POEHNER The Pennsylvania State University University Park, Pennsylvania, United States Dynamic assessment DA offers a conceptual framework for teaching and assessment according to
Trang 1Group Dynamic Assessment: Mediation for the L2 Classroom
MATTHEW E POEHNER
The Pennsylvania State University
University Park, Pennsylvania, United States
Dynamic assessment (DA) offers a conceptual framework for teaching and assessment according to which the goals of understanding indi-viduals’ abilities and promoting their development are not only com-plementary but are in fact dialectically integrated More specifi cally,
DA follows Vygotsky’s proposal of the zone of proximal development (ZPD)
by offering learners external forms of mediation in order to help them perform beyond their current level of independent functioning (Vygotsky, 1986) A major challenge to implementing DA in second language (L2) classrooms is that these contexts typically do not per-mit the one-to-one interactions that have characterized most DA work
to date (and ZPD research more generally) This article explores the use of DA with groups of classroom L2 learners rather than individu-als Group dynamic assessment (G-DA) applies the same principles of mediation as in individualized interactions but broadens the focus to potentially an entire class Vygotsky himself recognized the possibility
of constructing a group ZPD by negotiating mediation with more than one individual (see Vygotsky, 1998) Transcriptions of G-DA interac-tions involving L2 classroom learners are presented It is argued that organizing classroom activity in this way enables teachers to explore and promote the group’s ZPD while also supporting the development
of individual learners
In both general education and the second language (L2) teaching domain, there is renewed interest in relating assessment to teaching and learning, and it is within this context that dynamic assessment (DA) has begun to attract attention from researchers and practitioners DA dif-fers sharply from most other forms of assessment in its basic premise that fully understanding abilities requires active intervention in their develop-ment and consequently shifting focus from products of prior learning to the processes through which abilities are formed At the level of practice, the more familiar assessment model in which teachers observe student performance is replaced by one in which teachers and students jointly carry out activities, with teachers intervening as necessary to help learn-ers stretch beyond their current capabilities This mutual engagement functions simultaneously as instruction and assessment because teachers
Trang 2not only discover where learners encounter problems but they also
pro-vide support to overcome them
BACKGROUND
The theoretical basis for DA is found in the writings of the Russian
psy-chologist L S Vygotsky, and particularly his conceptualization of the
zone of proximal development (ZPD) Vygotsky held that humans, unlike
other animals, interact with the world in a mediated rather than a direct
manner, employing culturally specifi ed symbolic tools such as signs,
sym-bols, counting systems, and language to understand and transform the
world through concrete activity Furthermore, social interactions play a
leading role in our own development because cognition emerges through
engagement with others One of the hallmarks of Vygotsky’s theory is his
position that the mind is not contained within an individual’s head but
includes interaction with others and the use of physical and symbolic
arti-facts This perspective is the basis for his well-known distinction between
intermental and intramental functioning As Vygotsky argued, new cognitive
functions appear twice during the course of their development, fi rst on
the intermental plane as individuals work collaboratively with others and
with cultural artifacts and later on the intramental plane, when these
functions have been internalized and may be carried out by the
individ-ual without external support (Vygotsky, 1978)
Vygotsky argued that instruction should not target fully formed
abilities but should operate within the ZPD to optimally guide learner
development:
Since teaching depends on immature, but maturing processes and the
whole area of these processes is encompassed by the zone of proximal
development of the child, the optimum time for teaching both the group
and each individual child is established at each age by the zone of their
proximal development (Vygotsky, 1998, p 204)
In this way, the ZPD provides a theoretical framework for the integration
of teaching and assessment: For teaching to be most effective, it must be
attuned to the ZPD, which entails an assessment that provides mediation
to help learners perform beyond their present capabilities; at the same
time, this mediation constitutes teaching intended to support learners’
developing abilities As I have argued elsewhere (Poehner, 2007, 2008b),
the ZPD demands a revolutionary pedagogy in which assessment and
teaching are fused as a single, development-oriented activity This is
pre-cisely the aim of DA researchers, who have devised various approaches to
co-constructing ZPDs with learners (see Lantolf & Poehner, 2004)
Regardless of the specifi c approach followed, DA proponents have
shown a strong preference for one-to-one, mediator–learner procedures
Trang 3Vygotsky (1998) describes the ZPD as “the optimum time for teaching both the group and each individual” (p 204), but he does not elaborate the group ZPD in his published work Since Vygotsky’s time, the numerous studies of learning and development that have invoked the ZPD have like-wise favored a dyadic expert–novice or mediator–learner model The L2
