Edmondson's Psychological Safety

6 min read · SessionData

In 1999, Amy Edmondson published a study of hospital nursing teams that produced a counterintuitive finding. The best-performing teams did not report fewer medication errors. They reported more.

The difference was not competence. It was climate. The high-performing teams had something Edmondson called psychological safety: a shared belief that the team is safe for interpersonal risk-taking. Members could ask questions, admit mistakes, raise concerns and offer half-formed ideas without fear of punishment or humiliation.

The teams that reported fewer errors were not making fewer errors. They were hiding them.

The construct

Psychological safety is not the absence of conflict, nor is it a synonym for trust or niceness. Edmondson was precise about what it is: a team-level belief about how interpersonal risk will be received.

In a psychologically safe team, a participant can say "I don't understand this" without being labelled slow. A facilitator can say "that session didn't work" without it becoming a performance issue. A junior team member can challenge a senior colleague's assumption without career consequences.

In a psychologically unsafe team, the same statements carry risk. People learn this quickly. They adjust. They self-censor, defer to authority, and keep concerns private. The team still functions — but it stops learning.

Why it matters for learning

The connection to learning is direct. Every serious model of how adults learn — Kolb's experiential cycle, Argyris's double-loop learning, Senge's learning organisation — requires people to do things that are interpersonally risky.

Kolb's cycle requires reflective observation: honest examination of what happened and why. This means admitting what went wrong, naming what was confusing, and surfacing disagreements about what the experience meant. Without psychological safety, reflection stays shallow — polite, consensual, and single-loop.

Argyris's double-loop learning requires questioning governing variables: the deep assumptions that shape how we act. Challenging these assumptions — in yourself or in others — is an inherently risky act. It implies that the current approach may be wrong, that the people who designed it may have been mistaken, that the organisation's theory of how things work may need revision. Without psychological safety, defensive routines take over and the second loop never fires.

Senge's team learning discipline depends on what he called dialogue: a quality of conversation where people think together rather than defend positions. Dialogue requires vulnerability. It requires saying "I'm not sure" and "here's what I'm struggling with" — speech acts that are only possible when the cost of being wrong is low.

The leader's role

Edmondson's research consistently found that psychological safety is shaped more by local leadership than by organisational culture at large. A single team can be psychologically safe inside an organisation that is not, and vice versa.

Leaders create safety through specific behaviours: framing work as a learning problem rather than an execution problem, modelling fallibility ("I may be wrong — what am I missing?"), and responding to bad news and mistakes with curiosity rather than blame.

This matters for programme design. A session can be well-designed in every other respect — clear objectives, strong content, structured evaluation — and still fail to produce learning if the climate in the room suppresses honest engagement. Participants will complete the exercises, give polite feedback, and return to their teams unchanged.

Psychological safety and accountability

A persistent misunderstanding is that psychological safety means low standards — that creating safety means accepting poor performance. Edmondson has been explicit that this is backwards.

Psychological safety and accountability are independent dimensions. A team can have both, neither, or one without the other. The most productive combination is high safety and high standards: people can speak honestly because the standard is clear and the environment supports the struggle to meet it.

Low safety and high standards produces anxiety: people know what is expected but cannot raise problems or ask for help. Low safety and low standards produces apathy. High safety and low standards produces comfort — pleasant but unproductive.

The combination that produces learning is the one where people feel safe enough to be honest about the gap between where they are and where they need to be.

Measuring safety

Edmondson developed a seven-item survey to measure psychological safety at the team level. The items are deceptively simple:

If you make a mistake on this team, it is often held against you (reverse scored)
Members of this team are able to bring up problems and tough issues
People on this team sometimes reject others for being different (reverse scored)
It is safe to take a risk on this team
It is difficult to ask other members of this team for help (reverse scored)
No one on this team would deliberately act in a way that undermines my efforts
Working with members of this team, my unique skills and talents are valued and utilised

The survey works because safety is a shared perception, not an individual trait. Agreement within teams (and variation between them) is what gives the construct its explanatory power. It is a property of the group, not the person.

The upstream condition

Psychological safety is best understood as an upstream condition — not an outcome to pursue for its own sake, but a prerequisite for the learning processes that produce outcomes.

Without it, Kirkpatrick Level 1 data is unreliable (people tell you what they think you want to hear). Level 2 assessment is incomplete (people will not reveal what they do not understand). Level 3 behaviour change is fragile (people will not experiment with new approaches in an environment that punishes failure).

The theory does not claim that safety is sufficient for learning. It claims that learning of any depth — the kind that questions assumptions, revises mental models, and produces lasting behaviour change — is not possible without it.

Edmondson, A. C. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350–383.

Edmondson, A. C. (2019). The Fearless Organization: Creating Psychological Safety in the Workplace for Learning, Innovation, and Growth. Hoboken, NJ: Wiley.