A Journal of Teaching English Language and Literature

ISSN Print : 2229-6557, Online: 2394-9244

Capturing Individual Growth in Group Discussions Through Teacher Observations

Pankaj Narke


Human interaction requires a great deal of collaboration and one of the ways in which this can be taught in ESL contexts is through group discussions (GDs). GDs are a part of many proficiency courses and ongoing assessments, and they have specific criteria for evaluating the performance of the participants, but these are applied in a uniform manner. Therefore, such criteria are not able to capture the varied growth of individuals on parameters such complexity in speech, accuracy, participation, time management and use of different communicative strategies.  In this paper, I will attempt to capture the individual growth of learners in the use of communicative strategies with a focus on language. This will be done through in-class teacher observations, accompanied by retrospective teacher notes. For my paper, I observed the performance of three tertiary level ESL learners across six rounds of GDs. The results showed that these three participants grew consistently, yet differently.

Keywords: evaluation, formative assessment, ESL, collaboration.


Group discussion is one of the most effective tasks in the communicative approach as it is highly meaning-focused and requires learners to interact and negotiate meaning in order to reach the desired goal. Among the many prominent reasons for using GDs in ESL classrooms, one is that it gives enough space for learner-learner interaction unlike one-way speaking tasks. However, the literature available on this issue views both sides of the coin. One view is that student-student interaction can only lead to the exchange and eventual fossilization of errors (James, 1994). Therefore, it is commonly believed by the majority of language teachers that real learning happens only through teacher-learner interaction. Peer group work is at best considered as a social exercise and good only for project work. Its implications for language learning and teaching have therefore not been explored enough. In contrast with this view, a majority of research in recent years has attempted to find out the effectiveness of task types in second language acquisition. It was observed that in group tasks, a lot of meaning negotiation takes place as it demands high levels of learner-learner interaction, which in turn leads to language acquisition (Courtney,1995).

However, considering the nature of a group discussion, the output produced by the learners is bound to be extremely complex compared to a one-way or a dialogue task. Also, since 5-6 participants perform a GD task simultaneously, it becomes difficult for the teacher to set common criteria to assess their individual language development. Another difficulty is in eliciting the expected output from the learners in a discussion task even if they are exposed to the targeted forms of language (Ellis, 2005). One of the major reasons for this is that learners’ response in group discussions cannot be predicted and controlled, unlike in a solitary or dialogic task. Therefore, it will not be feasible to assess learners’ performances in discussion tasks on pre-decided language forms or functions. This suggests that the traditional model of summative assessment will not be of great help in tracking learners’ growth in group discussions, as there may not be any definite predictable areas of growth.

Formative assessment, in contrast, helps teachers to track learners’ ongoing growth through observation. It provides information to the teachers or learners, about strategies used, which can be used to modify teaching and improve learning (Cizek, 2010). Researchers have ranked formative assessment as the best among all the methods of language assessment for its effectiveness not only in measuring learners’ growth but also for contributing to learning. William (2011), tried to establish the connection between classroom instruction and assessment to foreground the importance of formative assessment. In one of the studies that he reviewed, the teachers studied recorded videos of learners’ performance in an oral problem-solving discussion task and identified the strategies that the learners used for problem solving. Teachers accordingly modified their instructions on the basis of this evidence, resulting in significant improvement in the learners’ problem solving skills, which is a crucial aspect of GDs.

Another factor that affects student performance in GDs is the use of appropriate communication strategies (CS). Faerch and Kasper (1983) define CS as potentially conscious plans for solving what to an individual presents itself as a problem in reaching a particular communicative goal. A group of researchers conducted studies in which they compared the CS of L2 learners with that of native speakers and found differences in their proficiency levels. Therefore, they advocated the teaching of CS in the L2 context, in order to gain native-like proficiency. Another group of researchers who compared L2 learners’ performance in the target language with their own L1 performance, found many similarities (Kellerman, 1991). They believed that strategy transfer is an obvious process and it will happen without much conscious effort. Therefore, this group of researchers did not advocate the teaching of CS in the classroom, rather they believed in teaching the language itself. However, in the Indian context, including CS in classroom instruction is required not to achieve native-like proficiency but to help learners become better communicators; and for this purpose, GDs are ideal tasks.