DA has also favored a dyadic model (see, e.g., Poehner, 2008a, 2008b; Ableeva, 2008) Targeting the development of an individual no doubt appears a more manageable undertaking than mediating a group of learn-ers, but in some ways it represents an unrealistic model for classroom teach-ers, who typically must engage classes of 15 to 30 learners (and in some cases many more) Although both general and L2 DA work have done much to offer teachers principles and models to consider, this matter of scale has not been adequately addressed The purpose of the present article
is to outline how DA may be organized to function within a class’s ZPD
In what follows, I argue that a major impediment to creating a room ZPD is that to date little serious work has been done to articulate a
class-theory of group Indeed, it is widely held that groups may be formed by
simply placing individuals together and assigning tasks to be performed, with little or no consideration given to whether such groups constitute socially coherent units This practice has become commonplace, not only
in experimental psychological research but in various oriented language pedagogies (Brandl, 2007; Lee & VanPatten, 2003; Savignon, 2002) What is missing from such work is an articulation of the relationship between individual development and that of the group, con-ceived as a psychological entity in itself Without such an understanding,
communicative-it is diffi cult to imagine how group activcommunicative-ities may be organized to reveal individuals’ abilities and to support their continued development Even recent research in the area of cooperative learning, despite its connec-tions to Vygotskian theory (e.g., McCafferty, Jacobs, & DaSilva Iddings, 2006), has largely overlooked the developmental dynamics of groups The same is true of models of group-based assessment (He & Dai, 2006; Hilsdon, 1995; Van Moere, 2007) As I will show, Vygotsky’s perspective
on the social origins of mind, and in particular the extension of his ory to the study of group functioning by the Russian psychologist
the-A V Petrovsky (1985), offers a powerful way of thinking about the relations between individual and group ZPDs that can guide the imple-mentation of G-DA However, more needs to be said fi rst about how individual–group relations have been traditionally understood
Group and Individual
As should be clear, Vygotsky assigned a leading role to the social world
in psychological development As his close colleague A R Luria (1970)
Trang 4put it, it is through internalization of culturally specifi ed forms of
media-tion that “the social nature of people comes to be their psychological
nature as well” (p 45) For education, the interconnection between social
activity and development of the mind implies that teaching need not wait
for learner readiness but that it may take a leading role in helping
learn-ers chart new developmental trajectories
Moving from a one-to-one model of teaching and assessment to a
group-focused approach requires an understanding of the relation
between development of individuals and development of the group In
Vygotskian terms, this is the relation between the individual’s and the
group’s ZPD Petrovsky (1985) explains three broad perspectives on
indi-vidual and group abilities that have emerged in psychology The fi rst,
which I will call group-as-context , has dominated research in the West
According to this view, the group itself is not psychological in its own
right but simply serves as a static backdrop to the performance and
devel-opment of the individual By limiting the defi nition of group to “a chance
association of people with nothing connecting them except time and
space” the group-as-context view ignores the question of group abilities
(Petrovsky, 1985, p 66) The group is nothing more than an “aggregate
of interrelated and interacting individuals” whose “purposeful activity” is
unimportant for understanding individual cognitive processes (p 106)
Petrovsky points out that, on this view, the group itself is regarded as a
contextual variable that exercises a “mechanical infl uence” on the
psy-chology of the individual, who remains the analytical focus (p 66)
The group-as-context perspective clearly informs many current
prac-tices in group-based assessment, where every effort is made to parse out
the performance of the individual to infer his or her underlying abilities
as they are observed in the group setting For instance, Webb (1992)
con-ducted a metastudy of group-based assessment procedures and concluded
that although group assessments are a useful approach to evaluating
learn-ing, their results do not provide an appropriate basis for making
infer-ences about individuals’ abilities Instead, he argues, one must distinguish
individuals’ abilities from what they are able to do when working with
oth-ers on the grounds that the group setting only obscures the true focus of
assessment, the individual (see also Lejk, Wyvill, & Farrow, 1996)
The other two approaches to relating individual and group abilities
that Petrovsky describes both understand the group to have
psychologi-cal status and stress the importance of the activity in which group
mem-bers are engaged They differ concerning the degree of individual
identifi cation with the group Specifi cally, in a group-as-cooperation model,
each individual retains his or her own goals while understanding their
interrelation with the goals of other group members In a group-as-collective
model, all are united in working toward a common goal Petrovksy
explains that the group-as-collective model involves the most highly
Trang 5devel-oped form of social relations among individuals because it is guided by the humanistic principle “for others as for oneself” or “to oneself as to others” (Petrovsky, 1985, p 191)
