42 questions, about 10 minutes per person. Every team gets six behavior scores, and the manager learns which one to fix first.
Attuned runs inside 300+ companies, including
“This assessment provides multiple layers of data that can be used to shape metrics and allows you to see the variation in responses…”
A psychological safety assessment is a short survey that everyone on a team answers to show whether people feel safe to take interpersonal risks: admit a mistake, ask for help, disagree with the plan, or raise a problem early. Amy Edmondson, whose 1999 study of 51 work teams in a manufacturing company defined team psychological safety, described it as a shared belief that the team is “safe for interpersonal risk taking.”
Because the belief is shared, results are read at team level. Attuned breaks the score into the six behaviors where safety shows up day to day:
Offering useful information, and asking for help without it counting against you.
Owning an error early, so the team can contain it and learn from it.
Challenging an idea or a decision, including one that came from the boss.
Telling a colleague something they could improve, and hearing it back.
Sharing a real perspective without having to hide who you are.
Taking responsibility for a call inside an agreed scope, without waiting to be told.
Every established psychological safety scale is a survey the team fills in about itself. They differ in what they score: one overall belief, a set of stages, or the specific behaviors a manager can change.
Edmondson’s team psychological safety scale. Short and reliable, and the standard in research. It tells you how safe the team feels overall.
Timothy Clark’s model moves from inclusion to learner, contributor and challenger safety. It tells you how far along a team has come.
Attuned scores the six behaviors where safety shows up at work. It tells the manager which one to work on first.
Edmondson’s 1999 scale is the starting point for nearly every psychological safety survey since. Each person rates seven statements about their team on a seven-point scale, and three of them are worded negatively and reverse-scored (marked R):
Edmondson (1999), Administrative Science Quarterly 44(2).
It’s a good scale. If you’re running a study, use it. The catch for a manager is that it hands back one number. A team that shares knowledge freely but never argues with the supervisor scores somewhere in the middle, and the middle does not say what to change. Google’s Project Aristotle, a study of 180 teams, found psychological safety mattered more than any other team dynamic. Then Google had to turn that finding into something a manager could do on Monday.
Four kinds of tool measure it: free research scales such as Edmondson’s survey, stage-based instruments such as LeaderFactor’s PSindex, facilitated scans such as the Fearless Organization Scan, and platforms such as Attuned that score specific behaviors and track them over time. Engagement platforms usually carry a speaking-up item or two, which flags a problem without locating it.
Swipe to see the full comparison →
| Instrument | What it measures | Built for | What you get |
|---|---|---|---|
| Team Psychological Safety scaleEdmondson, 1999 | Shared belief that the team is safe for interpersonal risk. | Research on team learning and performance. | One team-level score. Free to use, with no reporting or actions attached. |
| Fearless Organization Scan | Team psychological safety, built on Edmondson’s research. | Teams, run directly or with a certified partner. | A team score against a global benchmark, usually debriefed in a session. |
| PSindexLeaderFactor, after Clark’s 4 Stages (2020) | Inclusion, learner, contributor and challenger safety, in 12 items. | Leadership development programs. | Stage scores for an intact team, rated from several perspectives. |
| Project Aristotle questionsGoogle re:Work | Five team dynamics, with psychological safety first. | Team self-reflection. | A structured team conversation, with no benchmark. |
| Engagement survey itemsCulture Amp, Qualtrics and others | One or two speaking-up items inside a broader survey. | Company-wide engagement tracking. | A flag that something is wrong, without which behavior. |
| Attuned psychological safety assessment | Six behaviors, from Mistake Reporting to Making Decisions, in 42 questions written by psychologists and psychological safety practitioners. | Managers and People teams. | Team scores from 20 to 100 with High, Average or Low levels, ranked actions, and quarter-over-quarter tracking. |
Sources: Edmondson, A. (1999), “Psychological Safety and Learning Behavior in Work Teams,” Administrative Science Quarterly 44(2). Edmondson, A. (1996), “Learning from Mistakes Is Easier Said Than Done,” Journal of Applied Behavioral Science 32(1). Clark, T. R. (2020), The 4 Stages of Psychological Safety. Project Aristotle as reported by Charles Duhigg, “What Google Learned From Its Quest to Build the Perfect Team,” The New York Times Magazine (2016). For a broader view of team health, try our free team effectiveness assessment.
Only 3 in 10 U.S. workers strongly agree that their opinions seem to count at work.
Gallup’s estimate of what moving that ratio to 6 in 10 would be worth. Jake Herway, How to Create a Culture of Psychological Safety, Gallup, December 2017.
Engagement surveys ask whether people like working here. They rarely ask whether anyone would tell you about the bad batch, the shortcut, or the plan they think will fail. So two teams can post the same score and be in very different shape.
Problems get raised in the meeting. The supervisor hears about the jam on Tuesday, while it’s still a jam.
Problems get raised in the parking lot. The supervisor hears about it in the incident report.
Illustrative teams
One report per team, read at a glance.
Each team gets an overall psychological safety score from 20 to 100, plus a score for each of the six behaviors. Every score carries a level: High means keep going, Average means take action, Low means act now. Under each number sits the spread of answers, so a team that averages well but has three people far below the rest is visible as exactly that.
No manager sees an individual answer, which is what keeps the answers honest.
Illustrative team
Each team gets specific actions, ranked by urgency, for the behavior that scored lowest.
A score with no next step turns into a slide, then into nothing. Attuned ranks the actions for each team so the supervisor starts with the one that matters most, and changes one visible habit at a time.
The assessment also pairs with Attuned’s intrinsic motivation assessment. The psychological safety score shows which behavior is weak. Each person’s motivator profile helps explain why: someone who ranks Feedback highest and has had none for months tends to stop offering any, and a manager can read that silence as contentment. TalkCoach then preps the one-on-one where it gets fixed.
Illustrative actions
Run the same survey under the same conditions, and compare team with team.
One reading is a baseline. The second shows whether the change stuck. The dashboard benchmarks teams across the company, so you can see which managers need support and which have something worth copying.
Read changes with care. A new manager, a reorganization, or a different set of people answering can move a score as much as anything you did. Look at who responded and how the answers are spread before you credit the intervention.
Illustrative data
No IT project. A pilot on one team or business unit is usually live within a week. A company-wide rollout typically takes three to four weeks.
Invite any team by email or single sign-on. Each person answers the survey.
An overall score and six behavior scores per team.
The dashboard shows which teams need support first.
Retake under the same conditions and see which behaviors moved, team by team.
The leader who runs the team owns the fix. The People team owns the program. Both work from the same report.
Bring us one team you aren’t sure about. We’ll walk you through its report and what the manager would do first.
Book A Demo