Though there have been ongoing debates around the use of CS in L2, not much has been discussed about its role in evaluating learners’ individual performance. This evaluation can be achieved through teacher observations and retrospective notes for self-reference.

The Study

In the present study, GD tasks were used as vehicles for speech production. Out of the 24 tertiary level engineering students,  who took part in the study, the speech samples of three students have been analyzed. The data in this article comprises an assessment of their performance in three alternate rounds of GDs out of a total of six rounds to try and find markers of growth, for it was felt that it would be difficult to find actual growth in consecutive rounds.  However, the qualitative analysis of the data, to identify learners’ gradual growth in language, was done across all six rounds. Since investigating learners’ varied individual growth was the major objective of the study, no inter-student comparison has been attempted.  Through the study, I attempted to answer the following question:

What are the different kinds of growths that can be identified in ESL learners when they participate in group discussions? How can these growths  be recorded by the teacher?


Nature of participants and tasks

The participants in this study comprised three third year engineering (CSE) students of the twenty-four in the original study which is a part of my ongoing doctoral study. All three participants had communicative English as one of the courses in their syllabus, which included communicative language functions such as group discussions, report writing, interview skills and presentation skills. The participants had to take part in online synchronous discussion (OSD) at least one day before they took part in the face-to-face group discussions. Six rounds of OSD and face-to-face group discussions were conducted over a period of two months. The OSDs were conducted using Facebook chat outside the classroom (at a time and place convenient to the learners). In these OSDs, the participants had to discuss a topic which was part of a broader topic that was debated in the face-to-face discussion. OSDs were meant to make the GD task familiar to the participants, as well as to initiate their thinking on the schema (content and language) related to the topic. There were 3-4 groups comprising 5-6 participants in every round of discussion. While forming groups, I, as the teacher-researcher, made sure that every group was a mixed ability group (based on the results of a proficiency test conducted at the beginning of the study) and for each GD, the group members were shuffled. This was done in order to avoid repetition of content. Though the researcher was not a part of these discussions, he was a close observer and made notes to mark instances that showed an improvement or deterioration in the learners’ performance. Every face-to-face discussion was video recorded and posted on a common Facebook page (created especially for this study); all participants had access to this page. Participants were encouraged to reflect on their performance and to give peer feedback which was posted along with the researcher’s written and oral feedback. Since this feedback is not a part of this study, that data is not provided.

The data gathered from these face-to-face discussions was closely examined. The remarks of the observer with regard to the performance of the three participants during the three rounds (first, third and fifth) were also analyzed. Finally, these observations were corroborated with the researcher’s  notes and video recordings.

Data Analysis

First participant (P1)

In the first round, the researcher observed that the first participant (P1) faced problems while interacting, which included lack of appropriate words, uncertainty with grammatical forms, insufficient content, and hesitation in taking turns. In order to cope with these issues, the participant used certain CS, of which three were consistent across all the three GD rounds.  These strategies were: use of fillers, repetition and self-correction. Though these three strategies were used by the participant in all three rounds, they were modified in each round and their communicative function increased remarkably.

The fillers used by P1 in the first round seem to have helped him in recalling the content or vocabulary and in maintaining his turn. However, they did not fulfill any interactional function as they were just sounds such as “aaa”. Therefore, in the context of the group discussion, they could be disregarded as meaningless language units.

“so as Kaleem said, arranged marriages are good, it is a prediction [sic]. Aaa… if aa couple is aaa... doing [sic] a love marriage then aaa... the parents may be anti… so they have to start a new life, there is no back up to…”

In the third round of the GD, P1 made use of different fillers to cope with fluency issues. The fillers used in this round had a communicative function unlike in the previous rounds.

Yeah but physically they are not as strong as men…”

Yeah, I would like to make a comment here…”

P1 used these fillers at the beginning of his comments so he could get enough time to structure his sentences. Also, the fillers acknowledged the earlier participant’s view by showing acceptance.

The fillers used in the last round by P1 fulfilled an even more complex communicative function for not only did they acknowledge the earlier participants’ comments but they were also used to appeal to the other participants to consider his perspective.

