The effect of teacher feedback on EFL learners´
functional production in classroom discourse
Ana Llinares García, Universidad Autónoma de Madrid (Spain)
Index
1
Introduction
3 The study of teacher feedback
3.1 Classification of teacher feedback
in the present study
4 Taxonomy of children’s
communicative functions
5 Analysis of the data
5.1 Frequency of functions
5.2 The effect of teacher feedback
5.3 Summary of results
Bibliography
The
present study is based on the analysis of classroom interactions between
children and with their teacher both in first and second language contexts. The
subjects of our study are two five-year-old classes in an English bilingual
school in
The source
of our analysis is Halliday’s classification of the communicative functions
that children can convey in their mother tongue at the pre-school level
(Halliday, 1975). Our premise is that the functional variety of children’s
production and their use of language for their own communicative purposes is
highly related to both frequency and type of teacher feedback. This is especially
important in the case of second language learners, in order to avoid their L2
language production being limited to responses to their teacher’s questions.
1 Introduction
Halliday’s
(1975) functional analysis of child language shows the development of
communicative functions in the language of a child from nine to eighteen
months. He analyses the way language is used to serve certain communicative
needs that children have in their interaction with the people that surround
them. Against the common belief in the 1960’s that language is an autonomous
being, language development studies in the 1970’s move away from this
direction. Language is no longer seen as a "kind of spontaneous once-for-all
happening resting on a given biological foundation, to be achieved by a certain
maturational stage or not at all" (Halliday, 1975: 139).
Halliday
(1975) advocates that language is mainly a social phenomenon. The child
constructs a system of meaning that represents his/her own reality. It is a
cognitive process but it always takes place in contact with the others.
According to Halliday (1975), the social context is the generator of the
meanings that are learnt. The social context is based on the contextualised
language that children hear around them and, therefore, it seems necessary to
focus on the functions that the child employs to make meanings, rather than the
structures used to make those meanings manifest.
More
recently, Painter (1999) also analyses the development of the ideational
function in the language of a child from two to five, from a systemic
functional perspective. In her work, she states the importance of the analysis
of the child’s interpretation of the world in the interaction with the others.
In western
cultures one of the most significant social contexts that represent the reality
of the child is the classroom. The language that children use in the class, as well
as the language that they hear, is expected to be related to the context that
surrounds them, and thus, it will convey the functions needed for that
particular context. In my view, the functional analysis of classroom discourse
(Christie, 2002) needs to be supplemented with the analysis of language
development in foreign language contexts and, in order to have significant
results, it is necessary to work with learner corpora (Granger, 1998; Llinares
García, 2004).
2 Methodology of the study
This paper
analyses the different types of feedback used by teachers with their pupils,
comparing a native and a foreign language context, in order to see to what
extent the quantity and quality of the feedback enhances a functionally richer
learner production in the L2.
The data
for the present study comes from the UAMLESC[1] (UAM Learner English Spoken Corpus), which is collecting
spoken data from EFL learners at different educational levels with different
degrees of immersion in the L2. The English native data comes from the CHILDES
corpus. The total number of sessions analysed was 15 (five sessions in each of
the three groups: the native group and the two EFL groups).
This study follows the ones carried out by
Romero and Llinares (2001) and Llinares (2004; in press) based on a corpus
driven analysis of the communicative functions realised by pre-school children
in the classroom settings mentioned above.
3 The study
of teacher feedback
The study
of verbal interaction in the classroom dates back to Sinclair and Coulthard
(1975), who describe the structure of classroom discourse. Other interesting
studies are Stubbs’ (1976) observation of the control that teachers have of the
relevance and validity of children’s discourse, and the study carried out by
Barnes (1989), who analyses one day in a secondary school class in a British
school, focusing on the effect of the abstract language used by the teacher on
the pupils’ learning process. As far as second language contexts are concerned,
Richards and Lockhart (1994) focus on the importance of not only letting
learners know how well they have performed but also of increasing their
motivation.
