Psychological Safety Team Survey (Edmondson 7-Item)

It measures whether team members feel safe taking interpersonal risks on their team, such as speaking up, asking questions, admitting mistakes, and offering dissenting views. The core items are based on Amy Edmondson’s seven-item psychological safety scale and include both positively worded and reverse-scored statements. The follow-up questions help explain low scores instead of leaving them as unexplained numbers. It is designed to reveal team climate, not individual performance.

Use this template when you need a focused read on team climate, manager behavior, and whether people feel safe to be candid. It is especially useful after reorgs, leadership changes, conflict, missed deadlines, or when teams are quiet in meetings despite obvious issues. A general engagement survey is broader; this one is narrower and better for diagnosing speak-up barriers. If you only need one or two pulse questions, this may be more detailed than necessary.

For most teams, quarterly is a practical cadence because psychological safety changes with manager behavior, workload, and team composition. Monthly can work for small teams actively improving a known issue, but repeated surveying can create fatigue if nothing changes. Weekly is usually too frequent for this topic unless you are running a short pulse with a very specific intervention. Whatever cadence you choose, keep it consistent so trend lines are meaningful.

HR, People Ops, or an internal survey owner should administer it, with the team manager receiving only aggregated results. The anonymity guarantee should be explicit, because psychological safety questions lose credibility if people think answers can be traced back to them. If the team is very small, set a minimum reporting threshold or combine results with a broader group to protect identity. The point is to create honest feedback, not to identify dissenters.

Reverse-scored items help reduce response bias and make it harder for people to click through without reading carefully. They also capture the negative side of psychological safety, such as fear of punishment, exclusion, or reluctance to ask for help. The follow-up questions attached to low ratings are important because a low score without context is hard to act on. Those open-ended prompts help you identify the specific meeting types, interactions, or behaviors driving the problem.

A common mistake is collecting demographics before the core items, which can make anonymity feel less real and lower response quality. Another is using leading language or adding manager praise questions that blur the construct being measured. Teams also often forget to act on low-safety signals, which teaches people that speaking up does not change anything. Finally, avoid turning this into a broad engagement survey by adding too many unrelated questions.

Yes, but keep the meaning of the Edmondson items intact so you do not break comparability. You can adapt examples in the follow-up questions to fit your environment, such as design reviews, shift handoffs, incident reviews, or client calls. If you change the rating scale, use a clear 5-point Likert format with semantic anchors rather than raw numbers alone. Avoid rewriting the items so aggressively that they stop measuring psychological safety.

Ad hoc conversations can surface useful anecdotes, but they are inconsistent and often miss quieter team members. This template gives you a repeatable structure, a stable scale, and a way to compare results over time. It also captures anonymous feedback, which people are more likely to share when the topic is sensitive. Use conversations to explore the results, not as a substitute for measuring them.