Investigating the effects of task repetition on fluency and accuracy in English oral performance of low level adult students: A case study at Vietnam air defence and air force academy

v LANGUAGE TEACHING METHODOLOGY

INVESTIGATING THE EFFECTS OF TASK
REPETITION ON FLUENCY AND ACCURACY
IN ENGLISH ORAL PERFORMANCE OF LOW LEVEL
ADULT STUDENTS: A CASE STUDY
AT VIETNAM AIR DEFENCE AND AIR FORCE ACADEMY
LUONG THI PHUONG*
*
Vietnam Air Defence and Air Force Academy, 
Received:23/12/2018; Revised: 31/01/2019; Accepted: 28/4/2019

ABSTRACT
The major aims of this case study were to investigate the effects of task repetition on three adult
students’ accuracy and fluency in speaking English. Also, it sought to examine changes to the level
of anxiety when speaking, which is a secondary aim. Qualitative research method was applied,
with assistance of some quantitative analysis. The findings revealed that the participants generally
showed improvements in their accuracy, while fluency took slightly different ends. As regards the
level of anxiety, some minor decrease was observed as a result of repeating the task. Besides, all
the participants made modification of content in their second presentation, which opens a new
research domain for my future study.
Keywords: accuracy, fluency, task repetition, teaching English speaking, Vietnam Air Defence
and Air Force Academy

1. INTRODUCTION
This paper reports a case study done at Vietnam
Air Defence and Air Force Academy (hereafter
referred to as VADAFA), aiming at examining the
effects of task repetition on learners’ accuracy and
fluency when speaking about a past event. Besides,
it was expected that students would feel less

anxious when speaking English in front of others
as a result of being familiar with the task. The
students at VADAFA follow a ten-month intensive
English course which is specially designed for
the military officers who have been experienced
in their profession for at least three years. Though
the researcher is not the main instructor of this
class, so far she has taught them for a few weeks

26

KHOA HỌC NGOẠI NGỮ QUÂN SỰ

No. 19 (5/2019)

as a supplementary teacher. The course uses the
English File series as course books, and includes
English speaking as one of the major macro skills
to be taught and tested on. The target students are
mainly the ones who wish to develop further in
their career by pursuing higher education either
abroad or in-country. Even if some of them have no
intention to study further, being fluent in English
speaking may be a great advantage when they have
opportunities to join international conferences and
workshops.
During the period of ten months, the course
runs for five successive classes a week, from
Monday to Friday, consisting of five forty-fiveminute sessions per day. This syllabus in fact

LANGUAGE TEACHING METHODOLOGY

allows the students to practise a substantial
amount of speaking activities both in and after the
class. So far, a commonplace of a speaking lesson
has been the fact that the teacher provides one or
two short speaking activities, whether guided or
free, then while monitoring, s/he often gives onthe-spot feedback, either implicitly or explicitly,
on some basic mistakes such as the pronunciation
of sounds, verb tenses, or subject-verb agreement.
At the end of the course, the students will have
already practised a variety of topics ranging
from elementary to intermediate level, plus the
same amount of situations in which they need
to use functional English. Most of these topics
and situations will reoccur in the end-of-course
examination. After testing three different courses,
however, the researcher found that a majority
of students performed the oral task with limited
accuracy and fluency. Moreover, many of them
showed a high degree of anxiety or nervousness
when speaking in front of the others. As soon as
they start speaking English, for instance, some
of them displayed these feelings through their
trembling voice or hands, while some others
showed evidence of tension on their face. When
it came to communication with foreign friends,
these shortcomings became a real hinderance to
their mutual understanding.

While reflecting back on the teaching of
speaking skill that the researcher and other English
instructors at VADAFA have been applying, it
seems that although this intense syllabus may
result in the students “knowing” a huge amount
of speaking tasks, yet their being able to use the
appropriate language to “perform” the tasks is
open to question. In other words, our delivery of
speaking activities are somehow superficial. Lynch
(2001) has a related concern and he explains this
reality by stating that nowadays teachers try to
maximize learner activity and output at the expense
of allowing too little time for reflection (p.124).
In addition, our traditional ways of correcting
speaking errors and giving feedback appear to be

v

ineffective as many of the students tend to make
the same mistakes that have already been corrected
before. This may be because once the students are
really involved in the speaking, they normally
focus their attention on the content of what they
intend to say, “rather than on potentially helpful
signals from their interlocutor about what they
have already said” (Lynch, 2007, p.312).
Bearing those sources of the VADAFA students’
speaking weaknesses in mind, this case study
aimed to answer the following primary question:
• In what ways can task repetition improve