There are
different types of classifications of teacher feedback. Sinclair and Coulthard
(1975) focus on the “evaluative feedback” used by the teacher in classroom
discourse, which usually consists of the acts of accepting, evaluating and
commenting. Richards and Lockhart’s (1994) classification includes saying that
something is correct or incorrect, praising, modifying a student´s answer,
repeating, summarising and criticising.
3.1 Classification of teacher feedback in the present study
Based on
the analysis of the data, we have distinguished two main types of feedback: a pedagogic feedback, of the type
identified by Sinclair and Coulthard (1975), and an interactional feedback, which reflects the type of feedback used in
ordinary conversations outside the classroom. The different types are defined
below and are illustrated with examples from the corpus:
Interactional Feedback
(IF): comment made by the teacher, with no evaluative or
corrective purpose, which may enhance the learner’s linguistic production. This
type of feedback includes expressions of agreement, disagreement or
acknowledgement[2].
(1) CH: because the
battery’s finished
TCH: Oh, you need to get new batteries (IF)
CH: I can take
this one
Pedagogic Feedback
(PF): acknowledgment or comment made
by the teacher, with the purpose of correcting or evaluating the children’s
performance. Four main types of pedagogic feedback (PF) have been identified:
·
PF that evaluates
the learners’ production positively (PF1)
(2) CH: Mum put the... wallpaper on the wall
TCH:
Good (PF1)
·
PF that evaluates
the learners’ production negatively (PF2)
(3) TCH: No,
scardy isn’t a word (PF2)
·
PF that corrects
the learners’ production (PF3)
(4) TCH: No, scardy isn’t a word. You must say scared (PF3)
·
PF that gives the
learners a clue for the right answer (PF4)
(5) TCH: Yes, where does the bread come from:
plants or animals?
CH: I don’t know
TCH: You remember what it’s made from? It’s made from wheat (PF4)
·
PF that prompts
the learners to respond (PF5)
(6) TCH: Can you tell me the right order?
CH: I don’t know
TCH: Come
on! (PF5)
As Chaudron
(1988) points out, the teacher’s presentation of the correct answer after
correct and incorrect responses can lead to ambiguety: "... repetition of
a speaker´s utterance can serve several functions, of either a negative
(correcting) or a positive nature (agreeing, appreciating, understanding
…)" (Chaudron 1988:145). Thus, the learner may see the modification as a
mere alternative to his/her own utterance, because accepting, approving,
confirming repetitions occur frequently in the same contexts. In the taxonomy
presented here, we have distinguished between the feedback used by the teacher
that includes changes in the learner’s utterance (which has been classified as
pedagogic corrective feedback, PF3) and the feedback which does not change the
child’s utterance or, if it does, shows interest in the content and not the
form (this type has been classified as interactional feedback).
4 Taxonomy
of children’s communicative functions
One of the aims of this study is to find out whether the use
of different types of feedback by the teachers may lead to a richer functional
production in the learners’ L2. In order to understand what is meant by “rich
functional production”, it is necessary to summarise the main functional
categories proposed by Halliday (1975) and later adapted by Llinares García
(2004) to the EFL context. Each category is illustrated with examples of
learners’ production in the EFL context.
Instrumental (I)
It is the
function that language serves of satisfying the child’s needs. Llinares García
(2004) identified the functions of expressing wishes and asking for permission
as the main subcategories within this group, as shown in examples (7) and (8)
from the corpus:
(7) CH1: Give
me a pencil (I)
CH2: No
(8) CH1: I
can do it? (I)
CH2: No
Regulatory (R)
It is the
function used to control the behaviour of others.
(9) CH:Turn
it up (R)
Informative (Inf)
It is the
function of language to give information about people, things and actions.
Llinares García (2004) included in this category functions such as describing,
identifying and narrating.
