Psychological Safety has become one of those phrases that is being used constantly but seemingly understood loosely. (I write this from Aotearoa New Zealand, but see it in social media content from across the world.) Since I started training and working with it as a concept prior to Covid I’ve watched it get folded together with wellbeing, with emotional comfort, with health and safety and more specifically psychosocial safety.
Consequently, it seems to mean so many things it’s in danger of meaning nothing.
I think the confusion is expensive. The idea most people seem to reach for, using the wrong definition, creates confusion or, worse, the opposite of what was intended.
So let’s separate the strands.
What Psychological Safety really is
The research construct comes from Amy Edmondson, whose 1999 study introduced team psychological safety, then defined it precisely. She defined it as a shared belief held by members of a team that the team is safe for interpersonal risk taking.
First, it is a team level concept, not a feeling in the individual. Second, it’s a shared belief, not just an individual belief, so the whole team has it. Third, the risks are specifically interpersonal: speaking up, admitting a mistake, challenging a decision, asking a question that may appear ignorant, flagging a problem, and so on.
Edmondson’s study of work teams found this shared belief was associated with learning behaviour, and that learning behaviour, in turn, predicted team performance.
Thus, Psychological Safety is a performance concept. Safety is not the goal. Safety is the condition and context that lets a team learn, adapt, and perform.
Again, Psychological Safety is a performance concept.
Some confusion is understandable
To be fair, and Edmondson has acknowledged this, the term does lend itself to ambiguity. She didn’t invent the term, but aimed to clarify it by being clear about the idea of the team level concept. The concept and language go back to work by Edgar Schein and Warren Bennis in the 1960s and William Kahn in the 1990s.
In the last 60 years, health and safety, wellbeing, wellness generally, mental health, and psychosocial concepts have grown in complexity, nuance, and reach, as has our comfort with talking about them. There’s clear opportunity for overlap and, naturally, confusion.
With the risk of confusion and ambiguity comes a greater expectation and responsibility for clarity, rather than the assumption we all know what we all mean when we use words.
Misreading number one
This is where the waste enters.
The moment we say ‘psychological’ it’s super easy to think about how we personally think and feel. To be perfectly honest, I think it’s a natural thing to do. It just isn’t the right approach here. But because we then add ‘safety’, it immediately starts to look like thinking and feeling safe, which just means being comfortable, and protected from difficulty and discomfort. That’s miles different from what the concept is intended to mean.
When we optimise for this idea, let’s call it emotional safety for clarity, we get teams that avoid difficult conversations to avoid people feeling unsafe. We get criticism softened so much it’s worthless. We get problems that continue because we don’t want people to feel bad. We get poor behaviour because it isn’t called out. It might all be lovely and everyone’s feelings are intact but it does not produce good outcomes. In the very worst cases, it leads to paralysis, and being held hostage by the fear of hurting individual feelings.
Edmondson’s distinction
Edmondson’s psychological safety encourages, and creates, the reverse. It makes challenge and disagreement possible, even deliberate, precisely because people have the shared belief that the team is safe for us to take such interpersonal risks. Speaking up honestly won’t be punished. It’s the very thing that lets a team be direct with each other, and call out issues, and not the thing that prioritises individual feelings over the group.
It’s a critical distinction. Edmondson herself has spent considerable effort, including her 2018 book, aiming to correct exactly this misunderstanding. Psychological safety is about raising the bar while making it safe to take the interpersonal risks that meeting a high bar requires.
Again, psychological safety is a performance concept.
Misreading number two
The next confusion is more specific to how the term has landed in workplaces. While it genuinely is a workplace concept, the organisational environment easily locates ‘safety’ in ‘Health and Safety’, immediately aligning it with hazard management and compliance. It doesn’t belong there.
Edmondson’s construct is about team learning and performance, not workplace hazard control. In the Aotearoa New Zealand context, and in others where the legislation is reaching further (I’m thinking Australia at the moment), there is a genuine but adjacent concept that does belong in the health and safety domain, and that’s psychosocial risk. I also think it’s easy to see how the two terms can be so easily conflated.
Workplace factors like workload, role clarity, unidentified and unmanaged stressors, exposure (especially ongoing exposure) that can harm mental health and wellbeing are legitimate concerns for any organisations. In the emergency management space (although it’s not exclusive to them) there’s also moral injury, which is a topic for a different post. Further, Aotearoa New Zealand has mental injury under the Accident Compensation legislation, also relevant for emergency management, and also a topic for another day.
Finally, WorkSafe, or whatever your home’s regulatory body is, expects organisation so manage psychosocial risk like any other hazard under the Health and Safety at Work Act. That’s fair, and it’s not optional. It’s also not the same thing as Edmondson’s team level, shared belief construct. Collapsing the two together produces confusion, greater risk, and cost. Are we talking team culture or regulatory duty? The two need quite different responses.
It matters what we think
There’s actual psychological safety (Edmondson’s performance construct), there’s emotional safety (comfort, and often the opposite of useful), and psychosocial risk (a health and safety duty). There’s a Venn diagram you can imagine here, but it’s important we get the definitions right, because the intervention depends on the definition.
If you think it means emotional comfort and freedom from emotional threat, you’ll try to build it by being nicer, avoiding conflict, and protecting feelings. You run the risk of having a team that can’t have a hard conversation when it needs to.
If you take it as managing psychosocial risk, then you’ll put resources into policies, programmes, and compliance, but may not develop the necessary culture. It will sit with Health and Safety people, or HR, neither of whom are the right groups to develop organisational psychological safety which is a leadership issue in the first instance, and one leaders need to own. Worse, you’ll develop a culture that looks to tick boxes, rather than create the necessary performance context.
If you understand it as Edmondson does, you’ll build it by making it genuinely safe to take interpersonal risk, which means leaders who respond well to bad news, who actively invite challenge and disagreement, and who treat sensible error as information, not abject failure. Clearly, this can feed into psychosocial safety.
Here’s the so what
But if you spend a year, and your budget, on emotional safety or psychosocial risk management thinking it’s psychological safety, you won’t have the performance or capability outcomes you may have hoped for, and you may conclude psychological safety wasn’t worth it. What went wrong of course was the reading of what it was.
There’s also a credibility aspect to consider. A vendor or leader who uses the terms interchangeably is telling you they don’t know the field very well. Further, If the phrase gets so overused or used so loosely it means little, then we’ve lost what we know to be one of the great performance enhancers.
If, then, you have someone saying they want to build psychological safety, the very first question must be some version of “What do you mean by that?”, so we can determine which of the three they actually want, and respond accordingly. In different worlds, all three might be legitimate, they just aren’t the same goal.
If you want performance, you need psychological safety.
From where I stand, getting it wrong can be expensive.
Photo credit: Fotos on Unsplash