my learners’ accuracy and fluency in their
English oral performance?
And the following subquestion:
• How does the degree of learners’ anxiety
in English speaking change as a result of
being familiar with the task?
It was hoped that the transcribing process,
which makes the student’s speaking visible,
would be a favourable condition for the students
to first of all raise awareness of the strengths and
weaknesses in their oral product. More specifically,
learners are in fact encouraged to “externalize
their thoughts about the formal correctness and
semantic precision of their own output” (Lynch
2001, p.131). In the similar vein, Mennim (2003)
also states that the procedures of this innovation
not only encourage real-time language processing,
but a more explicit language focus that could allow
more time for the noticing of language forms as
well (p.133). Furthermore, the repetition process
enables the automatization of procedures to take
place, which offers learners more attentional
resources to devote to other components of task
execution (Finardi, 2008, p.138). This condition,
hopefully, would by nature result in more fluent
speaking. Consequently, the level of anxiety
when speaking English may decrease as a byproduct. All in all, this route to noticing, supported
KHOA HỌC NGOẠI NGỮ QUÂN SỰ

No. 19 (5/2019)

27


v LANGUAGE TEACHING METHODOLOGY
by reflective self-correction, interactive peer
correction, supplementary teacher intervention
with explicit feedback, would possibly make a
contributing factor to successful language learning
(Lynch, 2001).
A considerable number of professionals in
the research world have also made use of task
familiarity, for various purposes and in various
contexts. One of the pioneers who set the earliest
ground for repetition of task in second language
acquisition may be Bygate Martin and Porter
Don. The three-month study that Bygate and
Porter (1991) carried out in the United Kingdom
was to examine whether repeating a task for the
second time has any effects on students’ language
production in terms of fluency and linguistic
complexity. Other researchers interested in the
influence of repeating a task on fluency, accuracy,
and complexity of second language output include
Mennim (2003), Finardi (2008), Birjandi and
Ahangari (2008), Baleghizadeh and Derakhshesh
(2012). Additionally, Lynch (2001) investigates
the effects of task familiarity on noticing, while
Finardi (2008) examines the relationship between
this variation and working memory capacity.

2. THE STUDY
2.1. Participants
The participants of the current research were three
adult male students studying in one intact English
class, which the researcher described earlier in
this report. One of them is in his late twenties,
and the other two are in their early thirties. At the
time of the study, the English proficiency of two
students was roughly at elementary level, and the
other student was at beginner level. All of them
are military officers who share the same first
language, that is Vietnamese, and similar cultural
background. Although preparation for a speaking
task is something these students are familiar
with, they have been experienced in neither

28

KHOA HỌC NGOẠI NGỮ QUÂN SỰ

No. 19 (5/2019)

videorecording nor transcribing their speech
before. These learners voluntarily took part in the
project by signing a consent in which the purposes
as well as the procedures of the study were
thoroughly stated, and the participants’ rights were
clearly mentioned. Further, one of the ways the
researcher guaranteed the subjects’ confidentiality

was to name them pseudonymously.
2.2. Methodology
First and foremost, the researcher’s choice of
qualitative case study was initiated by Bent
Flyvbjerg’s viewpoints as he argues that the
closeness of the case study to real-life situations and
its multiple wealth of details are not only important
for the development of a nuanced view of reality,
but also for researchers’ own learning processes in
improving the skills needed to do good research
(Flyvbjerg, 2006, p. 223). More specifically, case
studies produce context-dependent experience and
knowledge, which is essential for any researchers
who aim to develop from rule-based beginners to
high level experts, as Flyvbjerg (2006) expands
his ideas.
Most importantly, qualitative case study is
pertinent to this research aim in that, according
to Baxter and Jack (2008), this design empowers
the researcher to answer “how” and “why” type
questions, and at the same time takes the influential
contextual factors into consideration. For such a
novice researcher, a case study is an exceptional
opportunity to gain enormous insight into a case
(Baxter and Jack, 2008, p.556). This is because
by means of gathering data from various sources,
the researcher can explore a phenomenon through
a variety of lenses, hence multiple facets of the
issue are able to be illuminated and understood
(Baxter and Jack, 2008, p.554). The concerned

issue is therefore scrutinized from a holistic view,
and biased interpretation of data is hoped to be
avoided as a result. Another plausible reason was
that qualitative studies enable the researcher to


LANGUAGE TEACHING METHODOLOGY

explore new areas of research via its “principle
of openness” (Flick et al., 2004, p.9), thus
professional reflection and development becomes
a nonstop process.
Yet two of the drawbacks of qualitative case
study are claimed to be the lack of rigour and weak
grounds for establishing reliability and generality
as a consequence of small sampling (Zainal,
2007, p.5). In an attempt to minimize these
disadvantages, the researcher triangulated the data
sources by using three tools, including physical
artifacts, videotapes, and face-to-face individual
semi-structured interviews. The first source of data
is in the form of the transcripts of student’s oral
performance in time 1 and time 2. Lynch (2007)
points out that by examining their transcription,
the students are more likely to be able to notice,
remember, and produce reprocessed forms
highlighted in the feedback (p.312). This visible
form of speech was obviously crucial for the
participants to analyse not just the use of language
forms, but the pronunciation of words as well.