(10) CH:
No, it doesn´t work (Inf)
Personal (P)
It is the
function used to express the uniqueness of the speaker and his/her perception
of things and events.
(11) CH:
((TALKING ABOUT DEATH)) I think we would not.. grow again (P1)
Interactional (In)
It is the
language used by the child to interact with those around him/her. This function
includes greetings, apologies, etc...
(12) CH: Oh, sorry I didn´t see it (In2)
Heuristic (H)
It is the
use of language to request information.
(13) CH: Is
it red?(H1)
TCH: No.
In our
opinion, the realisation of these functions by EFL learners would make their L2
production “functionally richer”, as they would be using the target language in
a similar way to the L1.
5 Analysis of the data
5.1 Frequency of functions
The
children’s production in the fifteen sessions was codified according to the
main functions presented above. The first step was to quantify the number of occurrences
of each of the functions per group, with a further differentiation between
teachers and pupils. This differentiation was made on the assumption that the
functions used by the children would heavily rely on the use made by the
teachers. We used the data from the native group (from the CHILDES corpus) as
an example of the language used by five-year-old children in classroom
contexts. It served as a contrasting group as we were interested in finding out
to what extent the two EFL groups were close or distant to the native group in
the type of functions used.
Figures
1a, 1b and 1c show the distribution of functions in the first language context
and in groups A and B in the EFL context:[3]
Figure 1a. Frequency of communicative functions in the EFL
context: Group A
Figure 1b. Frequency of communicative functions in the EFL
context: Group B
Figure 1c. Frequency of communicative functions in the First
Language context
Figure 1c
above shows that, in the first language context, the teacher uses most
functions with a higher frequency than the children, except in the case of the
personal function. The most common functions in the language of the teacher are
regulatory, feedback and heuristic, which correspond to the most common
realisations identified in teacher talk (Chaudron, 1988).
In group A
in the EFL context (see figure 1a above), we can also observe the leading use
of the regulatory, feedback and heuristic functions by the teacher. Again, we
can see that the personal function is used more often by the children than by
the teacher, and it is the most frequent function in the language of the
children.
Group B shows a similar pattern in the type of functions used by
the teacher. Also, as in the previous cases, the children’s most frequent
function is the personal one. However, it is important to stress that the main
difference with group A is that 55.2% of the realisations of the personal
function in this group are in Spanish. In fact, 53.1% of all the functions realised by
the children in this group are in the mother tongue.
These results confirm the importance of the
personal function in child language at this age. The results in the first
language context show that children need
to perform this function, in other words, they like talking about their
personal things. This also happens in the EFL context, but here the two groups
differ significantly: the children in group A use the L2, whereas in group B
they often use the L1 (Figure 2):
Fig. 2. Language used by Spanish
children in the EFL context to realise the personal function
In the
next section we will argue that the quantity and quality of the teacher feedback
enhances the realisation of the personal function in the L2 in group A, whereas
in group B the lack of appropriate feedback by the teacher leads to a more
frequent use of the L1.
5.2 The effect of teacher feedback
As figure
3 below shows, the quantitative analysis of the two main types of feedback
proposed in this paper indicates the most frequent use of pedagogic feedback (PF) in all groups. This
result is not surprising as classroom feedback tends to be used for pedagogic
purposes (evaluating, correcting, etc...). However, in the first language group
(FL) this frequency is not significantly higher than the use of interactional
feedback (IF). As far as the EFL context is concerned, there is again an
interesting difference between both groups. In group A, where the teacher
interacted more with the children[[4]], the teacher’s performance of interactional feedback is
much higher:
Figure 3:Distribution of the Pedagogic (PF) and Interactional
(IF) feedback in the three groups
In the EFL context, the frequency of the
function of feedback in both groups is very similar and even higher in group B
(21.3% in group A and 25.6% in group B of the total number of functions
realised). However, the type of feedback used is very different. In group A
38.5% of the feedback provided is of an interactional type, whereas in group B
it only represents 9%.
