immediately. Each deviation that doesn't result in an incident becomes justification for the next deviation. Until something does happen.
The production-safety trade-off: Workers observe repeatedly that when safety and production are in genuine conflict, production wins. They learn the actual priority order — regardless of what is stated in the safety policy.
The feedback loop failure: Workers report hazards; nothing changes. Workers report near misses; no one follows up. Over time, reporting rates decline. Management concludes safety culture is declining when the cause is management's own unresponsiveness.
Improving Safety Culture: What Works
Safety culture improvement is slow and nonlinear. Organizations that successfully move from Calculative to Proactive or Generative share several characteristics:
Visible, authentic leadership commitment. Leaders who conduct regular safety walks — not to check compliance, but to ask questions and listen — and who follow up on what they hear. Workers can tell the difference between a performative safety tour and a genuine commitment.
Closing the feedback loop on reports. Every hazard and near-miss report should receive visible follow-up. When workers see that reports lead to changes, reporting rates increase.
Investigating near misses and hazards as rigorously as injuries. A near-miss investigation that produces a genuine learning document has more safety value than an injury investigation that concludes with "remind workers to follow procedures."
Using data actively, not for performance. Culture improvement happens when safety data is used to understand and address systemic issues — not to calculate whether targets were met.
Building psychological safety. Workers report errors and concerns when they trust that the response will be constructive, not punitive. Demonstrating a just culture — distinguishing between honest errors and reckless behavior — is the foundation.
How FindRisk Supports Safety Culture Development
Near-miss and hazard reporting: FindRisk enables workers to report hazards and near misses directly from their mobile device — with photos and structured information — before the moment passes. Lower friction = higher reporting rates.
Trend visibility: Inspection data from FindRisk is automatically aggregated across inspections, areas, and time periods. Trend visibility is a prerequisite for systemic learning — organizations cannot address what they cannot see.
Corrective action follow-through: CAPA tracking in FindRisk ensures that reported issues are assigned and tracked to closure. Visible follow-through builds trust in the reporting system.
Management participation: Documented management safety walks and leadership inspections in FindRisk create a visible record of leadership commitment — and a feedback loop to workers.
Frequently Asked Questions
How long does it take to improve safety culture?
Safety culture change is measured in years, not months. Research on organizational culture change consistently shows that genuine culture shifts — where values and behaviors change, not just compliance — typically require 3–5 years of sustained effort. Initiatives that promise rapid culture change are typically improving safety climate (perceptions) rather than culture (underlying values and beliefs). Climate improvements are valuable but less durable without deeper culture change.
Can safety culture be measured reliably?
Safety culture assessment methods have known limitations, and no single method produces a definitive culture score. The most reliable assessments combine quantitative and qualitative methods across multiple data sources and are conducted by assessors independent of the management being assessed. Even with these limitations, systematic assessment is far more useful than no assessment — because it identifies specific dimensions where the gap between documented intent and actual practice is largest.
Is there a relationship between safety culture scores and incident rates?
The research evidence is consistent: organizations with stronger safety cultures (measured by validated climate surveys and behavioral indicators) have lower incident rates. The relationship is not one-to-one — safety culture is one of several factors influencing incident outcomes — but the effect size is significant. Organizations in the top quartile of safety culture typically have TRIR rates 30–50% lower than industry average.
What is the difference between safety culture and safety management systems?
A safety management system is the set of documented processes, procedures, and controls that specify how safety should be managed. Safety culture is how safety is actually managed in practice. The gap between the two — between what the procedures say and what people actually do — is the measure of culture-system alignment. High-performing organizations have both strong systems and strong culture. A strong system without strong culture produces compliance that disappears when oversight is absent.
Conclusion
Safety culture assessment is not an alternative to managing safety systems — it is the layer above those systems that determines whether they actually work.
The organizations with the lowest incident rates are not necessarily those with the most sophisticated management systems. They are those where workers at every level genuinely understand the hazards, actively report what they observe, and trust that the system will respond.
Reaching that state requires leadership behavior, not just leadership statements. It requires closing feedback loops, not just opening reporting channels. It requires treating near-miss and hazard reports as valuable data, not administrative exercises.
Download FindRisk to support the behavioral foundation of a strong safety culture — structured inspections, near-miss documentation, hazard reporting, and transparent corrective action tracking from any mobile device in the field.
Try FindRisk
Ready to modernize your safety workflow?
Conduct AI-powered risk assessments, generate reports instantly, and keep your team safe — anywhere, anytime.