In many of Dylan Wiliam’s talks and publications he references five ‘key strategies’ that support the implementation of effective formative assessment. The five strategies each get a chapter in his excellent book Embedding Formative Assessment (2011) which builds on the work he developed with other colleagues in the 90s and 00s.
The five strategies were expressed as early as 2005:
- Clarifying, understanding, and sharing learning intentions
- Engineering effective classroom discussions, tasks and activities that elicit evidence of learning
- Providing feedback that moves learners forward
- Activating students as learning resources for one another
- Activating students as owners of their own learning
Very commonly, Wiliam presents these ideas in this helpful table, linking the strategies to core assessment concepts:
In my work as a consultant and teacher trainer, I give a lot of ‘evidence-informed’ advice to teachers. Of late, this has been influenced largely by discussions about a knowledge-rich curriculum and my reading of Rosenshine’s Principles of Instruction, cognitive load theory, and various other papers linking cognitive psychology to classroom practice.
However, it occurred to me recently that most of this overlaps entirely with Wiliam’s five strategies and that is what I want to explore here. To some extent, schools and teachers often feel they have have ‘done AfL to death’ in countless CPD sessions over the last 15 years. Time was when you couldn’t get a job unless you said ‘AfL’ about 12 times in an interview. Sadly, my sense is that the wisdom at the heart of Wiliam’s ideas about responsive teaching/formative assessment gets washed out either a) by the delusion that the strategies are already embedded in day-to-day practice or b) by the sense that this is a box ticked and people are really ready to move to the new thing. Truth be told, a lot of ‘AfL’ was and is a mile away from the formative assessment practice Wiliam is talking about.
Essentially, I feel that, among the important things every teacher should know, the five strategies should be there, part of the core curriculum for teacher development. Here’s how I see it all connecting:
1.Clarifying, understanding, and sharing learning intentions
Wiliam says ‘if you don’t know where you’re going, you’ll never get there’. This is largely about curriculum planning. I read ‘learning intentions’ as meaning: what do we want all students to know and be able to do? In the detail, this means spelling out what knowledge – in all its forms – they should have and how to apply this knowledge in new contexts. It chimes perfectly with the wave of work being done around curriculum design. It also resonates with the strand of Rosenshine’s Principles of Instruction around sequencing concepts, providing models and appropriate scaffolding.
It also means ‘what does excellence look like?’. This connects to ideas about assessment and comparative judgement and teachers knowing the standards. Significantly, the implication from Wiliam is that in ‘clarifying, understanding and sharing’ – teachers, students and their peers all need to know both the knowledge requirements and the criteria for excellence in any performed task. This goes far, far beyond writing a mandatory one-line LO on the board at the start of every lesson! (Aarrghh!). It suggests a lot of very explicit exposition and discussion about the target knowledge and the features of any endeavour that constitute ever increasing degrees of success.
This, in turn, feeds into ideas about self-regulation and metacognition. Successful learners will be good at self-regulation, planning and monitoring their progress towards learning goals in a deliberate self-directed manner. Knowing the learning intentions very well is essential for that process to work.
So the links here are numerous: curriculum, knowledge, standards, self-regulation, scaffolding, modelling.
2. Engineering effective classroom discussions, tasks and activities that elicit evidence of learning
In some ways, this ‘strategy’ is a one-line summary of most of the rest of Rosenshine’s Principles of Instruction. ‘Discussions, tasks and activities’ covers a lot of possibilities. At the centre of it is the idea of ‘responsive teaching’. Instructional teaching has to be highly interactive so that teachers are getting feedback from their students about how well their schemas for the material in hand are forming and how fluent they are becoming retrieving and using what they’ve learned. The challenge for teachers is to involve as many students as possible which leads to the need for good questioning routines and good knowledge-check routines where the ratio of student involvement is high and the information received has a good diagnostic component.
Rosenshine talks about the need for checking for understanding and asking lots of questions in a probing style. Wiliam focuses on question design – including good diagnostic multiple choice questions – and the role of all-student response techniques.
Links: Rosenshine’s Principles of Instruction. Shimamura’s ‘Generate-Evaluate’ model. Ideas about retrieval practice. Nuthall’s ideas about ‘hidden lives’ and the idea that we can’t be remotely confident about learning taking place until we check – now, and again later.
3. Providing feedback that moves learners forward
Feedback is a thorny issue, woven into discussions about the use of formative and summative assessment, marking and workload, grading and the value of data as a tool to improve learner outcomes. The key in Wiliam’s work is the emphasis on moving learners forward. It’s this thinking that informed the ideas I expressed in this ‘feedback as actions‘ post.
