Psychological Safety Framework
Psychological safety is an emergent phenomenon: a property and state of a group that is felt (differently) by members of the group. For this reason, it’s a difficult (but worthwhile) concept to study, and many efforts have been made to codify and structure psychological safety in both conceptual and practical frameworks.
There are multiple frameworks of psychological safety, ranging from Tim Clarke’s “4 Stages” of psychological safety to Viktor Cessan’s Integrative Framework and the SAFETY model (which frankly, I can’t help but feel was an acronym looking for a model). When I began to work on psychological safety a few years ago, I was frustrated by the lack of a framework (that I knew of) that provided not just a way to understand psychological safety, but also included methods to sensitively measure it alongside practical and achievable actions suggestions for leaders and team members to take in order to build it.
The Psychological Safety Action Pack is designed around an implementation framework of the practices of psychological safety that not only provides a somewhat conceptual model, but also (primarily) practical ways to measure, build, and maintain psychological safety in an organisation. In this article, I’m going to look at and describe the elements and foundations of that framework.
The psychsafety framework stands on the shoulders of giants, and builds on previous excellent work by W E Deming, Dr Amy Edmondson, among others. I don’t intend to present it as entirely net new work, nor do I claim any ownership of the concepts. I only hope that we, as a community focussed on researching psychological safety, extolling the benefits, and empowering teams and organisations to create it, can collectively progress towards a state where psychological safety in teams is the new normal.
Many years ago, I learned about Deming’s PDSA (Plan – Do – Study – Act) cycle, often framed instead as PDCA (Plan – Do – Check – Act). Deming preferred the term “study” as it suggests less a perfunctory “check”, and more detailed analysis and true understanding of the system. What Deming is describing in the PDSA framework is essentially the scientific method, and an experimental approach. As we know through models such as Cynefin, the only truly successful way to operate and succeed in a complex, adaptive, sociotechnical system such as a large team or organisation, is through an experimental approach: Snowden describes this as a Probe – Sense – Respond cycle.
The Möbius Loop of the Open Practice Library adopts a similar approach:
This iterative, experimental approach to creating value is powerful, and encourages short, effective feedback loops in order to discover user needs, examine options, deliver, and analyse the results through discovery again.
And this is exactly the approach I took when creating the PsychSafety Framework during the Covid Pandemic of 2020:
The framework is based upon a cycle of Study – Measure – Build – Maintain – Reflect. The Study element is in homage to W E Deming, and is ideally the first stage, though you could, in theory, begin at any stage.
The simplified cycle can be seen as a reflective process of understanding and acting: to ensure that practitioners can not only decide what to do, but also reflect upon the success of their actions and take appropriate action. All too often, we get stuck in a Plan-Do cycle, forgetting to ensure we take the time to study and act on the learning.
The Five Phases
1 – Study. We begin with the study of psychological safety, because it’s critical for leaders to understand what they are trying to do, what the impacts and benefits may be (as well as the dangers). In this phase, we may also put together a business case for a programme, if we need to persuade senior stakeholders of the value of the programme. In this phase, we also build enthusiasm and energy, alongside the theory.
2 – Measure. As with any improvement effort, if we don’t take a baseline, we won’t know if we’ve improved, and we won’t know if the actions we’re taking are having a positive or negative effect. The measurement phase includes carrying out surveys (granular to a team or business unit level), alongside education and training, and building enthusiasm for psychological safety across the organisation. The survey consists of 10 statements based upon Dr Amy Edmondson’s research, which highlight any areas of concern in respect to the behaviours that engender psychological safety. More importantly, carrying out this phase also:
- Educates people about psychological safety itself
- Makes psychological safety a safe thing to talk about
- Encourages the very behaviours that build psychological safety.
That is to say, even if you discarded the results of the survey, there would still be some positive impact. Please don’t do that though, because the results inform the next phases of:
3 – Build. By highlighting areas of concern in respect to certain behaviours and practices (are mistakes used to blame team members, or used to create enquiry, or do team members feel they are able to suggest ideas?), we are able to identify crucial interventions and actions that are specific at a team level. Via the implementation of certain practices such as retrospectives, or workshops such as the fear conversation, we are able to build psychological safety within these team units. The same interventions, carried out at multiple layers of the organisation, create further momentum and improvement.
4 – Maintain. Psychological safety is not something we can build and forget about. We cannot measure it, run some workshops, and then go back to our “old ways” of command-control, blameful management. The maintain phase, whilst the least “exciting”, is possibly the most important phase. This is where practices and behaviours become habits in teams and across the organisation.
5 – Reflect. This phase involves self-reflection alongside team level retrospection. It’s absolutely crucial to ensure we learn from the previous phases before looping back to measurement once more. What practices and workshops succeeded? Which teams improved and where did we struggle? Do we need to modify the psychological safety survey to better suit the language and context of the people involved? These may be difficult lessons, but they are powerful and important.