Why is self-evaluation not working across schools in Wales?
In this second part of my three-part blog series I’ll be taking a detailed look at why a self-evaluation process that combines top-down decision-making with poorly informed professional judgement is destined for the “superficial, compliance-driven” self-evaluation that Estyn sees as evidence of a failure in the school improvement process.
Schooling in Wales needs to improve. The data we have on standards (Dauncey, 2023) tells us here in Wales, we are way behind the UK average. But why is the main driver of effective school improvement, self-evaluation considered by Estyn to be “superficial, compliance-driven and insufficiently focused on learners’ progress.”(Estyn, no date) If you haven’t yet read part one, click the link to read what the process of self-evaluation looks like in schools in Wales, when it isn’t effective.
The traditional approach to self-evaluation is to start with the idea that a school development plan needs to be populated, and leaders then use their professional judgement to match inspection recommendations with themes they are seeing in their collation of first-hand monitoring evidence.
Using our professional judgement to mentally collate and analyse disparate types of school data is fraught with danger. We are subject to many cognitive biases. There is significant data that shows teacher’s assessment of individual pupils is subject to cognitive bias, making it more likely that disadvantaged pupils’ achievement will be judged more harshly (Doyle et al, 2022). It has also been found that the process of identifying which teacher, or which teacher practices need to be improved, is also subject to the same level of cognitive bias (Luoto et al, 2023)
Often first-hand self-evaluation evidence is largely descriptive and surface-level. Teacher A asked questions, but pupils gave one-word answers, so teacher A needs to improve their questioning techniques. Or books provided little evidence of opportunities for pupils to write at length, so more opportunities should be provided. However, neither of these self-evaluations get close to identifying the root cause of either of these issues.
If you’re thinking this all sounds highly subjective, it is. And we have the additional complication of what works for one child, might not work for another. So, how on earth can we grasp the complexity of the self-evaluation process and improve it so that teachers really understand the root cause of what needs improvement in their classroom and that is what drives professional learning and the priorities in their school development plan.
Let me introduce you to Problem Driven Iterative Adaptation (PDIA) (Andrews et al, 2013). Originally proposed as a framework to help improve public sector capability in developing countries, it has now been widely adopted as a valuable method of tackling complex, adaptive problems. It is also exactly what is needed for school leaders to understand, address and measure the impact of school improvement actions.
So, we need a system of development like PDIA that starts with the actors closest to the action, or as Prof. Graham Donaldson would have described it a system of subsidiarity. The reason for this is because, this is where the real understanding of what is going on and the ability to fix it resides, with the teacher in the classroom.
The second key reason why self-evaluation is not working as it should, is because the cause of the problems being identified are as Estyn quite rightly says are “superficial”. Schools are failing to accurately identify the root cause of their learning and teaching problems, because the whole process is driven by the nee to populate the school development rather than to fix the learning problems as they occur in the classroom.
What schools therefore need is a practical, research-informed approach to flipping the self-evaluation process to start in the classroom and a tried and tested method to identifying the root cause of the problems teachers and learners are experiencing. Only then will self-evaluation have come close to its main purpose, that of driving school improvement.
Watch out for part three of this blog series, in which I will be answering the final of the three questions: “How could self-evaluation be done better?”
References
Andrews, M., Pritchett, L. and Woolcock, M., 2013. Escaping capability traps through problem-driven iterative adaptation (PDIA). World Development, 51, pp.234–244. https://doi.org/10.1016/j.worlddev.2013.05.011 (Accessed: 8 April 2025).
Dauncey, M. (2023) How did Wales perform in PISA 2022? Senedd Research. Available at: https://research.senedd.wales/research-articles/how-did-wales-perform-in-pisa-2022/ (Accessed: 30 March, 2026)
Doyle, O., Easterbrook, M.J. and Harris, P.R. (2022) Roles of socioeconomic status, ethnicity and teacher beliefs in academic grading. Available at: https://pubmed.ncbi.nlm.nih.gov/35998351/ (Accessed: 25 March 2026).
Estyn (no date) Annual report. Available at: Estyn Annual Report (Accessed: 25 March 2026).
Luoto, J., Klette, K., & Blikstad-Balas, M. (2023). Possible biases in observation systems when applied across contexts: Conceptualizing, operationalizing, and sequencing instructional quality. Educational Assessment, Evaluation and Accountability, 35(1), 105–128. https://doi.org/10.1007/s11092-022-09394-y