“How do we measure team psychological safety?” is one of the first questions I get from HR leaders and coaches. The good news is that a validated measure exists, and it is short. The harder part is choosing what to measure, at which level, and what you will do with the result.
This guide explains what to measure, gives a fair overview of the main options, and shows where our own assessments, Team Psychological Safety (TPS) for Team and Team Psychological Safety (TPS) for Leader, fit.
Start with what you want to measure
Team psychological safety is the shared belief by team members that the team is safe for interpersonal risk-taking. That is Amy Edmondson's definition from her 1999 research. It is a property of a team, not of a person and not of a company.
That has a direct consequence for measurement. You measure a real team: everyone in it, including the leader. You then look at two things: the team's average, and how much team members agree. A team where some members feel safe and others don't is not “average”. It has a different problem.
An organisation-wide score hides exactly this. In our data, the gap between two teams in the same organisation can be larger than the gap between two organisations. Teams eat culture for breakfast.
Three questions to ask any assessment
- What level does it measure? One team, an individual's feelings, or the climate of a whole organisation?
- On what research basis? Published, validated statements, or an intuitive framework?
- What happens after the report? Does someone help the team turn the result into an intervention, or does it stop at the score?
The main options
- Amy Edmondson's 7-item scale. It is the original measure from her research, published in her 1999 paper and used in research since. It is short and validated. On its own, you get a number per team, but no report, no benchmark and no guidance on what to do next. Three of the seven statements are worded in reverse, which some teams find confusing.
- The Fearless Organization Scan. It is a diagnostic built on Edmondson's work and run by licensed practitioners. It shares our research roots. It is often used across an organisation as well as within teams.
- LeaderFactor's PSindex. It is based on Timothy Clark's 4 Stages of Psychological Safety model. LeaderFactor reports a large benchmark database, with 1.3 million data points from 81,500 responses. The model is intuitive. When you look for published research that validates it, you will not find it.
- A few psychological safety questions in an engagement survey. These are easy to add to a survey you already run, and good for spotting trends across a large organisation. They are usually not built to measure one team, and the results rarely reach the team itself.
- Free question sets online. These are useful for a first conversation in a team. They come without validation, a report or a benchmark.
None of these is wrong. They answer different questions.
Team Psychological Safety (TPS) for Team
TPS for Team uses Amy Edmondson's seven validated statements, the same measure as her research. We made one change: all seven statements are worded in the same direction, with no reverse-worded items. We also add up to four open questions that you define, so the team can say in its own words what helps and what gets in the way.
- It takes 6 to 7 minutes. It is anonymous and online, in 14 languages.
- The report shows the seven elements. Our traffic-light chart shows at a glance which elements are strong and where the team disagrees.
- Each team is benchmarked against around 600 teams in 30 countries. The benchmark gives context. The most useful comparison remains the team against itself, over time.
- The report is a starting point, not an end point. A certified practitioner uses it to design an intervention that targets the two or three elements that need work, and the foundations behind them. Across 36 teams we measured before and after an intervention, 34 saw their average score go up, and 33 became more aligned.
Team Psychological Safety (TPS) for Leader
The leader is part of the team, and accounts for roughly half of the success or failure of any engagement. TPS for Leader is an extension of Edmondson's instrument that focuses on the leader. In 21 statements, it compares the leader's own view with the team's view on three dimensions: role-modeling, supportive responding and team enabling. The gaps between the two views are often where the conversation starts. It is available in 7 languages.
Which option fits your need?
| If you want to… | Look at… |
|---|---|
| Get a quick, validated number for a team, for research or a first look | Edmondson's 7-item scale |
| Track trends across a large organisation | Psychological safety items in your engagement survey, or an organisation-wide diagnostic such as the Fearless Organization Scan |
| Use a leader-focused stage model | LeaderFactor's PSindex |
| Measure a real team, see the seven elements and a benchmark, and design an intervention that improves its performance | Team Psychological Safety (TPS) for Team, with TPS for Leader for the leader's view |
How to get a report
Through a certified practitioner, a Master Facilitator, or TEAM.AS.ONE directly. Coaches and facilitators who want to run the assessments with their own client teams can become certified. For how certifications compare, see our buyer's guide.
Questions buyers ask
What is the best way to measure team psychological safety?+
Survey every member of a real team, including the leader, with validated statements. Then look at both the average and how much members agree. TPS for Team does this with Amy Edmondson's seven statements, a visual report and a benchmark.
Is the Team Psychological Safety (TPS) for Team survey validated?+
Its seven statements are Amy Edmondson's validated team psychological safety items. They are presented in the same direction, without reverse-worded items.
How long does the Team Psychological Safety (TPS) for Team survey take?+
6 to 7 minutes.
Can I measure a leader's impact on team psychological safety?+
Yes. TPS for Leader compares the leader's view with the team's view on role-modeling, supportive responding and team enabling.
How often should a team be measured?+
Measure before an intervention, and again after it, to see what changed. The team's progress against itself matters more than its place in the benchmark.
