There has, so far, been only one occasion that brought me to the brink of having a good cry in fear that I might not be able to complete my EdD. Most of the time, although my work has certainly not been easy, I tend to get by by maintaining my faith in the process. Step by step, if I plod along, follow the process and take the advice from my supervisors and others, I will eventually get there. This occasion was different. This was the day the my statistical analysis didn’t work. I had meticulously reviewed the relevant literature, broken it down into themes, and conducted a pilot study. From there, I had reduced and refined my survey questions using statistical analysis procedures to edge as close to construct validity as I could.

I was pleased and confident with the result, but the data from the second pilot study were… all over the shop. It was clear that something had gone badly wrong with my survey instrument design. The thought that the issue might be the small sample size filled me with dread. The first iteration of my survey was, frankly, horrible. I was very grateful that so many of my colleagues had persisted with it. I could not see how on earth I was going to be able to find a larger sample size from which to start the statistical analysis again. Eventually, with help, support and expert guidance, I found an alternative path through, which is more suitable and appropriate for my study. I accept, from some user feedback, that some teachers still find my survey too long, but it is in much better shape than it was! This mistake arose from the fact that I did not have the expertise to understand that I had selected an inappropriate methodological tool. Once corrected, I was able to proceed, and my work is better for it. My inexperience was not blameworthy and I did not feel shame per se, and it was not a pleasant experience. However, it was an important learning experience.

The science of failing well

I have recently read Amy Edmondson’s (2023) Right Kind of Wrong: Why Learning to Fail Can Teach Us to Thrive. She describes her own crushing PhD pivoting moment, in which the data seemed to nullify her hypothesis about the relationships between team dynamics and frequency of mistakes in hospital teams. Rather than good teams making fewer errors, she discovered that, the greater the degree of psychological safety perceived (and cultivated) in teams, the more mistakes they reported. Importantly, reporting small errors enables us to catch and correct them, which is essential in helping to avoid large, complex errors, which would have much more serious consequences.

Lessons for educators

How can this insight be made relevant to schools? I have been thinking about this. It is important because, ultimately, we all share common goals: Improved student outcomes and resilience. I (and many others) believe that the best pathway we have to achieving this is to extend the same aspiration to our teachers. But there are many barriers in the way. Some, but not all of them are beyond our control. Here, inspired by Edmondson, I suggest how making space for teacher professional learning can help us all.

Firstly, we need to stop blaming ourselves and each other for systematic problems. We fear blame because it is unpleasant and embarrassing. It feels personal because we tend to invest so much of ourselves in our work. We often hiding our mistakes to avoid coming under the shadow of the threat of the ‘informal support plan.’ Edmondson points out that most failures are not blameworthy. To create psychologically safe conditions in which we can learn, we need to understand the types of failures we might encounter. Only the third on this list is directly blame-worthy, and final one (in bold) ought to be actively encouraged!

• Basic, often preventable failures (notice and catch them early and correct)​
• Complex failures (lots of variables/unpredictable factors)​
• Malicious failures (negligence, ​recklessness, dishonesty etc.)​
• Intelligent failures – essential for innovation and learning ​
  (problematisation/experimentation/action research/co-creation/​
  contextualisation/practise)​

A much more healthy and productive approach is to think about the system and work together to address the inherent challenges that we face. In doing so, we can:

• Notice and report failure risks​ (to prevent bigger ones)
• Collaborate​ (so we don’t feel isolated, working to address problems together)
• Contextualise solutions​ (to make sure what we do works for us)
• Learn together​ (so we can all thrive)
• Identify systematic causes of small failures as well as the immediate ones (and proactively seeking solutions)

We also need to stop being heroes and martyrs. Edmondson found that nurses worked longer shifts when they did ‘work arounds’ for the problems they encountered. As well as increasing their workload (and that of their colleagues, as everyone reinvented the wheel) this also made the system more prone to complex failures. This is because this behaviour increased the risk of things getting missed. It is better to take the time to report and treat causes, not symptoms and make thing better, for longer, for more people.

Give it time (but don’t leave it to chance)

There are no quick fixes here. Positive change can only meaningfully happen over a sustained period of time within deliberately supportive ​systems in psychologically safe conditions that enable learning for all. Even the teachers. Even the experienced ones. We’re time poor and impacted by the chronic underfunding of education, but supportive systems for CPD seem a good strategy on which to place our bets.

In my EdD, I have explored five secondary school case studies, investigating the relationships between organisational cultures and teacher professional development and learning (development being the activities and learning being the sustained changes to beliefs, knowledge and practices). In each school, I gathered quantitative and qualitative data. The clearest correlation I noted was between high teacher perceptions of the conditions associated with sustained professional learning (agency, efficacy, collegiality, logistics, trust, resilience, reflexivity and autonomy) and what I describe as ‘second order’ professional development activities.

Unlike ‘first order’ PD, with its focus directly on a pedagogical strategy itself, second order activities focus on dedicating time and resources for teachers to contextualise and co-create their responses to first order interventions. This mitigates the sense of ‘faddiness’ or corporate uniformity that many teachers fear (and actively resist) because it demands that we, as professionals use our judgement to critically engage with and explore evidence-informed practices. These can then be trailed and refined as appropriate in their setting, and reviewed later in further second order PD dedicated time.

Unfortunately, busy people in busy schools struggle to make time for second order PD. The evidence from my small study shows that there are, at least, good reasons to commit to making a deliberate and systematic space for such practices. The dividends could be significant.

I would be really interested to hear from you if you work in a school where systems thinking is used, you feel psychologically safe and/or you take a structured approach to ‘second order’ PD activities. Do get in touch!