Accordingly, the use of and changes in students’
interlanguage may be easier to be evaluated.
Videotapes was a dual-purpose tool because they
provided audio source for the transcribing as well
as the analysing of sounds, while it was hoped to
better the researcher’s verification of the changes
in students’ affective variables. The reason for the
use of videotape is that the researcher could obtain
supplementary information on body language
and facial expressions from this source (Hinchey,
2008, p.85), though at some surface level. The
last method of data collection, which is semistructured interviewing, is significantly beneficial
to the research aims. One of the rationales for this
method lies in the fact that qualitative interviews
can provide rich and in-depth information about
the experiences of individuals, as concluded by
Dicicco-Bloom and Crabtree (2006). By means
of talking face-to-face to the researcher, the
participants have a chance to provide the most direct,
relevant, and hopefully reliable information about
the issues being studied (Hinchey, 2008, p.81).

v

2.3. Procedures and data collection
The research site was an ordinary classroom at
VADAFA with pleasantly quiet surroundings. The
stages of the two-week research are as follows:
Stage 1: On the 18th February 2013, each of
the three students (whose pseudonymous names

are Minh, Thanh, and Hieu) was given a topic
in which they had to narrate their last holiday in
about five minutes. Narrating has been a common
activity in this course, yet they were not told about
this specific topic until the first stage took place.
They had five minutes to prepare for the speaking,
and could make some notes on a piece of paper.
Stage 2: The students spoke without scripts in
front of the researcher and other two participants.
Their performance was videotaped.
Stage 3: At home, the students watched and
listened to their performance again using their
personal computer, and handwrote their entire
speech verbatim with double-spacing. They then
tried to correct any mistakes by themselves in red
pen, before doing peer correction with another
participant. (Transcript 1).
Stage 4: Four days after stage 3, the students
gave transcript 1 to the teacher, who then indicated
further corrections and feedback on what the
students had missed. (Transcript 2).
Stage 5: One day later, the researcher returned
transcript 2 to its original writer. At this stage,
the participants re-read their transcript, and they
could ask about any points they were unclear in
the feedback. On the same day, she spent forty five
minutes to revise some of the basic linguistic points
that the students had trouble with, and about one
hour to drill pronunciation of the words that they
failed to utter in their speech. Also, the researcher

introduced some supplementary materials for selfstudy, and had them practise pronunciation with
the Pronunciation Power software at home.
KHOA HỌC NGOẠI NGỮ QUÂN SỰ

No. 19 (5/2019)

29


v LANGUAGE TEACHING METHODOLOGY
Stage 6: The second trial speaking was
videorecorded on the tenth day of the study. This
time the transcribing was done by the researcher.
Stage 7: On the 4th March 2013, the students
were invited to take part in an individual indepth
interview, with the questions being sent to them in
advance so that they could have time to consider
their answers to those questions. The interviews
lasted about twenty minutes each, and were
taperecorded. This final stage aimed to delve into
the learners’ views on their own English speaking
problems, their understandings of the innovation
process, and most crucially, their appraisal of the
research as a whole in regards to their learning
style and context.
Basically, the case study followed Mennim
(2003) and some of the aforementioned scholars’
procedures, yet it was different from the previous
research in some fundamental points. First of all,
most of these studies were done within classroom

hours, and covered a fairly long time span, which
is up to three months, whereas this two-week
project took place after class hours, in students’ self
study time. This is because the three participants
had to attend normal lessons in the morning
with the rest of the class. However, in the selfstudy session, they also need to be present in the
classroom for two hours in the afternoon, without
teacher’s monitoring. In addition, the very busy
working schedules of both the researcher and the
participants restricted the length of time allocated
for the research, so the researcher had the learners
complete stage three at home in their own time.
The second difference is that while the subjects
in previous studies were mainly at intermediate
English proficiency level, all of the participants in
this study were at low level when they started to
take part in the research. Taking these contextual
features into great account, the speaking time for
the topic was restricted to as short as five minutes,
and only a limited number of basic linguistic
features was measured. That is to say the term
“accuracy” in my research question limits itself