Drawing on his own clinical work, Petrovsky (1985) relates these forms
of group to the shifting orientations of individuals participating in group therapy (pp 160–161) He notes that at the outset, participants under-stand the object of group therapy to be their own recovery; the group merely offers a context for interaction Group members view one another
as merely chance associations As sessions progress, patients come to regard others as important for their own recovery That is, individuals are still each oriented toward overcoming their own problems but they now begin to value cooperation with others and the role others can play in facilitating their recovery In some instances, the groups eventually adopt
a collective orientation, in which they not only see others as instrumental
in their own recovery but the goal of the activity shifts from the individual
to the group, and each works to bring about the recovery of every group member, an orientation Petrovsky captures with the axiom “The health
of each is the wealth of all” (p 161)
Group Activity
The importance of both cooperative and collective activity for chological development is that as individuals recognize the interde-pendence of their goals, “the specifi c character of the relations being formed [among group members] demands from each person involve-
psy-ment in joint intellectual activity , and the pooling of psy-mental efforts in order to
overcome diffi culties” (Petrovsky, 1985, p 183, italics added) Petrovsky’s
notions of joint activity and pooling efforts require some explanation Joint,
in contrast to solo, activity denotes contributions from more than one individual to realize individual and group goals The initially competitive nature of social relations among group members gives way to coopera-tion as they come to understand their own development to be a func-tion of the group’s activity, the success of which ultimately depends on each member’s contributions (pp 186–187) The shifting dynamics of individual relations and goals within the group transform the group into much more than an assemblage of individuals The group itself begins
to function as a psychological entity composed of individuals with ferent forms of expertise working cooperatively to carry out activities that no single group member could do independently To be sure, this way of thinking about group activity is markedly different from what one fi nds in many language classrooms, where group work is often con-strued as a time for learners to use the language with little regard for the object of their communication and its signifi cance for development
Trang 6dif-(see Magnan, 2008) For Petrovsky, group work is much more than an
opportunity for communication—it is purposeful activity in which
com-munication is crucial because it is the sharing of knowledge and abilities
that moves the group forward in its ZPD while also benefi ting
individu-als (Petrovsky, 1985, p 183) The changing nature of social relations and
goals, then, implies that development of the group and development of
the individual are increasingly interconnected
Hutchins and Klausen (1996) offer an example of interconnected
social relations and goals in their analysis of how a fl ight crew interacts
while fl ying a commercial airliner According to these authors, “a larger
unit of cognitive analysis” foregrounds not the actions of individual crew
members but rather the coordinated efforts of the group, each with
dif-ferent expertise, as they fl y the plane—something none could accomplish
independent of the group (p 17) 1 As instructive as this example may be,
it is also the case, as Petrovsky (1985, pp 171–172) reminds us, that
col-lectives emerge more readily during some activities than others For
instance, the divisions of labor and different expertise that may be
observed in the cockpit example are inherent in certain highly complex
activities and essential to achieving desired outcomes (e.g., surgical
pro-cedures, industrial manufacturing), whereas other contexts, such as
for-mal schooling, tend to favor individual achievement Petrovsky (1985)
explains that “genuinely collective academic activity” is extremely diffi cult
to realize in practice both because there is often no obvious division of
labor (other than, say, teachers and students) and also because schooling
culture “mainly envisages an individualized process of mastering
knowl-edge” (p 181) 2 However, Petrovsky does recognize collectives in school
settings as a possibility and assigns primary responsibility to the teacher:
“Correctly organized group work does not reject, but necessarily
stipu-lates the leading role of the teacher” (p 186) Teachers take the leading
role not because they can simply transform groups of learners into
collec-tives Rather, by engaging learners in tasks that are challenging to all and
providing support to benefi t all, teachers may foster a more cohesive
1 The Hutchins & Klausen (1996) example of cognition in an airline cockpit fi ts with models
that have been broadly defi ned as distributed cognition (see, e.g., Salomon, 1993) Research
on distributed cognition emphasizes the importance of a strict division of labor with
contri-butions made according to individuals’ previously developed forms of expertise In this
regard, labor activities would appear to differ markedly from educational contexts, where
the object is not simply to contribute based on prior learning but to develop new abilities
through collaboration This would most likely be achieved through shifting roles and
responsibilities among group members rather than through a static division of labor