Some of the key messages that Wiliam offers in relation to feedback that I cite very often are:
- Feedback is only successful if students’ learning improves – and this depends on their capacity to understand it and inclination to accept and act on it. It’s got an interpersonal, motivational element that can’t be brushed aside. Giving feedback isn’t a purely technical, objective task – although it does have to suggest actions students can actually take rather than offering a nebulous retrospective critique.
- The goal is to change the students’ capacity to produce better work, not just to improve their work. Austin’s Butterfly is wonderful – because it shows what effective feedback can achieve – but Austin has only truly benefitted if, later, he is more able to ‘look like a scientist’ or draw beautiful butterflies without feedback: he needs to be able to generate his own feedback and become more independent.
This links formative assessment to metacognition and self-regulation and Rosenshine’s ideas about moving from guided to independent practice. If we’re still reliant on external feedback to tell us if we’ve succeeded (SatNav style), then we’ve still got a long way to go. Effective learners can link their work to the success criteria and generate their own ongoing self-correcting feedback narrative.
Links: Ethic of excellence, Rosenshine guided to independent practice, self-regulation.
4. Activating students as learning resources for one another
I think this is the feature of Wiliam’s five strategies that deserves more attention. All too often teachers create major bottlenecks by forcing all classroom interactions to pass through them. However, if teachers develop strong routines where students support each other’s learning in a serious structured manner, then the ratio, quality and frequency of student interactions with the knowledge in hand can increase significantly. We can’t have a dialogue with every student at once but they can all be involved in meaningful dialogues with each other to support the process of working out ‘where the learner is’ and ‘how to get to where the learner is going’. This is where disciplined ‘think pair share‘ becomes so powerful.
Wiliam cites Slavin in showing that well-designed collaborative learning can yield significant gains – but it has to be done such that everyone is learning. There are so many ways to do this e.g students checking their partners’ answers using all manner of quizzing formats and generative processes and elaborative-interrogative questions (why? how?). Pairs are probably the most efficient and effective use of this strategy – because of the ease of switching in and out of the interactions. If one person in a pair acts as the verifier for the other, using exemplars, fact sheets, mark schemes as a reference, the extent of retrieval practice and feedback can be increased hugely. Another example might be using structured dialogues for practising the use of language or rehearsing explanations and arguments. Provided that there is a strong process for evaluating students’ responses for accuracy and quality, a high volume of peer-to-peer interactivity is powerful.
Links: Hattie’s ‘reciprocal teaching’, Shimamura’s ‘think it, say it, teach it’, Slavin’s collaborative learning, Sumeracki and Weinstein on elaborative interrogative questions and retrieval practice.
5. Activating students as owners of their own learning
In all honesty, I find that implementation of the strategy behind this feel-good-phrase, often falls into the dust of ‘noble intent’ rather than delivering something tangible. However, it is actually highly actionable and links directly to many other ideas. ‘Owning your own learning’ is at the heart of strong self-regulation and metacognition: setting learning goals, planning, monitoring and evaluating success in tasks links to those goals; forming effective schemata that take account of big-picture questions and themes that inform subsequent conscious rehearsal and elaboration. However, these ‘goals’ are not broad brush life goals; they are learning goals – the next steps in improving writing fluency, science knowledge, confidence with maths and languages, physical fitness etc.
The point is that these characteristics of effective learning can be fostered by setting up good routines and expectations. Teachers can help students to know where they are going and where they are on the curriculum journey. This can be supported by:
- giving students access to long-term topic plans, the syllabus, the wide scope overview before diving down into the details;
- setting out milestones in the progress journey so that students can take their bearings and plan their own next steps through appropriate forms of practice, becoming increasingly independent.
- setting out clear relational models for conceptual schema building – as per Shimamura’s Relate in MARGE.
- providing exemplars of performance at various levels of success up to a high/exceptional level so students can compare their own work against a scale and see for themselves where they are and what short-run learning goals might be achievable to move forward.
If a student knows for themself what they need to do in order to improve and gains the experience of being able to achieve success through applying effort to these self-determined goals, then they begin a positive upward spiral of confidence building, growth mindset-inducing, self-regulation that fuels even more success.
Links: Rosenshine: practice; Shimamura: Relate; Growth mindset; self-regulation.
To some extent I feel that the issue has been that ‘AfL’ or even ‘formative assessment’ has been too broad a term; too much of a catch-all, thereby allowing various degrees of corruption and dilution to take root. I think that it’s when you get into understanding and deploying the five separate strategies that it finds form. That’s the understanding of formative assessment that teachers need. It’s powerful stuff, right there, where it’s been for years.
Wiliam, D., & Thompson, M. (2007). Integrating assessment with instruction: what will it take to make it work? In C. A. Dwyer (Ed.), The future of assessment: shaping teaching and learning(pp. 53-82). Mahwah, NJ: Lawrence Erlbaum Associates.
Wilam, D (2011) Embedded Formative Assessment. Solution Tree Press.