Psychological Safety Survey: Edmondson’s 7-Item Scale

It’s Tuesday morning and your product team has just met to discuss a delayed software release.

An engineer recently discovered a serious flaw in the software 3 weeks ago, but after a manager dismissed a colleague’s warning, she kept quiet.

Now the company has to pause the launch, fix the code, and explain the delay to 12 major customers.

She knew what was wrong. But speaking up seemed riskier than staying quiet. Everyone would share the embarrassment of the delay, but she would face the embarrassment on her own.

Harvard Business School professor Amy Edmondson encountered a strange phenomenon while studying clinical teams. She found that better-led teams reported more errors.

In other words, they weren’t making more mistakes. They were just more willing to talk about the mistakes they were making.

Edmondson used the term psychological safety to describe the shared belief among people that they can take interpersonal risks without fear of humiliation or punishment.

Psychological safety is not about avoiding conflict, lowering standards, or protecting weak performers from feedback.

Meet the mind behind the movement: amy edmondson’s accidental discovery

Edmondson’s early research on clinical teams had an odd side-effect.

It looked like better teams were making more medication errors.

Turns out, what she’d stumbled upon was a difference in reporting errors.

In strong teams, members were more likely to detect and report errors.

In weaker teams, errors were more likely to be covered up, leading to fewer reported errors.

The point for HR teams is this: Most of your "safety" signals are affected by the factors you’re trying to measure.

How many incidents are reported? How many whistleblower calls are made? How many complaints about quality levels are registered?

How many project risks are flagged? All of these are influenced by how comfortable people feel reporting these issues.

This makes it difficult to interpret changes in reported issues. If you see an increase in reported issues, are you witnessing a genuine increase in work issues? Or is the increase due to people finally trusting the reporting system? You can only interpret the number of people speaking up by also looking at operational outcomes and what happens after a report is made.

Edmondson also shifted the conversation away from individual bravery to creating conditions in which bravery wasn’t needed.

Instead of asking why someone didn’t speak up, she suggested asking what the team environment taught them to expect.

What are the four pillars of psychological safety according to edmondson’s framework?

Edmondson did not specifically list "four pillars" of psychological safety.

We’ve put together a list of four items that help us think through the themes of Edmondson’s framework.

The first item on this list is the importance of a team’s attitude toward risk and failure. People pay attention to how risk is managed by looking at how mistakes are handled.

It’s important for teams to distinguish between intelligent failure and avoidable negligence. An intelligent risk is a sound decision made based on incomplete information where the agreed-upon safeguards are in place. This differs from carelessness.

Ensuring people make this distinction - and that leaders make this distinction - further enhances team learning and psychological safety.

The second item on this list is the importance of open conversation. The key here is how leaders respond to receiving undesirable news.

If a leader responds poorly to someone delivering undesirable news, this will have a huge impact on whether that individual and others feel they can continue to be open.

The third item on this list is the willingness to ask for and offer help. When asking for help is viewed as a sign of weakness, people will be so concerned about managing their image that they fail to actually resolve the issue at hand.

On the other hand, if managers emphasize the importance of asking for help and offer support by asking "What support do you need?" before a visible delay happens, they can reset the norm.

Finally, the fourth item on our list of pillars of psychological safety is inclusion and respect for difference. This is an area where teams often recognize the desire to do better, but fall short in practice.

They may have formal programs in place for inclusion, but the reality of meetings may favor a certain style of communication.

It’s helpful to reevaluate meetings based on observable characteristics: Who speaks first? Whose ideas are quietly attributed to someone else? Who gets cut off?

Whose voiced objections are given less credibility and require greater proof?

How do you move into a learning zone?

Edmondson’s two-by-two charting of psychological safety and performance standards is helpful for HR professionals trying to understand different work environments.

When you understand this model, you can diagnose where your team is and focus on the right things.

Understanding these potential failure modes is helpful because they require different solutions.

For example, if your team feels like a comfort-zone team because you have nice personal relationships but a soft delivery culture, you need to focus on raising standards and encouraging harder conversations. You don’t need another team offsite.

If you have a team that feels like an anxiety-zone team - perhaps they’re high performers, but they hide their mistakes for fear of the consequences - you need to focus on changing how leaders respond to receiving bad news. You don’t need to set new goals.

It’s also helpful to be aware of common missteps.

Leaders may think they need to boost performance standards and psychological safety at the same time by raising the bar while emphasizing safety. This often leads to...

An anxiety zone: Employees are asked to do better while they’re still worried about the repercussions of speaking honestly.

The better approach is to first demonstrate that delivering bad news is okay and then...

Raising the bar.