2 Donato (1994) provides an example of what he argues is a collective orientation among
university undergraduate French L2 learners engaged in group work He notes that one
group’s cohesion became apparent in their discourse (e.g., a prevalence of “we” rather
than “I” and “you” as they addressed one another) as well as in the ease with which they
changed roles, and the lack of overt turn-taking markers (p 40) However, this orientation
appears not to have been the result of teacher intervention but emerged on its own
Trang 7orientation to classroom activities on the part of students, an orientation
in which learners share a common goal of solving the problems at hand and appreciate the contributions of others for the realization of both this common goal as well as more individualistic goals, such as demonstrating profi ciency to earn grades Put another way, by organizing instructional activities around the group’s ZPD, teachers may help the group itself to become psychological in the sense that the development of the group and the individual are interrelated In G-DA, as I show in the next section, two approaches have emerged to bring about such group cohesion
GROUP DYNAMIC ASSESSMENT
Group-based and one-to-one DA procedures follow the same general principle of offering learners mediation to help them co-construct a ZPD, but they differ in that that G-DA must also take account of the group’s ZPD As explained earlier, G-DA entails understanding the group to be not merely a context for individual performance but a social system in its own right that might be supported to function in ways that are beyond the present capabilities of any individual member G-DA then must engage the group in an activity that no individual is able to complete indepen-dently but for which all members require mediation, albeit at different levels and different quantity Of course, the forms of mediation needed will likely vary from learner to learner at any given point in time Bearing
in mind the notion of a shifting focus of instruction, teachers may fi nd it useful to track development of the individual’s ZPD within the broader group ZPD by foregrounding the support offered to an individual within the group and to examine his or her responsiveness However, for G-DA
to effectively promote the development of all group members, the teacher must actively engage the entire group in G-DA interactions
Of course, engaging the entire group does not mean that the teacher should not offer mediation to individuals, but that every mediating move should also be directed to the group In this regard, it is useful to distin-
guish primary from secondary interactants in G-DA In the event that the
teacher offers mediation in response to a given learner’s diffi culty, that learner and the teacher are the primary interactants as they negotiate the support that is needed However, because the exchange occurs in the social space of the class and before the other group members, it has medi-ating potential for the rest of the group as well, who are secondary inter-actants but participants nonetheless Similarly, the teacher may pose a leading question or provide a prompt to the entire group to which individ-ual learners will respond Once again, the respondents may engage more overtly with the teacher but the exchange may simultaneously benefi t sec-ondary interactants, a notion substantiated by research on private speech
in language classrooms (Lantolf & Yáñez-Prieto, 2003; Ohta, 2001)
Trang 8In what follows, I offer examples of two different approaches to G-DA
with L2 learners In the fi rst, which I refer to as concurrent G-DA , the
teacher dialogues with the entire group To be sure, the teacher may
pro-vide mediation in response to an individual, but the interaction shifts
rapidly between primary and secondary interactants as one learner’s
question, struggle, or comment sets the stage for another’s contribution
In this way, concurrent G-DA may appear to an observer to be similar to
whole class instruction, but of course the absence of extended
one-on-one interactions does not preclude development within individuals’
ZPDs In the second approach, cumulative G-DA , the teacher conducts a
series of one-on-one DA interactions as the group works toward mastery
of a problem That is, individuals take turns engaging directly as primary
interactants with the teacher, with the understanding that each
subse-quent one-on-one exchange will have the advantage of building on
ear-lier interactions that the class witnessed This approach is cumulative in
that the goal is to move the entire group forward in its ZPD through
negotiations with individual learners in their respective ZPDs Cumulative
G-DA attempts to move the group forward through co-constructing ZPDs
with individuals, but concurrent G-DA supports the development of each
individual by working within the group’s ZPD
The following discussion offers a brief illustration of concurrent G-DA
with L2 learners and then moves to a more extended discussion of L2
cumulative G-DA The reader should not interpret the relative length of
discussion to mean that cumulative G-DA is more effective or somehow
preferable to concurrent G-DA Rather, it refl ects the recent emergence
of cumulative G-DA research—at the time of writing, ongoing projects
have yielded more developed examples of cumulative than concurrent
G-DA 3 For the purpose of illustration, I will briefl y comment on Gibbons
(2003), which describes a classroom teacher’s efforts to scaffold a group
of English language learners I will limit my remarks to explicating how I
believe the teacher co-constructs a group ZPD with her students (for
fur-ther discussion, see Lantolf & Poehner, 2004)
Concurrent G-DA with English Language Learners in an