Another common feature in all groups is that the most frequent
type of pedagogic feedback is positive evaluation (PF1). This result contradicts
Nunan and Lamb’s (1996) conclusion that error correction is one of the first
tasks of the language teacher. According to our data, this does not seem to be the case in pre-school
contexts, where positive evaluation is more frequent than correction.
As far as
interactional feedback is concerned, as we have already indicated, it was more
frequent in the language of the teacher in the native group than in the
teachers in the EFL groups. In the case of Group A, where it represents a
higher percentage than in group B, almost 70% of the realisations of this type
of feedback generated a personal function from the children. Moreover, in group
B 18 out of the 22 realisations of this feedback by the teacher (more than 80%)
generated the children’s realisation of the personal function in the L2.
5.3 Summary of results
The main results of the analysis can be
summarised as follows:
·
The personal
function of language is the most common function in the language of five-year old
children in the classroom, both in the L1 and the L2.
·
Pre-school first
language contexts show a similar frequency of use of pedagogic and interactional
feedback. However, in EFL contexts the first type is more frequent.
·
In both first
language and foreign language pre-school contexts, “positive evaluation” is the
most common type of pedagogic feedback,
contrary to previous research that point to “corrective feedback” as the most
frequent.
·
The teacher’s use
of interactional feedback in the EFL context seems to encourage the learners’
realisations of the personal function in the L2.
From this
study, we can conclude that EFL young learners can realise the same functions
in the L2 as native speakers of the same age, if their teacher encourages them
to do it. The frequent use of the personal function by the children in the
native context indicates its importance as a communicative tool at this age,
also in classroom contexts. Therefore, it seems to be a function that should be
enhanced in EFL contexts.
The
present study shows that interactional
feedback, which is frequently used in first language contexts, is also
necessary in EFL classrooms if children are expected to use the L2 to convey
functions such as the personal one. This has been observed in our analysis of
Group B in which the teacher’s low frequency of use of this type of feedback
has resulted in the children’s use of their mother tongue to compensate for
their lack of functional competence in English.
Finally,
it is essential, in our opinion, that we establish clear comparative
correlational models with the native language classroom context that could help
the non-native teacher of English to establish the parameters of use of the
different discourse functions. This type of studies are especially aimed at the
promotion of children’s language production and communication in a foreign
language from the early stages of their learning process. This is a factor
which will certainly facilitate their oral communication skills in a foreign
language as they grow older.
Bibliography
Barnes,
Chaudron, Craig (1988): Second Language Classrooms: Research on
Teaching and Learning.
Granger, Sylviane (ed.) (1998): Learner English on Computer.
Halliday, Michael (1975): Learning how to Mean. London: Edward Arnold.
Llinares García, Ana (2004):
“La interacción lingüística en el aula de segundas lenguas en edades tempranas:
análisis de un corpus desde una perspectiva funcional”. Premios Nacionales de Investigación Educativa 2002. Madrid: Ministerio de Educación y Ciencia.
Llinares García, Ana
(in press): “Young learners’ functional use of the L2 in a low-immersion
context”. ELT Journal.
Nunan, David and Clarice Lamb (1996): The self-directed teacher.
Painter, Claire (1999): Learning through Language in Early Childhood.
Richards, Jack and Charles Lockhart (1994): Reflective Teaching in Second Language Classrooms.
Romero Trillo, Jesús and Ana
Llinares García (2001): “Communicative constraints in native/non native
pre-school settings”. International
Journal of Corpus Linguistics, vol. 6/1, pgs. 27-46.
Sinclair, John and Malcolm Coulthard
(1975): Towards an Analysis of Discourse.
Stubbs, Michael (1976): Language, Schools and Classrooms.
Corpora:
The
CHILDES Corpus
The
UAM-Learner English Spoken Corpus
____________________________________________________________
copyright @ ANGLOGERMANICA ONLINE 2005
____________________________________________________________