30

KHOA HỌC NGOẠI NGỮ QUÂN SỰ

No. 19 (5/2019)

to the ability to build simple sentences using

appropriate verb forms, subject-verb agreement,
and correct forms of plural nouns. Accuracy here
also means producing understandable segmental
aspects of pronunciation, which are individual
sounds. Evidence of increased fluency may take
various forms, yet due to the small scope of the
present study, the amount of pausing and selfrepair was taken into account in order to examine
the effects of increasing task familiarity. According
to Bygate and Porter (1991), pausing is taken as
an indication of the number of selection or access
problems engaged by the speaker, and includes
individual filled and unfilled pauses (p.42). Repairs
are considered as false starts, repetitions of words
or utterances, incomplete fragments, redundant
repeated words, which indicates the speaker’s
hesitation about lexical decision (Bygate & Porter,
1991, p.42).
As far as the project was carried on, there
appeared a few practical ethical issues that are
worth reporting. The first constraint was again
related to the time arranged for the stages. As two
of the participants often had unexpected personal
business to do during the research process, the
researcher had to rearrange stage 6 and stage 7
twice so that every subject could manage to be
ready for the activity. Given the fact that people
are not always willing to speak truthfully during
interviews (Hinchey, 2008, p.81), on the one
hand, the researcher clearly advised them of the
protection of the learners’ anonymity and the

confidentiality of their information, and she tried to
develop a rapport with these learners on the other
hand. Dicicco-Bloom and Crabtree (2006) argue
that rapport, which includes trust and respect for
the interviewee and the information s/he shares,
plays a vital role in the interview (p.316). By setting
up a secure and comfortable environment for the
videoing as well as interviewing, this positive
relationship was successfully created during this
entire research process, particularly during the
interviewing time. Besides, it seems that as the


LANGUAGE TEACHING METHODOLOGY

subjects are all adults, they were free from any fears
of telling the truth of their speaking shortcomings
and the reasons behind that. Another potential
issue was the English translation of the interviews.
As all of the students were low-level English
speakers, the interviews were done in Vietnamese,
which might result in the unreliable translation
into English as well as misrepresentation. The
researcher therefore asked the participants to
check the interview’s transcripts for clarification
and admendment if needed. After that, she invited
another competent colleague to read the translation
and gave feedback on unclear points. Furthermore,
while analysing the interviews, she relistened to
the original recording so as to ensure the nature

of the learner’s intonation, pauses, and hesitations.

3. DATA ANALYSIS
The data will be presented on a theme-by-theme
basis, using qualitative content analysis approach,
yet including some quantitative summary. The
researcher compared the rehearsed transcription
with the final one to evaluate the changes to
students’ use of verb forms, subject-verb agreement,
and plural forms of nouns. Also, the changes to
the pronunciation as well as the level of fluency
were investigated through both the transcripts and
videorecordings. Changes to levels of anxiety was
evidenced mostly from the interview, and some
from the videos. Surprisingly, some changes in
content will be discussed here, though they are not
the original focus of the research. It is important
to note that the students were supposed to speak
in five minutes, but on both occasions they spoke
for less than the allowed time, maybe due to their
inadequate interlanguage. This results in the
amount of evidence exemplified hereafter.
3.1. Verb forms
In general, the group successfully modified their
use of verb forms after reviewing their rehearsed
transcripts. On the first artifacts, the total number

v

of mistakes spotted by both the students and the

researcher is 11, and about half of them were
successfully repaired in the second trial speaking.
Table 1. Response to feedback: Verb forms
Student

Number of
suggested
repairs
on 1st
transcript

Targetlike
repairs in final
presentation

Structure left
out in final
presentation

Minh

3

2

1

Thanh

6

3

2

Hieu

2

1

1

Hieu made the fewest mistakes of verb forms,
but he did not repeat one of them in his final
speech. Thanh reused more than half of the verbs,
he left one highlighted verb unrepaired:
Rehearsal

Final Performance

We get up early and ... walking

We got up early and...and...
walk...and walking

Thanh showed his confusion about the form
of the verb “walk”, which might be explained by
the fact that he had had little practice on using
simple sentences with compound predicates. This

is coupled with his opinion in the interview about
his English speaking weaknesses:
“...One of the most serious difficulties I have
been facing in speaking English was the insufficient
knowledge of grammar...,... while the teachers have
not created regular speaking practice activities”.
Both Minh and Thanh replaced their suggested
repairs with more familiar verb forms.
Student

Rehearsal

Final Performance

Minh

We enjoy it so much.

We loved the holiday so much.

Thanh

I like saw the fields.

I liked the fields.

