![]()
Dynamic Assessment: fairness through the prism of mediation
Matthew E. Poehner
Department of Curriculum and Instruction, Pennsylvania State University, University Park, PA, USA
Dynamic Assessment (DA) originated in the writings of L.S. Vygotsky more than 80 years ago, and despite its popularity among a small community of specialists it is not widely pursued by educational researchers. This paper seeks to strengthen dialogue between DA proponents and the broader assessment community by identifying potential contributions DA may offer to considering such pressing questions as how assessment may support teaching and learning and how fairness in education may be pursued. It is argued that the defining feature of DA that cuts across the varied approaches that have been developed by psychologists and educators working in diverse contexts around the world is a commitment to looking beyond learner independent performance and examining contribution to joint activity as central to diagnosing abilities. This position prompts a view of teaching and assessment as integrated activity and approaches fairness through the provision of culturally available forms of support. Attention then turns to trends in DA research that are elaborated in the articles in this special issue.
Introduction
Writing in the context of US public schools, where large-scale, standardised tests are employed with ever-increasing frequency to determine student learning and teacher effectiveness, Delandshere (2002, 1480) asks the following:
When the same test is given to all sixth graders in a state to find out whether their educational experiences yield similar achievements, is it because we are working from a theory stating that if students have all been taught the same thing, they all will learn it in the same way at the same time? It seems unlikely that any educator would articulate such a theory. Yet without this perspective, how can current forms of state-mandated assessment be justified?
The use of standardised tests to inform high-stakes educational decisions is by no means a new phenomenon and certainly resonates beyond America’s borders. The full effect of Delandshere’s remarks is that they go beyond simply pointing out discrepancies between assessment and teaching. Indeed, a good deal has been written on the disconnect between practices intended to promote learning and those designed to measure learning outcomes and the potential for these activities to be more closely integrated (e.g., Assessment Reform Group 1999; Black and Wiliam 1998; Torrance and Pryor 1998). Delandshere’s query raises the more fundamental issue of theoretical assumptions about knowledge and learners that underlie teaching practices on the one hand, and assessment on the other. Following Delandshere, awareness of these assumptions is essential for determining how well educational practices in fact reflect intended values and priorities or whether they need to be refashioned to better meet the needs of individuals. At a time marked by growing concern over the role assessment plays in educational policy and the consequences of assessment for individuals, institutions, and society, Delandshere’s query is imbued with a particular urgency.
The articles in this special issue contribute to this discussion by offering a radically different conceptualisation of assessment and its relation to learner development. Dynamic Assessment (henceforth, DA) derives from the Sociocultural Theory of Mind (SCT) formulated by the Russian psychologist L.S. Vygotsky (1987) more than 80 years ago. Vygotsky’s writings compel us to view assessment and teaching not as distinct activities but as dialectically fused. Within this framework, efforts to understand, or assess, learner abilities necessarily involve promoting their development through instructional intervention. Put another way, the object of assessment is fully understood by actively seeking to change it. This orientation requires a shift on the part of the assessor, also referred to as a mediator, whose responsibility is no longer limited to neutrally observing learner performance but now involves engaging as a co-participant with learners. Interpretations of learner knowledge and abilities are broadened beyond observations of independent performance to include their contributions to, and responsiveness during, joint activity with a mediator. In addition, the instructional quality of the interaction begins the process of helping learners move toward overcoming current difficulties.
Despite the tantalising possibilities for organising educational activities offered by DA, it remains on the margins of psychoeducational assessment and is not well known outside of small groups of specialists. To fully understand the contributions DA stands to make to pressing educational problems – as well as the inherent challenges it poses – sustained dialogue is needed between DA proponents and the wider community of educational and assessment researchers. This special issue is intended to create opportunities to advance such dialogue.
The articles in this issue were selected to represent the range of educational contexts and problems to which DA principles have been applied. In addition, these papers offer a selection of the varied approaches to DA, some of which have been around for decades, while others are more recent and are pushing DA work in new directions. As will be clear, Vygotsky’s theoretical proposals have generated many different practices, and indeed his work figures more centrally in some DA approaches than others. What is common across DA approaches is the premise that learner independent performance reveals only a part of their capabilities and that greater insights are gained through interaction intended to support their development. This is a hallmark of SCT, and the prism through which DA addresses questions of fairness and access.
