How psychological safety emerges and changes over time in the workplace.

Given that psychological safety itself is a relatively young field, there’s not a huge amount of published research that addresses the longitudinal dynamics of psychological safety in the workplace: how it emerges and changes over time. We get into this a lot in workshops, but until now have had little in terms of scholarly evidence to back up our insights. That’s why this paper is much appreciated! Paradise Lost (and Restored?): A Study of Psychological Safety over Time, by Derrick Bransby, Michaela Kerrissey, and Amy Edmondson. And here’s the HBR article summarising it.

The paper, which examined longitudinal data representing more than 10,000 health care workers, shows how psychological safety changes for new hires to an organisation. Many of us, when joining a new organisation, feel a sense of optimism and excitement, and may feel a greater sense of psychological safety due to the support and encouragement that we receive as a new joiner. This is important – the first few months in a new role is exactly when we need our learning capacity to be at its highest: our capacity to ask questions, raise issues, ask for help, and suggest ideas is critical at this early stage. Indeed, this permission to ask the ‘unaskable’ is one of the superpowers of being a new joiner. 

The paper includes some neat insights into why new joiners may experience slightly higher psychological safety when they join a team, which then declines. New joiners enjoy the luxury of being able to say “Sorry, I’m new here..” as a prefix to asking questions – this newness provides an implicit permission to ask questions, and often, questions that others can’t, such as “Why do we do it this way?”. And we can ask the otherwise “stupid” questions, because there’s little risk of the (deadly to psychological safety) reply of “You should know that by now.”

However the paper demonstrates that in nearly all cases, that early psychological safety soon tails off as we discover the “real” organisation; what Schein and Van Maanen (1979) call “reality shock”. And when psychological safety erodes quickly, so does our power to ask these insightful questions. 

Fortunately, if we’ve joined a team with high psychological safety, the evidence shows that we recover within a reasonable timescale, and may even feel greater psychological safety over time. Sadly, if we join a team with low psychological safety, that initial decline is fast and steep, and never fully recovers. 

This is also one of the few papers that acknowledges that person-level differences (such as beliefs, traits, experience and expertise) influence psychological safety. Whilst the paper doesn’t explicitly mention it, I’d also propose that intersectional privilege plays a significant role here. 

The authors propose a multilevel framework (see the figure below) that integrates individual traits and group-level climate factors to explain the emergence and fluctuation of psychological safety. This perspective helps to reconcile perspectives in the academic literature that “neither disavows psychological safety as an emergent property of a group nor claims it as inherently individual” (Edmondson & Bransby, 2023: 69) – essentially there has typically been a (sometimes heated and ironically psychologically unsafe) debate about whether psychological safety is purely a group phenomenon, or an individual belief about the group. Of course, in reality, it’s effectively both, and this framework helps to elucidate that. 

As practitioners, what we can take away from this paper is that if we are recruiting people into our teams, we can be explicit with them that there is no cut-off date for asking the ‘stupid’ questions. By being aware of this phenomenon, we can put in place practices and behaviours to try to reduce the dip, and increase the pace at which psychological safety picks back up again. 

We also need to acknowledge that (as we might expect) recruiting people into already psychologically unsafe teams may not have the positive outcome we hoped for – in fact, due to that initial sharp decline in psychological safety, a new joiner may justifiably decide to cut their losses and leave after only a relatively short tenure. And this is an important point – if this has happened to you, you’re perfectly justified in leaving in the hope of finding a better environment to work in. But also remember – just as we all possess the ability to destroy psychological safety for those around us, we also all have some capacity to improve it too.

Many thanks to both Amy Edmondson and Kiron Bondale for the heads up on this great piece.

References:

Bransby, D. P., Kerrissey, M. J., & Edmondson, A. C. (2024) Paradise Lost (and Restored?): A Study of Psychological Safety over Time. Academy of Management Discoveries. https://doi.org/10.5465/amd.2023.0084 
(Available at the internet archive: https://web.archive.org/web/20240416072614/https://www.hbs.edu/ris/Publication%20Files/edmondson_bransby_kerrissey_ba07666b-9b1a-41da-810a-1f40079da327.pdf)

Van Maanen, J. E., & Schein, E. H. 1979. Toward a theory of organizational socialization. In B. M. Staw (Ed.), Research in Organizational Behavior, vol. 1: 209–264. Greenwich, CT: JAI Press.

Further reading

Psychological Safety – A Timeline
Top 10 Ways to Foster Psychological Safety in the Workplace
Sociotechnical Theory
The Adaptive Cycle and Self-Organised Criticality

Psychological Safety in Practice

We’ve just released another Psychological Safety Practice Playbook!

This playbook is specifically focused on practices and tools to help boost inclusion in your teams and organisations. It’s adaptable and appropriate to working in all domains. 

Inclusion needs to be built into everything we do. And if we’re working on building and maintaining psychological safety in your teams, inclusion goes hand-in-hand with that. These practices should be used alongside those in the Action Pack to ensure your ways of working, team behaviours and communications are as inclusive as possible.

It includes:

• Multiple Channels of Communication Guidance
• Time To Think Practice
• Team Calendar Template
• Inclusive Workplace Blueprint

Download for £3.99 here

Boeing

Boeing has agreed to plead guilty to a criminal fraud conspiracy charge after the US found the company violated the deal meant to reform it after two fatal 737 Max crashes, caused by flawed MCAS software, that killed 346 people. By pleading guilty, Boeing will avoid a criminal trial, which is what many of the victims’ families have been calling for. It also means that the Boeing now has a criminal record. I’m not sure how it works when a company has a criminal record, but I imagine it will make it somewhat more difficult for Boeing to be involved in commercial, government, or military supply chains, just as having a criminal record presents challenges for an individual. 

Ofsted and School Inspections

You may be aware that in the UK, we have a new government as of last week. And it’s already resulting in some changes. Here’s an interesting piece in TES about Ofsted (the UK school inspection authority) and how headteachers shouldn’t be sacked for a poor report. Not only do poor Ofsted reports not truly represent the complex and challenging nature of schools, but it argues that trust leaders should know well before an Ofsted report if there is a problem with the headteacher. From a psychological safety perspective, this is helpful for school leadership: a perception that you could be sacked for a poor report very likely leads to avoidance of intelligent risk taking and innovation, and certainly a lower chance of speaking up about concerns or mistakes. 

The User Research Sweet Spot

I really like this piece by Adrian Howard on user research, and how to distinguish between the different levels of risk and cadence – which then informs decisions about whether those tasks can be self-service, or researcher-led. Low-risk, high cadence activities are best suited for self-service, and that means we should ask ourselves a number of questions in order to better facilitate self-service and learning for user researchers:

• Can we break a big research question into smaller ones?
• Can we turn a big study into several smaller ones?
• Can we turn one-off work into longitudinal work?
• Can we find smaller bits of valuable research work that we’re currently ignoring?

This week’s poem:

The Gods Grant Few Pleasures Beyond These, by Fernando Pessoa

A verse repeating

A cool breeze,

Summer in the fields,

And the soul’s courtyard

Vacant and sunlit…

Or, in winter, the snowy

Summits in the distance,

The fireside where we sit,

And a poem to tell all this…

The gods grant

Few pleasures beyond

These, which are nothing.

But they also grant

That we want no others.

You can find this poem in Pessoa’s — A Little Larger Than the Entire Universe: Selected Poems

Thanks Jade for the excellent poem find.