Basically, the students were able to make their
own alternations to the use of most verb forms in
the final speaking, yet to some extent, they seemed
to be afraid of repeating the same mistakes, even

KHOA HỌC NGOẠI NGỮ QUÂN SỰ

No. 19 (5/2019)

31


v LANGUAGE TEACHING METHODOLOGY
when those mistakes had been already corrected
by the teacher on the first transcript.
3.2. Subject-verb agreement
As the student narrated a past event, the instances
of subject-verb agreement were not frequently
observed. Totally, 4 errors were noted on the
rehearsals.
Table 2. Response to feedback: Subject-verb
agreement
Student

Minh

Suggested
repairs on 1st
transcript

Targetlike
repairs in final
presentation

Structure left

out in final
presentation

- It were next
to...
- It look like a...

- It was next to...

1

Thanh

- She live with ...

- She lives with ...

0

Hieu

- Its were in June.

- It was in June.

0

-Ø

Unsurprisingly, all the examples of this type of

fallacies lie in the third person singular subject-verb
agreement, which is a very common trouble when
the VADAFA students use present simple verbs.
Three of the errors were noticed and successfully
modified by the students themselves as they
worked with their peer before the researcher made
the final feedback.
3.3. Plural forms of nouns
The amount of suggested corrections of this
category vary greatly across the subjects. The
largest number of mistakes was found in Minh’s
speech, and he responded very well to five out of
six spotted occasions. Besides, he was the only one
who misused two new plural nouns in the second
presentation. This may mean that Minh needs
more time to practise this specific linguistic area.
Thanh made the most mistakes on verb forms,
but he showed a good grasp of plural forms of
nouns, though he used only 3 plural nouns in his
first speech. Also, very few plural nouns were

32

KHOA HỌC NGOẠI NGỮ QUÂN SỰ

No. 19 (5/2019)

Table 3. Response to feedback: Plural forms of
nouns
Student Number of Targetlike

Nouns left
suggested
repairs
out in final
repairs on 1st
in final
presentation
transcript presentation

New
suggested
repairs
in final
presentation

Minh

6

5

1

2

Thanh

0

0

0

0

Hieu

2

2

0

0

observed in Hieu’s first transcript, and he included
changes to both instances in the final performance
as a result of collaborating with one of his partners.
It is worth noting that both Minh and Hieu left out
the final letter “s” in all the 8 nouns highlighted
in the first transcripts. This can be explained by
that they did not know when to use a plural noun,
or that they understood the rules of using plural
nouns, but failed to utter them. The former could
be a more reasonable explanation because at the
time the study was implemented, they learners
had merely had some practice on spelling rules
of plural nouns without being provided with the
clarification of their use. However, Thanh was an
exception because before this course, as he stated

in the interview, he had learnt English for longer
time than the other two participants and might
have practised using plural nouns more often.
3.4. Pronunciation of sounds
The most frequently occurring inaccuracies noted
in the students’ rehearsal presentations belong
to this theme. Totally, 27 mispronounced words
were highlighted, yet only about half of them were
correctly altered by the students. It seems that most
of the recalled words sound more similar to some
of Vietnamese sounds such as also, ice creams,
holiday, rice, as crowded as.
In addition, all the participants had difficulty
pronouncing newly-added words in their second
speaking, with Hieu facing the most challenges.


LANGUAGE TEACHING METHODOLOGY

Table 4. Response to feedback: Pronunciation of
sounds
Student

Number of Targetlike Words left
New
suggested
repairs
out in final
suggested
repairs

in final
presentation repairs in final
on 1st
presentation
presentation
transcript

Minh

11

5

3

2

Thanh

8

4

0

2

Hieu

8

5

0

5

He failed to fix three words (especially, ancient,
much) when repeating the task, while including
five more cases of mispronunciation in his second
trial, which are went, last, stayed, restaurant,
talk. These examples indicate Hieu’s inconsistent
imperfection in different aspects of sounds, that is
he had troubles with pronouncing not only vowel
and consonant sounds, but also cluster sounds.
Minh made his own alternations to
pronunciation of less than half of the spotted
words. The three repeated errors include next, West,
wild, and the two new errors include August and
walk. He appeared to find cluster sounds the most
challenging because he pronounced next as /nek/,
West as /wet/, wild as /wai/, August as /ɔːgə/. Unlike
Minh, Thanh was unable to fix the words with /ʃ/
sounds, including fishing, fresh, she, and some final
sounds, as in will and wife in his second speech.
Although the teacher’s feedback, one-hour
pronunciation drilling lesson, and student’s self
practice with Pronunciation Power software
could result in only half of fruitful targetlike
modifications, these interventions helped the

students, though to a modest extent, achieve such
nontargetlike repairs as word stress, schwa and
sound linking.
Schwa
Minh: the bank of the lake, around
­Thanh: a long time ago
Sound linking
Thanh: and I

v

Hieu: was in
The numerical evidence in Table 5 clearly
demonstrates that these students found
pronunciation of sounds more challenging than
other aspects of English language mentioned in
this study. This is in accordance with their opinions
in the interviews.
Minh:
“... One of the problems I have in speaking
English is the pronunciation of individual sounds
because I could not pronounce the words correctly
right at the beginning,... so it becomes my
pidginised language.”
Thanh:
“...even some of the teachers do not have good
pronunciation, so it is difficult for the students to
know whether they themselves pronounce correctly
or not... , ...the accuracy in pronunciation is one of
the most serious problems to me... ”