Dynamic Assessment – theoretical background
Vygotsky’s social ontology of human abilities
At the heart of SCT is the premise that human mental abilities develop as a process of gaining awareness of and voluntary control over basic cognitive abilities (e.g., attention, memory, perception) through cultural means. The principle Vygotsky used to explain this process was mediation, positing that instead of acting in the world in a direct manner, humans are unique in acting indirectly through the use of physical and symbolic tools. Wertsch (2007, 178) explains that for Vygotsky mediation is the ‘hallmark of human consciousness’ because it is through their appropriation of the ‘forms of mediation provided by particular cultural, historical, and institutional forces that their mental functioning [is] sociohistorically situated’. Among important symbolic tools are language, alphabetic and numeric systems, maps, models, charts, mnemonics, and even art (Vygotsky 1981). We are also mediated through dialogic interaction with others. Of course, as Karpov and Haywood (1998) observe, these two forms of mediation are not mutually exclusive but often co-occur as individuals rely on both to participate in activities that are beyond their independent capabilities. Through participation, individuals come to rely less exclusively on these external forms of mediation and more on representations of mediation that they have internalised. That is, the operations that were carried out on the external, intermental plane created among individuals come to be carried out on the internal, intramental plane as individuals function more autonomously (Vygotsky 1978).
Wertsch (1985) explains that it was this ontological premise that led Vygotsky to a genetic approach to research, that is, one that traced cognitive functions through the process of their formation, from when responsibility for a function is distributed among individuals to when it is subsequently internalised as part of an individual’s independent capabilities. For this reason, Vygotsky’s (1987) experiments often involved children working through challenges artificially created by the researcher, who was typically present as an interactant, offering symbolic mediation (e.g., wooden blocks or coloured cards) as well as dialogic support. Given the active role of the researcher and the expressed intent to promote development as an essential part of the experiment, it is not difficult to understand how these same principles – in some cases, the very same activities – could be cast as educational opportunities. In other words, the genetic method was a means to overcoming perceived divides between research and practice as both were guided by a commitment to understanding developmental processes through intervention. As will become clear, this dialectical orientation is also at work in DA.
The ZPD – assessment as collective, developmental activity
While Vygotsky himself never employed the term DA, his proposal of the Zone of Proximal Development (ZPD) formalised his argument of assessment and teaching as a dialectic unity. The ZPD is most often described in terms of the difference between an individual’s independent performance and his/her performance in cooperation with others (Vygotsky 1978). This description emphasises the distinction between inter-and intramental functioning, or performance that is externally mediated through interaction with others and performance that is largely or entirely internally mediated through interaction with the self. To this, it is important to add that the ZPD is predicated upon a dynamic social situation wherein mediator and learner contributions and responsibilities are in flux, with the forms of mediation changing in step with learner needs and with learners assuming greater responsibility for performance as they are able. In this sense, the ZPD, despite the spatial metaphor invoked, is not a ‘zone’ into which individuals are placed so that learning might occur but a collective undertaking of individuals working together with mediational means (Holzman 2009). This activity involves learner development through participation, participation that is mediated in a manner responsive to learner needs. Following Aljaafreh and Lantolf (1994), mediation in ZPD activity then is necessarily contingent as mediators must continually position learners to assume as much control over the activity as possible. This is achieved by offering support that is sufficiently explicit to be helpful to the learner but not so explicit that the mediator takes over more of the activity than is necessary.
In addition to its contingent quality, mediation must also be oriented toward uncovering and refocusing the processes learners follow during the activity. In this respect, mediation intends to probe learner reasoning, help formulate plans, draw attention to relevant features of the task, provide hints and reminders, maintain engagement, establish connections with similar or prior tasks, offer feedback, and prompt reconsideration of actions. Given the dialogic nature of ZPD activity, these mediating intentions emerge through negotiation as mediator and learner jointly pose questions, explore possibilities, brainstorm alternatives, and evaluate outcomes (Feuerstein, Rand, and Rynders 1988). Thus, as Poehner (2008a) argues, learner participation in this activity is far more extensive than simply responding to the mediator. The process may be more appropriately described as co-regulation (Fogel 1993) and involves learners’ explicit or implicit moves to request, reject, or clarify mediation.1
Fairness through access to mediation
In reviewing various ways in which the concept of the ZPD has been adopted and applied under the rubric of DA, Minick (1987) astutely observed that the dynamics of the social situation co-created by mediator and learner are the source of its potential to promote learner development but that they also raise important questions when viewed alongside other forms of assessment. Unlike many conventional assessments, particularly in the arena of high-stakes testing, where standardisation of procedures and identification and measurement of discrete behaviours or responses is favoured, the ZPD is predicated on the notion of change. As explained, mediators actively seek to bring about change during DA. ZPD activity then assumes a decidedly activist stance toward learner development, a position that is more implicit than explicit in some conceptualisations of formative assessment (cf. Torrance and Pryor 1998). In the aptly titled book, Don’t Accept Me as I Am, Feuerstein (Feuerstein, Rand, and Rynders 1988), a leading figure in DA, argues that this orientation may best be understood in terms of the assessor’s role in the activity. For Feuerstein, assessors are no longer passive acceptors willingly acknowledging assessment performance as a sufficient and authoritative indicator of an individual’s potential and entire life trajectory; instead, assessors are active modifiers whose priority is to undo predictions based on assessment performance by cooperating with individuals to create a new developmental trajectory.
