Blog / Team Psychological Safety and Team Efficacy: Confidence to Deliver Is Not Safety to Raise a Risk
Team Psychological Safety and Team Efficacy: Confidence to Deliver Is Not Safety to Raise a Risk
Believing ‘we can deliver’ and believing ‘I can safely raise an uncertainty’ are different team beliefs.
October 11, 2026 · 5 min read · By Cepingku Editorial Team
Ready to start?
Start the Teamwork Test →Team efficacy concerns a shared belief that a team can accomplish its task. Team psychological safety is a different shared belief: that the team is safe for interpersonal risk taking. In her 1999 study, Amy Edmondson defines psychological safety this way and treats feedback seeking, information sharing, asking for help, discussing errors, and experimentation as learning behaviour. ‘We can deliver’ and ‘I can safely raise a risk’ therefore do not answer the same question.
| Team belief | What it directly concerns |
|---|---|
| Team efficacy | Whether the team can accomplish the current task. |
| Psychological safety | Whether it is safe to ask, admit an error, or request help. |
A team may be highly confident in its capability and schedule while members fear looking incompetent if they ask a question or blame if they disclose an error. Conversely, members may be able to raise uncertainty safely without that proving that the team has enough resources or will deliver. Keeping the beliefs distinct helps a team look for the process that is actually missing.
A fictional project review shows three observable actions
This is a fictional review, not site data or an organisational diagnosis. A team is preparing a client report. Chen notices an unexplained gap in the data table. During the meeting, others repeat, ‘We are experienced; we will deliver tonight.’ Chen worries that raising the gap will look like slowing the project down, so says nothing.
The efficacy claim here is that the team can deliver. The safety question is whether Chen can say, ‘I could not verify this section and need help.’ Both can be present at once. If Chen raises the issue, observable learning behaviour can include asking about the data source, having another member check it, recording the consequence of the gap, choosing a correction or limitation statement, and updating the delivery plan. Silence prevents those actions; it does not demonstrate that the team lacks capability. Psychological safety lets Chen raise an unknown; it does not mean every claim is accepted, verification is optional, or decision responsibility disappears. The team still checks source data, records what remains unsupported, and has the accountable decision-maker choose correction, limitation disclosure, or delay with reasons.
Read the study result within its measures
Edmondson studied 51 manufacturing work teams. The abstract reports that psychological safety was associated with learning behaviour, while team efficacy had no stable independent association when psychological safety was controlled. The paper also reports differences between self-report and observational measures. The careful conclusion is that results were not fully consistent across measures. It is not that efficacy has no role, and the study’s associations do not prove that psychological safety causes learning behaviour or performance.
Learning behaviour is a process record: asking, seeking help, discussing an error, checking information, and revising action. It is not simply a claim that a team has a good atmosphere, nor can one employee self-report establish an organisation’s culture. A review can make the process more visible by naming who raises unknowns, how risks are recorded, who can request a check, and how mistakes are discussed without labelling a person as a failure.
The Teamwork Test can prompt reflection on self-reported team interaction. It does not establish this site’s validity, a team’s culture, an organisational causal effect, or whether any project will deliver.
Want to see where you land?
Take the Teamwork Test and get your results in minutes.
Take the Teamwork Test →