Make the business case: don’t forget to link psychological safety to business performance

Psychological safety impacts whether an organization can get the most out of the information it’s already paying for.

In other words, it’s the difference between buying a great sports car and actually getting to drive it.

When Google’s Project Aristotle set out to determine the dynamics of effective teams, it ranked psychological safety as the most important of the five factors (the others were dependability, structure and clarity, meaning, and impact).

So when you’re pitching the benefits of psychological safety to executives, make sure you’re drawing clear lines to business outcomes.

Instead of saying psychological safety "drives profit" say how it encourages people to raise problems early to avoid costly mistakes and allow teams to deliver results faster.

Or you can talk about how psychological safety improves employee retention rates.

People rarely quit because of one bad experience. They quit after a string of experiences where they weren’t heard, where they weren’t given credit, or where their honest feedback resulted in punishment.

There’s also a credible way to estimate the cost of poor psychological safety without making up numbers.

You can ask your team to pick a recent, expensive failure within your organization and trace the event back to see who had information, when they had it, and why they didn’t share it.

A well-documented "someone knew and didn’t say" story can be more persuasive to a skeptical board than any benchmark from another company.

Taking the temperature: how to measure psychological safety

Before measuring psychological safety, ask yourself: What will I do differently based on these results?

Measuring psychological safety can hinder your efforts if, for example, you score low but the reason you can’t improve is due to an untouchable cause or lack of appetite for change.

In these scenarios, asking your employees and then disregarding their responses can be punitive and reduce psychological safety.

If you plan to measure psychological safety, a great way of doing so is Edmondson’s seven-item team scale:

• If someone makes a mistake on this team, the team often holds it against them.
• Members of this team can raise problems and difficult issues.
• People on this team sometimes reject others for being different.
• It is safe to take a risk on this team.
• It is difficult to ask other members of this team for help.
• No one on this team would deliberately undermine another member’s efforts.
• Working with this team, people value and use my skills and talents.

Items one, three, and five are reverse-scored. Make sure you pilot adapted or translated versions of this survey to ensure there aren’t issues with careless responses or translation drift.

Ensure that your survey questions are consistent from cycle to cycle. If you rewrite the questions, you’ll struggle to differentiate real change in your responses from measurement artifacts.

Whenever possible, collect individual responses from teams within your organization. An average score across the entire enterprise may result in a forgettable number that masks very fearful or very open teams within it.

A team may receive an average score of psychological safety that is "fine" despite a large minority feeling like they can’t share ideas due to severe punishments for mistakes.

Pay attention to the distribution of responses, and how many people disagree with a given statement.

Consider the participation rate when reviewing survey results. A high score from a small, self-selected group of respondents isn’t comparable to a high score based on a larger, more representative group of responses. Treating these results as equal may lead to incorrect conclusions about which teams require support.

In addition to your survey, collect behavioral evidence about how often employees raise risks before deadlines, how leaders react to failed attempts at work, who participates in meetings, or if people ask for help when they need it.

Keep in mind that your first measurement of psychological safety in a previously fearful team may drop, since awareness improves and people feel more comfortable sharing how they feel.

Warn leaders of this possibility ahead of time, so they don’t view a drop as a failure and use it as a reason to cease their efforts to improve psychological safety.

Asking teams to explain low survey scores in follow-up sessions, especially in the presence of the implicated manager, will only confirm their fears and defeat the purpose of assessing psychological safety.

Instead, ask questions about situations and routines, rather than asking questions about who complained about what.

Sparkbay allows you to collect feedback at regular intervals (many clients opt for monthly pulses).

Because psychological safety can fluctuate quickly based on events such as a reorg or change in leadership, our platform allows you to catch these shifts in your monthly feedback survey instead of waiting for an annual survey conducted several months later.

You can embed focused psychological safety survey questions and track their results over time to understand their trend.

These results are then reported against a clear engagement score out of 10, so you can easily understand your psychological safety survey questions in the context of overall engagement. eNPS is also available as a secondary signal instead of a primary focus.

Leveraging the power of audience segmentation, you can segment your data based on different attributes such as manager, department, tenure, and others to understand the story behind an average score and determine if a specific team leader or the entire organization has a negative impact.

Our platform is highly configurable, ranging from the wording of your surveys and dashboards to the content shown on each dashboard. This way, you can ensure the language used is appropriate for your workforce while also maintaining consistency in your survey questions to reliably track trends over time.

Sparkbay is an ISO 27001 certified platform.

For large enterprises we offer automatic report access based on the organization hierarchy. This way, each manager only has access to their teams, and team members cannot view other individuals’ survey results. Additionally, we provide a configurable minimum response number (five by default) below which survey results will not be shared with managers.

This feature ensures no one can reverse-engineer who said what.

