1. LR

1. LR

1. Using video recordings of lessons and interviews with teachers, this article explores the way in which teachers enact Assessment for Learning (AfL) practices in their classrooms

   • This study took as a starting point the earlier findings of Black and Wiliam’s (1998a) literature review in which pupils were found to improve performance as a result of the application of formative assessment. Similar gains were found in the KMOFAP study.

2. Starting with the hypothesis that AfL is built on an underlying pedagogic principle that foregrounds the promotion of pupil autonomy, we analyse the ways in which teachers instantiate this principle in practice. A distinction is drawn between lessons that embody the ‘spirit’ of AfL and those that conform only to the ‘letter’. The nature and sequence of tasks and especially ‘high organization based on ideas’ appears crucial to the former. This adds a dimension to more familiar formulations of AFL practices. We also ask whether the teachers’ beliefs about learning contribute to the different ways in which they interpret the procedures of AfL.

3. Interviews with teachers indicated that those whose lessons captured the spirit of AfL were more likely to take responsibility for success and failure in the promotion of pupil autonomy. Thus they had a sense of their own agency and sought to use it to overcome barriers to learning.

4. Defining AFL

   • David Hargreaves described Assessment for Learning as ‘a teaching strategy of very high leverage’ (2004, p. 24).
     • The KMOFAP work explored four main areas of classroom practice in relation to formative assessment: questioning, feedback, sharing criteria with the learner, and peer and self-assessment.
       • These procedures were always seen as a ‘Trojan horse’ for a particular view of pedagogy (Black et al., 2006
         • In particular it was conjectured that AfL practices might enable pupils to become more independent learners and for this reason it was central to the Learning How to Learn (LHTL) Project (James et al., 2006). The concept of learning how to learn is explored elsewhere (Black et al., 2006); here the connection between formative assessment and pupil autonomy is pursued in order to contextualize the analysis of classroom observations and teachers’ beliefs about learning, which is the main focus of this article.
           • Black and Wiliam (1998a) argued that ‘the quality of the interaction [between pupil and teacher] … is at the heart of pedagogy’ (p.16). Earlier work on AfL foregrounded the relational element of assessment even more strongly. Arguing against a model of assessment of ‘checklists, precision, explicit criteria [and] incontrovertible facts and figures’, Drummond (2003, p.13) explains her use of the term to ‘describe the ways in which, in our everyday practice, we observe children’s learning, strive to understand it, and then put our understanding to good use’. This view of assessment eschews the model of assessment in which ‘the assessor collects the evidence, makes judgements on the basis of that evidence, and then certain events follow’ (p. 14). Instead the process is seen as ‘essentially provisional, partial, tentative, exploratory and, inevitably, incomplete’ (p. 14). Torrance and Pryor (1998), who distinguish between ‘divergent’ and ‘convergent’ formative assessment, discuss the possi
             • Implicit in the language of divergence and convergence is a sense that certain classroom activities afford more opportunity for pupils to develop independence than others. This echoes Perrenoud’s (1998) understanding of AfL, which he characterizes as the regulation of learning. For Perrenoud the nature of the tasks planned for a lesson significantly impacts on the scope and potential of subsequent interaction and feedback between pupil and teacher as the lesson progresses. He differentiates between ‘traditional’ sequences of activities within lessons, which merely allow for the remediation of narrowly prescribed concepts at the end of the sequence, and those lessons where the tasks are not ‘imposed on the pupils but [adjusted] once they have been initiated’ (Perrenoud, 1998, p. 88) in order to take the learning forward.
               • Yet it is Vygotsky’s and Dewey’s activity-based approaches to learning (see Bredo, 1997) which are most helpful in establishing and extending the link between assessment and progression towards independent learning.1 Much of the work on formative assessment has come out of what might broadly be called cognitive constructivist theories of learning (James, 2006). It is possible, however, to work from a model which arises out of a more socially situated understanding of how learning takes place.
                 • This is less common in the literature of formative assessment though it formed part of the theoretical basis of the KMOFAP work. While Vygotsky and Dewey differed at important points in their understanding of learning, they shared a broadly similar view of the socially constructed nature of learning and desire to promote autonomy (Bredo, 1997; Glassman, 2001).
       • Crucial to appreciating the relevance for AfL of Vygotsky and Dewey’s understanding of learning is the notion of progression toward autonomy, and the teachers’ role in facilitating this through the activities in which they encourage pupils to engage. Most obvious is Vygotsky’s concept of the ZPD (Vygotsky, 1978). Significant, also, is Dewey’s (1966) definition of ‘progressive’ education as ‘high organization based upon ideas’ (pp. 28–29); the challenge is, ‘to discover and put into operation a principle of order and operation which follows from understanding what the educative experience signifies’ (p. 29).
         • Dewey acknowledges that it is ‘a difficult task to work out the kinds of materials, of methods, and of social relationships that are appropriate’ (p. 29). In a sense this is what the teachers on the LHTL project were attempting to do. The implementation of AfL in the classroom, then, becomes about much more than the application of certain procedures—questioning, feedback, sharing the criteria with the learner and peer and self-assessment—but about the realization of certain principles of teaching and learning.
           • The question for the LHTL project, and anyone engaged in professional development, was whether teachers need to share these underpinning principles in order to effect change within the classroom, or whether altering classroom practice through the application of certain techniques is sufficient. The Trojan Horse analogy of KMOFAP arose from the premise that: Teachers will not take up attractive sounding ideas, albeit based on extensive research, if these are presented as general principles, which leave entirely to them the task of translating them into everyday practice. (Black & Wiliam, 1998b, p. 15)
             • For this reason the KMOFAP model, adapted by the LHTL project, emphasized the practical procedures, in training sessions with teachers, as a vehicle for encouraging teachers to engage with the pedagogic principles of AfL. The US-based Capital Project (Coffey et al., 2005), which worked in parallel with KMOFAP, took a slightly different approach, however. Underpinning their research was the ‘perspective’ that: Practice is ripe for modification when teachers begin to understand the nature of the gap between their own current actions and the picture they have of themselves as professionals. In the process of becoming the person or professional they want to be, contradictions between beliefs and actions may be confronted, new belief systems may be constructed, existing beliefs deepened, and often times, risks are taken as new actions or behaviours are tried in the classroom. (Coffey et al., 2005, p. 170).
               • The question of the role of teachers’ beliefs about learning, and their impact on practice, became increasingly significant in the analysis of the data in the LHTL project as a whole, which examined practice/values gaps in teachers’ Questionnaire responses (see Pedder, 2006).