Moving towards the formative assessment of CPD

SAM GIBBS, Trust Lead for Curriculum & Development, GMET

NIKKI SULLIVAN, Deputy Headteacher, Beckfoot School

In this article, we consider why the summative assessment of teaching and professional learning may persist, along with the associated problems. Secondly, we explore how the formative evaluation of CPD (continuous professional development) might support leaders in not only assessing the impact of professional development but also creating empowering staff learning cultures where school improvement and people development sit hand in hand.

High impact, high accountability and a ‘fuzzy feedback loop’

We know that there is a worrying disparity between the impact of highly effective versus less effective teaching, especially for students from disadvantaged backgrounds (The Sutton Trust, 2011). We also know that leadership that promotes and participates in teacher development has a positive correlation with student outcomes (Robinson, 2007). School leaders rightly seek to hold themselves accountable for the quality of their CPD provision.

Therefore, it is perhaps unsurprising that in an education system that also operates with high levels of external accountability (OECD, 2015), summative assessments of teaching quality persist, with TeacherTapp surveys indicating that 10 per cent of schools still engage in graded lesson observations (Fletcher-Wood, 2023).

However, such methods are likely to be highly inaccurate, due to the ‘fuzzy feedback loop’ (Mccrea, 2023) between teaching and student learning – learning is invisible and happens over time. As Wiliam asserts, ‘even if we combine classroom observations, measures of student progress and student surveys, the ratings of individual teachers are still not very accurate… we would need to collect data on each teacher for eleven years’ (Wiliam, 2018, p. 43).

If we think of CPD as teacher learning, it is this same ‘fuzzy feedback loop’ that makes the impact of CPD persistently hard to evaluate. In fact, when we combine the invisibility of learning with the ‘knowing–doing gap’, the issue is further compounded. Were we to (unwisely) engage in simplistic summative assessments of teacher knowledge, this would not necessarily mean that a positive change in classroom practice had occurred.

If we wait until ‘download day’ in August to attempt to determine the impact of our professional learning, the multitude of inputs means that we would struggle to directly attribute any improvement (or lack thereof) solely to school CPD. In addition, when we focus on what Wiliam (2016) refers to as ‘lagging indicators’ of success, we may overlook the ‘leading indicators’ and consequently lack clarity about how we might improve provision.

The consequence: CPD is important and we need to get it right (and, some might feel, try to evidence that we are getting it right?!). However, it can be complex to evaluate impact. Ultimately, inherent challenges of assessment in our work with students – reliabilityIn assessment, the degree to which the outcome of a particular assessment would be consistent – for example, if it were marked by a different marker or taken again and validityIn assessment, the degree to which a particular assessment measures what it is intended to measure, and the extent to which proposed interpretations and uses are justified, correlation versus causation – also manifest when we attempt to evaluate teacher learning and associated CPD provision.

‘Shifting the dial’: Wiliam’s five formative assessment strategies 

Recognising, therefore, the limitations of summative methods as a ‘blunt tool’ with which to evaluate teaching and CPD, a focus on formative methods might serve us better. This formative lens – the purpose of which is to ‘move the learner forwards’ – could further shift the dial to a system where we seek to ‘improve not prove’ (Moyse, 2019).

Fortunately, there already exists a range of formative frameworks upon which schools can draw. Kirkpatrick (1959) produced a taxonomy for the evaluation of training programmes, beginning with reaction, before progressing through to learning, behaviour and results. Kaufman and Keller (1994) later built upon this model, creating a distinction between training resources and delivery within learner responses, and through the addition of a final level entitled ‘societal outcomes’.

Brinkerhoff’s Success Case Method (2003) compares the most and least successful participants in order to better understand what next steps to take, considering two key questions: ‘How well does a programme work in a best-case scenario? and ‘When a programme doesn’t work, what’s the reason for this?’

We propose that Wiliam’s formative assessment framework (2011) provides potentially useful parallels. The prerequisite conditions for meaningful assessment are clarifying, sharing and understanding learning intentions. CPD should be based on need, where school evaluation has identified priorities that in turn become our professional learning intentions. Teachers and leaders need to know where they are going and what success will look like. As Fletcher-Wood says, ‘teacher learning is just learning: plan accordingly’ (2017).

