Psychological Safety

Psychological safety at work: a definition

Psychological safety at work (الأمان النفسي في العمل) is an employee’s belief that they can ask a question, admit a mistake or disagree with their manager’s view without that being counted against them.

The definition names three acts: asking a question, admitting a mistake and disagreeing with the manager. In each of them the employee says something that can bear on how they are seen at work. A question can show what they do not yet know. An admitted mistake shows something they got wrong. Disagreement sets their view against the view of the person they report to. Psychological safety is the belief that none of the three will be held against the employee who does it.

The term describes a belief, and the belief is the employee’s. It concerns what the employee expects to follow if they ask, admit or disagree: whether the act will be taken as part of the work, or counted against them.

What psychological safety is not

Psychological safety is not the absence of accountability. An employee in a psychologically safe team remains answerable for the work, including for a mistake they admit. What the definition excludes is the admission itself being counted against them.

Psychological safety also does not mean that work is free of pressure or of criticism. Pressure and criticism can both be present where psychological safety is present, and the concept removes neither of them.

A psychologically safe team may in fact be more candid in its criticism than a team without psychological safety, because criticism in it is directed at the work and not at the standing of the person who did the work. When criticism of a piece of work is not taken as a judgement on its author’s position, it can be stated plainly, and the author can hear it as a comment on the work rather than on themselves.

How the absence of psychological safety shows

The absence of psychological safety is not declared. It is inferred from behaviour that can be observed:

• Mistakes come to light late, once their effect has grown, because raising them early is costly.
• No one asks questions in meetings, and the same questions are then raised afterwards in side groups.
• Disagreement comes only from the person in the most senior position, or does not come at all.
• The same mistake recurs in different teams, because no one reported it the first time.

Raising a mistake early has a cost for the person who raises it, and that cost is what the definition describes as the mistake being counted against them. While the mistake goes unreported its effect keeps growing, so that when it does come to light it is larger than it was when someone first knew of it.

The questions in the second sign do not disappear. What changes is where they are asked. They are not put in the meeting itself, and they reappear afterwards in side groups, away from the meeting in which they went unasked.

Whether disagreement is voiced depends on position. The person in the most senior position may disagree. The others do not, and if that person does not disagree either, no disagreement is heard at all.

A repeated mistake follows from its first occurrence having gone unreported. One team made it, word of it did not reach the other teams, and another team then made it again.

What actually builds psychological safety

The decisive signal is the manager’s reaction to the first piece of bad news. That reaction is what the team observes when someone first brings a problem forward.

Behaviour that is rewarded once in front of the team is taken as a rule. Behaviour that is penalised once is taken as a warning, and that warning carries more weight than any written policy. In both cases a single occasion is enough: one reward or one penalty is taken as showing what will happen the next time someone does the same.

For that reason a written policy that invites employees to raise problems does not outweigh one occasion on which raising a problem was penalised. The team takes its guidance from what the manager did on that occasion, not from what the policy says.

Tools such as a reporting channel or an anonymous survey are useful for detection, but they do not create psychological safety. They can bring to light a problem that is not being raised openly. Having them in place does not change what happens when an employee raises the same problem openly.

The existence of a channel for reporting a problem does not mean that raising the problem openly has become safe. Psychological safety, as defined above, concerns what an employee can do openly as themselves: ask a question, admit a mistake or disagree with their manager. A reporting channel and an anonymous survey help an organisation find out what is not being said in that way. Psychological safety concerns whether it can be said in that way.

Why psychological safety concerns HR

Psychological safety is a mediating variable. It is hard to measure in itself, and its effect shows in employee engagement, in the turnover rate and in the quality of the information that reaches management.

Describing it as a mediating variable means that psychological safety is approached through its effects rather than measured directly. Engagement, turnover and the information reaching management are the three places named in which its effect shows, and they are where HR can observe it.

The third of these effects concerns information. An organisation that does not receive bad news in time makes its decisions on a picture of the situation that lags behind reality. The information on which it decides describes the situation as it was, not as it is at the moment of the decision.

Bad news that does not arrive in time reaches management after the point at which it was first known, and the decisions taken in the meantime rest on the earlier picture of the situation.

This is an explanation of the concept, not legal advice.