Hieu:
“...I find vocabulary and pronunciation the
most problematic”.
Paradoxically, at VADAFA, the majority of the
current English teachers were originally trained
as Russian teachers. When English language
became one of the main foreign languages to be
taught at this school about twelve years ago, those
teachers joined a two-year English course as an
ad hoc solution to the new working situation. As
a result, not only their pronunciation of English is
strongly influenced by Russian accent, but many
of them underestimate the importance of teaching
pronunciation. Most of the teachers who have been
teaching these three participants are no exception,
which may explain why Minh could not pronounce
the words correctly at the first time, while Thanh
found it difficult to know whether he pronounces
the words accurately or not.
KHOA HỌC NGOẠI NGỮ QUÂN SỰ

No. 19 (5/2019)

33


v LANGUAGE TEACHING METHODOLOGY
3.5. Fluency
The researcher followed Bygate and Porter’s
(1991) procedures of observing changes in

students’ fluency, in which the ratios of the amount
of pauses and the amount of self-repair to number
of words were calculated. This is to say the higher
the ratio is, the less fluent their speech is.
Table 5. Comparison of fluency measures on the
task at Time 1 and Time 2.
Student
Ratio
pauses: number
of words
Repairs: number
of words
Total

Minh
Thanh
Hieu
Minh
Thanh
Hieu
Minh
Thanh
Hieu

Narrative
task (Time 1)

Narrative
task (Time 2)

0,11
0,31
0,36
None
0,11
0,06
0,11
0,42
0,42

0,11
0,10
0,16
0,02
0,04
0,03
0,13
0,14
0,19

Two out of three students (Thanh and Hieu)
showed dramatic improvements in fluency in the
familiar mode, with regard to a decrease in pauses
and in self-repairs. However, given that Hieu had a
lower general English proficiency than Thanh, he
created more pauses at Time 1 and at Time 2 than
Thanh. Also, Hieu did not make many self-repairs
because perhaps his interlanguage was limited.
Minh was the only one who went against the trend
of the other two subjects as he made no changes to

the proportion of pauses, and even increased his
self-repairs at Time 2. Nevertheless, it is important
to note that amongst the three subjects, Minh
committed the least pauses as well as repairs,
which also means he was a more fluent English
speaker than Thanh and Hieu.
3.6. Degree of anxiety
This affective aspect proved to be the most
difficult to be measured. When scrutinizing the
videorecordings, the researcher found very little

34

KHOA HỌC NGOẠI NGỮ QUÂN SỰ

No. 19 (5/2019)

evidence of improvement in students’ confidence.
The data therefore came mostly from the interviews.
Thanh was the only one who stated that “so far
anxiety when speaking English is not a problem for
me”, yet he also confirmed that “After seeing and
understanding my weaknesses in the transcript,
and being actively involved in diminishing these
weaknesses, I felt more confident”. Unlike Thanh,
anxiety when speaking English was a real problem
for Hieu and Minh. As a result of being familiar
with the task, both the two students expressed
that in the second speaking, they felt a little bit
more confident, but the feelings of anxiety and

nervousness still existed.
Minh:
“I felt a bit less anxious because after my
mistakes were corrected, I reviewed them,
familiarized myself with the words and structures,
and was no longer afraid of making these mistakes,
especially the ones on pronunciation”.
3.7. Elaboration of content
This was an unexpected area of student focus,
yet it is worth some analysis as it may open new
aspects for future investigation. Quantitatively, in
their second trial, Minh and Hieu increased the
amount of independent clauses, while Thanh did
not produce extra sentences.
Table 6. Number of independent clauses and time
of speech
Student

Number of
independent
clauses-Time 1

Number of
independent clausesTime 2

Minh

16

18

Thanh

19

19

Hieu

13

15

Nevertheless, on a qualitative basis, all of the
three participants not only inserted new contents
but also modified their language so as to make


LANGUAGE TEACHING METHODOLOGY

it more comprehensible in the final presentation.
In many instances, the elaboration of content
aimed to expand the ideas and add more detailed
descriptions.
Student

Content Time 1

Content Time 2

Minh

...the weather was so
good.