See if we go deep into this topic, this reservation might go…it mean [sic] deviate...it shows unequality [sic] some times. See if we give reservations to the particular people…”

This change in the use of fillers shows a clear growth in the participants’ use of CS, from simple to complex. The first filler (“aaa”) is merely a sound, whereas the latter two (“yeah” and “see”) are meaningful words and they perform a complex communicative function during the GD.

Another commonly used CS by P1 across all GDs was repetition of words, phrases and sentences in order to gain time to organize language or content related knowledge. The participant exhibited growth in this strategy as well. His improvement could be traced from merely repeating words/phrases to restructuring sentences. In the first round, P1 simply repeated certain phrases, but he could not complete the sentence and lost his turn:

So there will be, there will be aaa a backup”

By the third round, he progressed from simply repeating phrases, to repeating them to restructure the sentence and make it more meaningful in the context of the GD:

First we have to change theI think we have to change the mindset of the men.

“Yeah, it’s a very good point that aaa women… I agree with Shanthi. So it’s a very good point that many people are killing before the woman has [sic] born.

Similarly, in the last round P1 did not just repeat the phrases but also restructured his sentences and made them communicatively more significant:

“Government should mainlythe main purpose of this reservation is to help the people financially.…”

The progression in the use of repetition, from merely repeating phrases to repeating and restructuring arguments shows the development in the participant’s use of CS.

The third most commonly used CS by P1 in all rounds of GD was self-correction (the other two being fillers and repetition). P1 used self-correction when he realized his mistakes in grammar or vocabulary, which shows that he was aware of his mistakes. In the first round, self-correction was restricted to a syntactic level:

“they have to start a new life, there is no back up to… backup for them, no supports [sic] for them…”

In the next round, it was at a semantic level:

“So I think reservations can be provided in some restricted areas, can’t be provided in restricted areas.

These observations reflected the growing use of CS by the first participant. P1 used a few other strategies as well (such as, paraphrasing, word appropriation, use of fixed expression, etc.), but as these were not consistent across the GDs, they were not considered as good representative samples.

Second participant (P2)

Group discussions being the most uncommon and least practiced task, all participants had communication related issues. The two most significant communication related issues faced by the second participant (P2) were lack of content/vocabulary and uncertainty with regard to sentence structure.  To cope with these problems, P2 used different CS, out of which restructuring of sentences and self-correction were observed consistently across all rounds of GD. Both these strategies helped the participant in dealing with the above communication related problems.

In the first round of the GD, the participant used restructuring, which was very close to repetition. Through the restructuring strategy, P2 did not change the argument but presented it in a more organized manner:

So in that, what is my opinion is, in that time period the [sic] we can betterly [sic] understand our partner in that time.”

However, in the fifth round, P2 used the same strategy for a different communicative function, which may be graded higher on the scale of communicative complexity. P2 restructured the agreement in order to not to show absolute disagreement with another participant and to maintain the flow of the GD:

“You are saying that reservation is good. But the people... yeah I will also agree for [sic] that. But reservation …”

The two different instances of using restructuring as a CS is indicative of P2’s progress—being able to modify the statement according to the context and also the use for a simple to more complex function, namely, from restructuring as repetition to using restructuring to maintain the flow of the argument.  Use of self-correction as a CS is also evidence of P2’s progress in GD performances across the three rounds. In fact, his progress can be traced from the complete absence of self-correction in the first round to its advanced use in the last round. In the first round, P2 seemed to be unaware of his mistakes. Therefore, even though he had the opportunity to correct an adverb form, he did not do so. For example:

“…we can betterly [sic] understand our partner...”

In contrast, in the last round, P2 corrected himself to avoid wrong usage of a word.  He was aware of the inappropriate use of a word category and he corrected himself immediately.

Reservations are like politicals...politicians...they want votes…”

He can’t get a goodbetter college compare to…”          

These examples of self-correction illustrate the growth in the participant’s communicative ability in GD tasks. Along with restructuring and self-correction, P2 used a few other CS, such as hedging markers and coining words. 

Third participant (P3)

The third participant (P3) also exhibited excellent use of CS to tackle communication related problems. He consistently used word appropriation, paraphrasing, and restructuring across all rounds of GDs.