This focus on clarifying intentions is reflected in the work of Guskey, quoted in Earley and Porritt (2010): ‘Good evaluation does not need to be complex; what is necessary is good planning and paying attention at the evaluation at the outset of the professional development program, not at the end.’ This clarity subsequently enables us to determine the ‘effective discussions, tasks, and activities that elicit evidence of learning’ – another of Wiliam’s (2011, p. 2) strategies. We need to be looking to shape a CPD curriculum (Enser and Enser, 2021) built around a ‘rhythm of inputs’ (Cottingham, 2022), resulting in ‘small micro effects [which] cumulatively create a macro impact’ (Weston and Clay, 2018, p. 56).

From here, CPD leads can provide responsive feedback, while also creating opportunities for peer-to-peer collaboration, where colleagues become learning resources for one other. The ultimate goal is for assessment to make itself redundant: students – or teachers – are owners of their own learning, able to self-diagnose and direct their own path. Each colleague has a different starting point and will acquire and develop knowledge and skills at an individual pace. Some will be able to self-direct before others; assessing their learning formatively, however, ensures that teacher educators are focused on matching provision to need.

We contend that not only does the application of Wiliam’s model (2011) avoid crude summative judgements, but it also necessitates a partnership between the teacher and those involved in their learning: the responsibility of the teacher to engage with and direct their professional learning is balanced with that of the CPD lead to continually improve the quality of the provision. It requires us to focus on the questions that help us to improve our offer: Have we been sufficiently clear about what teachers need to know? Have we provided the right balance of activities? Is feedback moving teachers forward? Are colleagues moving towards self-direction? How should we adapt provision if impact is less than we intended? Crucially, we are not just able to consider ‘what is working well and what is working less well’, but we are also able to consider for whom. While considering the individual needs of colleagues and the quality of their learning journey, we also gather meaningful data that enables continual improvement of provision.

Current context and addressing complexity through collaboration

There appears to be an increase in momentum around the interest in quality CPD across the sector. This is reflected in the report from Conyard et al. (2024), who describe recent CPD developments in the system (including the recent ‘golden thread’ reforms) and the associated ‘sharp upturn’ in the interest in CPD from leaders and teachers.

Although we have this positive context, recent CPD provision remains inconsistent. TeacherTapp (Allen et al., 2024) reported that ‘only just over one-in-ten teachers “strongly” agree that they have taken part in professional development activities that have helped them become better teachers over the last 12 months’ (p. 13).

Additionally, if we are not careful, what we might end up with is a CPD ‘Matthew effect’: schools who are already outward-facing continue to engage and take advantage of large-scale developments in the system, while others continue to be less aware of the opportunities available. The CPD-rich get richer and the CPD-poor get poorer. Again, we are pleased to see this mirrored in the report from Conyard et al. (2024), who state the need for CPD to be ‘an expectation… embedded [as] part of a professional teaching career, rather than an entitlement through which CPD is accessed’ (2024, p. 6).

As the system continues to focus on effective and impactful CPD, we also need to seek and share solutions to the effective evaluation of CPD; consequently, crude summative judgements, where they still exist, can be replaced by effective and purposeful lenses.

This is why we are equally excited by the trusts, schools and networks looking to find solutions, not just for themselves, but across their systems and for the schools down the road. Networks such as the Trust-Wide CPD Leaders’ Forum create spaces for professional engagement and critical discussion. The real strength of such developments is, as Stoneman discusses, where leaders take collective responsibility for their own students and for students across the sector. In the same way as ‘they are all our children’ (Stoneman, 2022), so too are they all our staff if we want to help to increase equity for our young people.

To conclude

Despite the context of high accountability where schools are assessed summatively, we have a responsibility to create a greater shared understanding of how to formatively evaluate the impact of CPD based on a broader conception of quality.

Creating enabling cultures for teacher learning and development must be a high priority for school leaders. Collaboration between schools and MATs (multi-academy trusts) is essential to achieve this end, as we develop a shared understanding in the system.

One thing is certain: attempting to summatively assess teaching quality and the impact of CPD is inaccurate and ineffective – a focus on formative frameworks, threading through from clear learning intentions to enaction in the classroom and outcomes for learners, is where we need to be focusing. In doing so, we might not only gather information that we can use to refine our CPD provision, but also create the conditions for our teachers to thrive.