...the weather was so good. It
was sunny and...windy, and
very...pleasant.

...and
my
two
sister were very...
interested in them.

...My sisters were very
interested in them. They fed the
monkeys bread and banana...

Thanh

...my grandmother...
can walk

My grandmother ...can walk.
Si...very kind, very funny. And I
and my wife love si very much...

Hieu

I liked food here...so
much

Weee ...had a party ...inn a
restaurant near the sea. I...I
liked food here so mut.

Interestingly, all of the students made some
alternations to contents, that is they left out some
information in the first speech, and/or added
different ideas in the second one. For example, the
sentence “We take photograph in My Khe beach”
was not repeated in Hieu’s final presentation.
Instead, he ended his narration by making a
completely new utterance: “I enjoyed traveling
to Danang, I hope I will return Danang soon.”
Obviously, the modification of content helped
the speech more refined, and accordingly, more
sophisticated.

4. DISCUSSION
4.1. Discussion
In general, the learners showed a certain extent
of improvement in both accuracy and fluency in
performing the familiar narrative task after about
two weeks, with only one exception of Minh on
fluency. Bearing in mind that Minh’s English
proficiency was the best of the three students, as
shown in his classroom performance and regular
test marks, task repetition may not be much

effective in accelerating fluency of higher level

v

students, whereas the effects were more clearly
identified in lower level learners. In Minh’s
case, the reason could be that he added more
new information in his second story than the
other two participants, thus executing more selfrepairs was inevitable. This result was partly in
line with Bygate and Porter’s (1991) research in
which their two students coming from English as
a Foreign Language (EFL) backgrounds gained
more fluency in the repeated task than the one who
spoke English as a second language.
As subject-verb agreement and verb forms are
not totally new linguistic areas to these students,
they made relatively successful alternations to
these features. However, it is possible that one
of the ways these students avoided repeating the
mistakes was to eliminate some of the sentences
which included the highlighted verbs in the first
transcript, even when these verbs were already
feedbacked by the researcher. The changes to
plural nouns were somehow productive partly
because adding letter “s” to a noun is a more
simple way than adding “es” or utterly changing
the spelling of it. As a result, the students perhaps
found it fairly easy to pronounce this kind of plural
forms. However, because only a limited number of
plural nouns were used by the three learners, plus

Minh had the same trouble with two new nouns
in his second speaking, it might be concluded that
these participants need more explanations and
practice on how to use plural nouns.
Pronunciation of individual English sounds,
unfortunately, has never been systematically
introduced to these students before. Neither had
they had a chance to practise pronunciation with
a software. In addition, the teacher’s intervening
lesson was too short to provide a thorough drill
on the entire sound system, whereas student’s
self study with Pronunciation Power software
in fact depended much on each individual’s
time management. These might explain why the
subjects demonstrated fairly modest advancement
in this particular section. Despite this fact, a bright
KHOA HỌC NGOẠI NGỮ QUÂN SỰ

No. 19 (5/2019)

35


v LANGUAGE TEACHING METHODOLOGY
signal was that all the students agreed that the
transcribing process helped them be aware more
of their pronunciation than previously, and they
paid more attention to the sounds when starting
to speak. This short study could not lead to a
significant decrease in anxiety, especially in the

cases of Minh and Hieu, yet the confidence was
somewhat observed in their second performance.
Moreover, the most striking result was possibly the
modification of content, which was most overtly
seen in Minh’s and Thanh’s transcript. It could
be understood that the planning time allowed the
students to extend their thoughts, and that some
changes in the content might have helped them
avoid repeating the mistakes.
Generally, the learners followed the stages of
task repetition closely, except for the transcribing.
Despite having been clearly instructed and
modeled by the researcher, all of them failed to
write their articulation word for word at first. They
either missed the pauses and repetitions of words,
or wrote down the words they thought they had
said rather than the ones they actually articulated.
The researcher therefore had to edit some pieces
of their information. The reasons behind this could
be because these learners had had no experience of
transcribing verbatim so far, and they did this step
at home, thus could not get immediate help from
the researcher when necessary.
4.2. Reflections on future work
The present project was supposed to be the starting
point of the researcher’s ongoing cyclical reflective
practice. For more fruitful results in future
research, she may need to adjust a few points.
First of all, the rehearsed transcribing should be
done at the presence of the researcher, just after

students finish their recording. It is hoped that
this could ensure the exactness of the transcript.
Moreover, when students are familiar with such
written reproduction, the final performance may
be also transcribed by the speaker themselves. It is