Australian Primary School Setting
Gibbons (2003) describes a primary school class of English language
learners who are attempting to gain control over appropriate scientifi c
terminology to discuss the outcomes of their laboratory experiences
3 The interested reader may also wish to consult the ongoing work of Hamp-Lyons, Davison,
and colleagues under the rubric interactive assessment , which shares DA’s commitment to
providing support during procedures including those in group contexts
Trang 9In one particular lesson, the class worked in groups to conduct an ment on magnetism and then convened as a class to relate their observa-tions and inferences As learners describe their experiments, the teacher offers mediating support that increases in explicitness with each move, as
experi-in DA What is important is that the teacher does not run through the full range of mediating prompts with a single learner before beginning again with another individual Instead, the teacher’s focus remains fi xed on the entire class, and although he may call on a particular student to answer a question or may address an individual learner at a given point in time, his next remark will be directed to another learner and will build on the pre-ceding contribution In this way, the teacher engages the class in the activ-ity of using scientifi c terms to talk about magnetism and offers mediation intended to benefi t all As mentioned, the primary and secondary inter-actants during such an exchange are in a constant state of fl ux
To appreciate how this type of interaction unfolds, consider the two following brief exchanges reported by Gibbons (2003, p 264) and dis-cussed by Lantolf and Poehner (2004, pp 64–65) These exchanges occurred consecutively during the same lesson In the fi rst, the teacher (T) elicits a response from Beatrice (B) and in the second from Michelle (M) Note how the progression from implicit to explicit mediation from the teacher continues across the two exchanges rather than within them separately:
Exchange 1
1 T: Tell us what happened
2 B: Em we put three magnets together/it still wouldn’t hold the
3 gold nail
4 T: Can you explain that again?
5 B: We/we tried to put three magnets together to hold the gold
6 nail even though we had three magnets it wouldn’t stick
At this point, the teacher turns to another learner nearby:
Exchange 2
1 T: Tell us what you found out
2 M: We found out that the south and the south don’t like to stick
3 together
4 T: Now let’s/let’s start using our scientific language Michelle
5 M: The north and the south repelled each other and the south and
6 the south also repelled each other but when we put the/when we put the
7 two magnets in a different way they/they attracted each other
As Lantolf and Poehner point out, the prompt given to Beatrice to attempt the explanation again is far less explicit than the one give to
Trang 10Michelle, which directs her to use scientifi c language In fact, the
teach-er’s prompt to Beatrice does not even state that there is a problem with
her description whereas the prompt to Michelle reveals the nature of the
problem Viewed as two distinct interactions, one might conclude that
Beatrice was disadvantaged and the teacher was more favorable to Michelle
by taking a more leading role in guiding her development However,
shift-ing the analytical focus from the individual to the group allows a very
dif-ferent picture to emerge From this perspective, the prompts to Beatrice
and Michelle are alternating turns in a dialogue that the teacher
nego-tiates with the entire class, a dialogue in which mediation is continually
fi ne tuned to remain in step with the learners’ level of responsiveness
Of course, this dialogue might be interpreted simply as an example of
good teaching; indeed, Gibbons does not refer to DA but understands
the exchanges as teaching-focused interactions that take account of the
ZPD Recall, however, that Vygotsky’s formulation of the ZPD posits a
dia-lectical relation between teaching and assessment Offering learners
mediation, as the teacher does in these examples, serves simultaneous
teaching and assessment functions, a diagnosis of abilities that are still in
the process of forming as well as an intervention to support their
devel-opment Given that these interactions occur within a group, it is also
worth noting that Michelle’s success in arriving at an appropriate response
does not necessarily indicate that she is at a higher level of development
than Beatrice Although she may be, it is also possible that she is more
responsive because, unlike Beatrice, who only received one mediating
prompt, Michelle stood to benefi t from two—one directed to her and
one directed to Beatrice The question, which we are unable to answer
defi nitively, is what impact the prompt directed to Beatrice had on
Michelle’s performance The matter of secondary interactants benefi
t-ting from mediation offered to a primary interactant is equally important
for cumulative G-DA and is taken up in the next section
Cumulative G-DA with Spanish Language Learners in a
U.S Primary School Setting
Lantolf and Poehner (in press) report the results of a study of DA
prin-ciples implemented in the context of a laboratory primary school affi
li-ated with a major urban university in the northeastern United States 4
The school employs a full-time L2 Spanish teacher, Tracy (a pseudonym),
4 This research was funded by a grant from the United States Department of Education
(CFDA 84.229, P229A020010-03) However, neither the arguments presented here nor the
contents of Lantolf and Poehner’s (in press) paper necessarily represent the policy of the
Department of Education, and one should not assume endorsement by the United States
government