In the initial round of GD, P3 was not aware about the usage of context specific words. Therefore, no instances of word appropriation were observed in his performance. However, in the later rounds, P3 attempted to use low frequency words. To avoid ambiguity of meaning as the participant was using these words for the first time, he used the strategy of word appropriation.  For example:

“we give most of our property…what…our income to the govt...”

P3 also had a tendency to coin new words/phrases that were grammatically inaccurate. To overcome this problem, he used the strategy of paraphrasing in the later rounds of the GD. Though this strategy was missing in the first round, to avoid the risk of inaccuracy he chose to adopt it the later rounds:

“people of lower caste which [sic] are thought to be less money, the people who have less money.”

 P3 used the strategy of restructuring sentences to avoid syntactic accuracy and semantic ambiguity.  Like the earlier CS of paraphrasing, this strategy was also used at a later stage of the study, when the participant was more familiar with the task and also more competent.  For example:

And the reservation when the… actually I feel is, when the food is given to all the people…”


The data analysis reveals that there was an observable growth in the participants across the rounds of GD and with a conscious effort on the part of the teacher, this growth can be recorded and tracked. For this, teachers, can maintain an observation diary/observation portfolio for individual learners. They can mark the progress of the participants in the use of CS from time to time, and provide instructions to them for more advanced CS on the basis of their observations.

As observed in the analysis section, the participants did not use CS in a uniform way; rather they used them in varied ways. The difference in the choice of CS of the three participants was discernible. P1 made use of fillers, restructuring and self-correction. Analysis reveals the existence of all these strategies in the participant’s performance right from the first round. However, the participant made advanced use of these strategies as the GDs progressed. For example, the strategy of self-correction was used at a syntactic level in the first round, whereas the same strategy was used at a semantic level in the last round.

P2 used different CS—self-correction and restructuring—which were consistent in his performance across all the rounds of GD. Restructuring, which was very close to repetition in the first round was later used to indicate a change of opinion. Though this strategy was used by P1 as well, the cause and effect of use was different.

Completely different from P1 and P2, P3 made use of a new strategy—word appropriation—which helped him avoiding unintended meanings. This strategy was a completely new addition to the set of strategies used by the other two participants. However, P3 did not make use of self-correction unlike the other two participants. It is evident from these observations, that different learners grow differently irrespective of the same environment. 


This study showed that though there are no prescribed parameters to track individual learners’ growth in GD tasks, this can be achieved through careful observations in the class and retrospective teacher notes. Such a study, though carried out with a small number of participants, has implications for language teaching. The teacher can observe the pattern of CS used by the learners and use the data to tackle communication related problems. He/she can accordingly modify classroom instructions and train learners to use CS at an advanced level. However, certain modifications will have to be made to the kinds of observations possible, if such a method had to be applied to large classes.


Courtney, M. (1995). Task-based language teaching for academic purposes. In K. Wong & C. Thinking Language: Issues in the study of language and language curriculum renewal, 83-95.

Durairajan, G. (2016). Evaluating and teaching through observing and recording student growth. EFL Journal, 7(1), 93-107.

Ellis, R. (2005). Planning and task performance in a second language. Auckland: University of Auckland.

Færch, C. & Kasper, G. (1983). Strategies in interlanguage communication. Harlow, UK: Longman.

Faucette, P. (2001). A pedagogical perspective on communication strategies: Benefits of training and an analysis of English language teaching materials. Second Language Studies, 19(2), 1-40.

James, G. (1994). Oral testing at tertiary level, In Boyle J. & Falvey P. (Eds.), English Language Testing in Hong-Kong. (pp. 183-196). Hong Kong: Chinese University Press.

Kellerman, E. (1991). Compensatory strategies in second language research: A critique, a revision, and some (non-)implications for the classroom. In R. Phillipson, E. Kellerman, L. Selinker, M. Sharwood Smith, & M. Swain (Eds.), Foreign/second language pedagogy research: A commemorative volume for Claus Færch (pp. 142-161). Clevedon, UK: Multilingual Matters.

William, D. (2011). Embedded formative assessment. Bloomington, IN: Solution Tree Press.

Pankaj Narke is a research scholar at the English and Foreign Languages University in the School of English Language Education in Hyderabad.