Once managers receive their results, Sparkbay offers a library of concrete actions related to meeting participation, responses to mistakes, seeking help, or structured challenges. These actions help managers complete the loop by altering their own behavior and then checking the impact of these actions on the next feedback pulse instead of making assumptions.

If you’re interested in learning how Sparkbay can help you build a more engaged workforce, you can click here for a demo.

How to understand the warning signs and symptoms of a low-safety culture?

Understanding the warning signs and symptoms of a low safety culture can help your organization identify and fix safety issues.

• Artificial consensus: Decision-making appears to be unanimous, but later discussions or one-to-one conversations reveal disagreements.
• Late escalation: Employees raise concerns or risks only after a deadline has passed, a customer has reported an issue, or a lot of resources have already been invested.
• Post-mortems focused on blaming others: Post-mortems focus on finding the person responsible for a failure instead of analyzing assumptions, information transfers, or safety controls.
• Information hoarding: Employees keep useful information to themselves for status or protection, rather than freely sharing it.
• Low help-seeking behavior: Employees try to fix problems on their own and only seek support once a lot of resources have been invested.
• Solely positive feedback from open-text responses: Employees praise the company-wide values in open-text response questions, but don’t provide specific feedback about their local leadership teams, which may indicate concerns about the anonymity of their responses.

Be aware of discrepancies between formal channels and actual employee behavior. Your organization may have a hotline, suggestion box, or survey for employees to voice their concerns, but employees may fear using any of them.

Your best indicator of a poor safety culture is likely to be discrepancies in responses at the manager level. If your organization receives very different safety culture responses from employees in similar roles, this is likely due to differences in local leadership behaviors, which indicates where support should be focused.

Finally, avoid jumping to conclusions about the intent behind safety culture issues. There may be other reasons for a poor safety culture survey results within certain teams, such as an excessive workload, active restructuring, role ambiguity, or a recent bad event within the team.

From insight to action: how to build psychological safety on your team

What can you do to boost psychological safety within your team?

First, try thinking about your work as a learning problem. As a leader, this means articulating what’s uncertain instead of pretending you’ve got everything figured out.

For example, you might say, "We know what we think the customer needs, but we haven’t tested whether this is scalable." This signals to your team that it’s okay to deliver information that contradicts your beliefs while maintaining the expectation of high performance.

Second, leaders should model bounded vulnerability. This means that when you admit a mistake, you tie it to a corrective action. Otherwise, it becomes a performance designed to put the team at ease.

For instance, it’s helpful if you say, I approved a deadline without double-checking the status of the dependency. I’ll go back and rework these assumptions with the tech lead before we commit to anything. A generic, "I’m not perfect either," doesn’t add much value.

Third, design opportunities for your team to dissent. Creating a safe environment by simply asking, "Does everyone agree?" often leads to silence.

Try asking questions like, "What do you think our weakest assumption is here?" or "Where do you see potential for failure?"

It’s also important to create environments where the most junior people feel comfortable speaking up before the executives share their thoughts. This helps erase the impact of anchoring and social hierarchy.

Fourth, when someone on your team speaks up, remember that the first thirty seconds are the most critical. Everyone is watching to see how you respond to bad news, and they will take cues from your reaction.

Try to thank the person for speaking up, ask factual questions, and reiterate the immediate risk.

At this point, it’s helpful to also articulate the "next step." What are you going to do now? What happens after this conversation?

This isn’t the time to express visible anger or blame. These reactions push future concerns down the road until they’re no longer actionable.

Fifth, create a different avenue for accountability. This is why you want to keep disciplinary processes different from learning reviews. Consider how a learning review can focus on a discussion of conditions, assumptions, handoffs, and controls.

If the review reveals real issues of misconduct or repeated disregard for company norms, there’s a separate, fair process for dealing with that.

This process fuels progress without turning every learning review into an investigation.

Sixth, managers should be held accountable for carrying through on actions coming out of learning reviews. It’s important to require one or two actions tied to results. These actions should have named owners and a date where they are revisited.

This isn’t a long to-do list of vague intentions. It’s a specific action that prevents someone from putting a learning review on the back burner.

Moreover, it’s valuable for managers to articulate what will change within their teams and what won’t change. This level of honesty creates credibility. It’s better than making big promises you can’t fund.

Finally, it’s important for managers to be honest about the fact that they themselves may not be able to change everything within their team.

They might not shield their team entirely from an intimidating layer above them or a company culture where bad news is punished at the highest level.

If that is the case, it’s worth stating this honestly and escalating the concern through the proper channels.

If you’re interested in learning how Sparkbay can help you build a more engaged workforce, you can click here for a demo.