4 Myths About Safety Culture Surveys That Leaders Still Believe

A safety culture survey can give leaders a clear score and still leave the most important question unanswered. Do people experience the controls, decisions, and responses that the organization says it values when production changes, staffing is thin, or a supervisor must choose between escalation and delay?

The survey matters because perception is evidence. It becomes dangerous when leaders mistake that evidence for a complete diagnosis. The following four myths create false confidence by turning a partial signal into a conclusion about the whole operating system.

Why survey results need operational evidence

Survey responses describe how work is experienced by the people who answer. That perspective can expose fear, silence, inconsistent supervision, or a gap between policy and practice that a document review will miss. It also has limits because people interpret questions through local relationships, recent events, language, workload, and their belief about what will happen after they answer.

Andreza Araujo’s work on safety culture emphasizes the distance between stated values and the decisions people can observe. That distance is why a survey should sit beside the evidence visible in daily work, rather than replace it.

Leaders should ask what the result makes possible to investigate. If the score says workers trust supervisors, trace a recent concern and examine the response. If the score says procedures are followed, observe a task whose conditions changed. The survey opens the door; field evidence shows what is on the other side.

Myth 1: A high score proves that safety culture is strong

This myth seems reasonable because a high score feels like confirmation. When workers select favorable answers, leaders may conclude that expectations are understood, supervision is consistent, and people will speak up before exposure becomes harm.

A positive response can mean those things, but it can also reflect a narrow question, a recent leadership message, or the fact that respondents do not expect anything to change after the survey. People may approve of the organization’s intentions while still lacking time, authority, or protection to challenge an unsafe decision.

James Reason’s distinction between active failures and latent conditions is useful here. A survey can reveal an attitude toward safety while leaving hidden weaknesses in staffing, maintenance, work design, escalation, or control ownership untouched. The result is not false, but it is incomplete.

Leaders should validate a high score with a small evidence trace. Select one critical activity, compare the written expectation with the task as performed, interview the people who can stop or change the work, and check what happened the last time a control was unavailable. A strong culture is not established by agreement alone. It is established when favorable perceptions correspond to reliable decisions.

Myth 2: A low score means workers have the wrong attitude

When results are poor, the fastest explanation is often that employees are resistant, disengaged, or insufficiently committed to safety. That interpretation protects the system from scrutiny because it turns a signal about work into a judgment about the people doing it.

Low scores may instead point to conditions that workers cannot repair through motivation. They may be describing conflicting priorities, repeated action delays, unclear stop-work authority, weak handovers, or supervisors who ask for concerns but provide no answer. A worker can care deeply about safety and still report that the system makes safe decisions difficult.

Edgar Schein’s work on organizational culture helps explain why visible statements are weaker than repeated consequences. People learn what the organization truly prioritizes by watching which problems receive resources, which delays are accepted, and which concerns are welcomed only when they do not interrupt production.

The practical response is to test the operating condition behind the score before adding training. Ask workers to identify one situation in which the stated expectation becomes difficult to meet, then observe that situation with them. The goal is not to defend the survey or disprove it. The goal is to find the decision that the organization has made hard to execute.

Myth 3: Anonymous responses create enough psychological safety

Anonymity can lower the immediate fear of identification, which makes it a useful design choice for some questions. It does not, by itself, create the confidence required for people to challenge a supervisor, report a weak control, or describe a pattern that senior leaders would rather not hear.

Psychological safety is visible in what happens after a concern is raised. Workers need to see that the concern is examined fairly, that retaliation is not tolerated, and that the response explains the decision even when the requested change cannot be made. A confidential form cannot provide that experience if the organization has no response rhythm.

A survey can therefore produce a paradox. The organization receives anonymous criticism, reports that participation was strong, and still leaves the workforce unsure whether speaking directly is safe. The channel has collected information without strengthening the relationship that turns information into action.

Leaders should pair anonymous responses with protected listening sessions, clear escalation routes, named owners, and visible closeout. The freedoms that support honest reporting become credible only when people can see how a concern moves from a message to a decision.

Myth 4: Repeating the survey will show whether culture improved

Repeated measurement is useful when the organization has changed a condition that the survey can reasonably detect. It is not useful when leaders repeat the same questions, publish a new score, and leave the underlying decision process untouched.

A higher score can reflect better communication, different participation, changed expectations, or survey fatigue. A lower score can reflect increased honesty after leaders make speaking up safer. Neither movement should be interpreted without knowing what changed in the work and how the respondent group experienced it.

Measurement also becomes distorted when leaders reward improvement in the number rather than improvement in the condition. Teams may learn to frame answers favorably, avoid difficult topics, or treat the survey as a reputation exercise. The score rises while the exposure remains.

Before repeating the survey, define the operational change it is meant to test. If the issue is response to concerns, track acknowledgment time, named ownership, closeout quality, and recurring themes alongside the next survey. If the issue is control reliability, combine perception data with field verification. The comparison becomes meaningful when it tests a decision, not just a sentiment.

What leaders should do after the survey closes

Start with two or three themes that could change exposure, decision quality, or worker trust. Do not publish a long action register that treats every percentage point as equally important. Prioritize the themes that have a clear connection to work and an accountable decision owner.

Then compare each theme with another evidence source. Use field observation, action aging, incident investigation findings, control verification, worker interviews, or a trace of a recent escalation. The difference between climate signals and broader culture diagnosis matters because it determines what the organization can responsibly claim.

Close the loop in language that workers can recognize. State what was heard, what was checked, what will change, who owns it, and when the organization will report back. If a request cannot be adopted, explain the constraint and the alternative control. Silence after a survey teaches people that participation is an administrative event.

A survey earns its place in safety management when it changes the quality of a decision. The number is a starting signal. The culture is found in the response.

FAQ about safety culture surveys

Can a safety culture survey replace field observation? No. The survey explains how work is perceived, while observation tests what people actually encounter. Both are needed when leaders are deciding whether a control is reliable.

What should leaders do with a high score? Treat it as a hypothesis worth testing. Trace a recent concern, inspect a changed task, and compare the response with the favorable perception.

What should leaders do with a low score? Investigate the work condition before blaming attitude or launching generic retraining. A low score may identify a design or leadership problem.

What is the best measure after a survey? Use the measure that matches the theme. Response time, ownership, closeout quality, control availability, and repeated concerns are often more actionable than a single culture index.

Turn culture data into safer decisions. Explore Andreza Araujo’s books and practical safety resources for methods that connect leadership, culture, and prevention.