Feuerstein’s decidedly humanistic view of assessment is certainly in line with Vygotsky, in particular Vygotsky’s work with special needs learners, which carries a strongly activist orientation (e.g., Vygotsky 2004). According to Gindis (2003), Vygotsky distinguished between primary and secondary disabilities. While the former were rooted in biology (e.g., blindness, chromosomal disorders) the latter resulted from the social response to the primary disability, a response that all too frequently involves lowered expectations or even restricting access to schooling. In many cases, Vygotsky argued, the socially created, secondary disabilities are far more damaging to the individual’s life trajectory because s/he does not have the opportunities to appropriate the mediational means of development. As Gindis explains, Vygotsky’s approach to special education was to reconstruct the social and cultural environment to afford alternative and more appropriate developmental opportunities. In other words, the kinds of activities and forms of mediation that are appropriate to learners without disabilities may not be effective for those with special needs, but this does not mean that these individuals are ‘lost causes’. Rather, ‘a child with a disability requires different methods of teaching and learning for his or her appropriation of psychological tools. The sociocultural meaning, however, remains the same and is to be delivered via alternative means’ (Gindis 2003, 209).
Conceiving of assessment as a form of activism entails recognition that, as Feuerstein et al. (1981, 218) put it, the stakes involved ‘affect the lives and destinies of real people’. Rather than systematically withholding developmental opportunities from individuals who perform poorly on tests, educational systems must increase access to available forms of mediation (and act to create new ones). This involves determining the sources of difficulty individuals experience and the forms of mediation to which they are most responsive. In other words, every effort is made to arrange the individual’s social situation to create opportunities for ongoing development and in this respect there is close affinity with some models of formative assessment (e.g., Gipps 1994).2 Fairness in education, from this perspective, does not involve treating all individuals as if they were the same, because doing so ignores that they are not. Fairness requires doing everything possible to maximally support individual learner development, with the understanding that some individuals will need more time and resources than others. Assessment represents a crucial first step toward establishing educational fairness by mediating learners toward success and taking stock of this process as it unfolds in order to determine, with learners, where to go next.
This perspective aligns with what Lantolf and Poehner (2004), following the work of Valsiner (2001), have described as a future-in-the-making model of development. In this view, development is not assumed to proceed along predetermined lines but emerges from participation in activities. ZPD activity renders visible individuals’ future intramental functioning during the course of intermental interaction in the present. However, this future is by no means a given. Rather, the future co-created during ZPD activity is a potential that is contingent upon continued access to appropriate mediation. In this way, the ZPD compels us to view educational practices as fair to the extent that they promote the development of all.
Dynamic Assessment – promises and challenges
In the decades since Vygotsky’s work, a wide range of practices have been developed in various nations that fall under the heading of DA. Consideration of all these approaches is beyond the scope of this paper, and readers are referred to the work of Lidz and colleagues (Haywood and Lidz 2007; Lidz and Elliott 2000). My remarks are limited to identifying orientations to DA that are subsequently taken up and in some cases extended by the contributions to this special issue. In brief, the distinctions among DA approaches considered here fall along two axes, one that concerns the organisation of DA procedures and another that addresses the quality of mediation.
According to Haywood and Lidz (2007), the most widespread approach to organising DA is as a process of test-intervention-retest, wherein a dynamic, mediating session occurs between two non-dynamic administrations of the same or a parallel test. Sternberg and Grigorenko (2002) have described this organisation as a sandwich format because mediation is sandwiched between two testing sessions. In fact, sessions may be delineated as intended either for diagnosis or intervention – although the sequence as a whole may be referred to as the DA. This format resonates more strongly with formal, measurement-oriented approaches to testing than with classroom-based formative assessments (for discussion, see Poehner and van Compernolle this issu...