36

KHOA HỌC NGOẠI NGỮ QUÂN SỰ

No. 19 (5/2019)

believed that learners will benefit from the second
transcribing the way the benefit from the first one,
not to mention they can see first hand the changes in
their interlanguage, which could be a motivational
factor. Additionally, this helps the researcher save
a lot of time for the whole research process. Also,
to deal with the problem of large class, she will
have them record their speaking in the language
labs where there are enough computers for more
than fifty students to work synchronically.
Secondly, various types of task will be used
so as to investigate the multifaceted effects of
repetition. This idea derived from two of the
students’ opinions in the interview when the
researcher asked them about what she should do
differently to improve task repetition in the future.
A circle of two weeks is believed to be enough for
reviewing and repeating a task, yet it is unsure to

guarantee the longer term effect of the innovation.
The future projects would hence include successive
cycles of task repetition throughout the course in
order to make it become a regular practice among
students. Finally, the researcher may adapt this
intervention for writing skill because basically,
reviewing a writing piece is to a large extent
comparable to reviewing a transcript.

5. CONCLUSION
The case study’s primary aims were to explore
in what ways repeating a narrative speaking task
would result in student’s changes to accuracy and
fluency. It also sought to examine the changes to
level of anxiety when speaking as a secondary
aim. Regarding accuracy, the findings revealed
in a fairly similar pattern for all the subjects, that
is they generally showed improvements after
two weeks. However, it can also be concluded
that the influence of task repetition on accuracy
could have been more effective if the students had
been presented and had practised those linguistic
features more profoundly earlier in the course.
The results of fluency took different ends since the


LANGUAGE TEACHING METHODOLOGY v

highest level student was immune to the repetition
of task, while the other lower level students

responded rather well to it. Some minor decrease
in anxiety was also observed, yet there should
be more similar practice in the future in order to
achieve significant improvements in this affective
variable. Interestingly, the modification of content
recorded in the learners’ second presentations
opens a new research domain. It was hoped that
this small scale study would set foundations for
the researcher’s future innovations, and that by
means of gaining such little changes over the
course, the VADAFA students would consequently
make substantial gains in the foreign language./.
References:
Baleghizadeh, S. & Derakhshesh, A. (2012). The effect of
task repetition and noticing on EFL learners’ oral output.
International Journal of Instruction, 5(1), 41-152.
Baxter, P., Jack, S. (2008). Qualitative case study methodology:
Study design and implementation for novice researchers.
The Qualitative Report, 13(4), 544-559.
Birjandi, P. & Ahangari, S. (2008). Effects of task repetition
on the fluency, complexity and accuracy of Iranian EFL
learners’ oral discourse. The Asian EFL Journal, 10(3), 2852.

Bygate, M. & Porter, D. (1991). Dimensions in the acquisition
of oral language. Language acquisition and the second/
foreign language classroom, 28, 38-48.
Dicicco-Bloom, B. & Crabtree, B. (2006). The qualitative
research interview. Medical Education, 40, 314-321.
<Doi: 10.1111/j.1365-2929.2006.02418.x>.
Finardi, K. (2008). Working memory capacity and L2 speech

production in a picture description task with repetition.
Revista de Estudos da Linguagem, 16(2), 129-144.
Flick, U., von Kardoff, E., Steinke, I. (2004). A companion to
qualitative research. London: SAGE Ltd.
Flyvbjerg, B. (2006). Five misunderstandings about case-study
research. Qualitative Inquiry, 12(2), 219-245.
Hinchey, P. H. (2008). Developing a research plan and
identifying a research question. Action research. New
York: Peter Lang Publishing.
Lynch, T. (2001). Seeing what they meant: transcribing as a
route to noticing. ELT Journal, 55(2), 124-132.
Lynch, T. (2007). Learning from the transcripts of an oral
communication task. ELT journal, 61(4), 311-320.10.1093/elt/ccm050>.
Mennim, P. (2003). Rehearsed oral L2 output and reactive focus
on form. ELT journal, 57(2), 130-138.

ẢNH HƯỞNG CỦA VIỆC LẶP LẠI NHIỆM VỤ TỚI ĐỘ TRÔI CHẢY VÀ ĐỘ CHÍNH
XÁC TRONG KỸ NĂNG NÓI TIẾNG ANH CỦA HỌC VIÊN CẤP ĐỘ THẤP: MỘT
NGHIÊN CỨU TÌNH HUỐNG TẠI HỌC VIỆN PHÒNG KHÔNG-KHÔNG QUÂN
LƯƠNG THỊ PHƯỢNG
Tóm tắt: Mục đích chính của nghiên cứu tình huống này là điều tra sự ảnh hưởng của việc lặp lại
nhiệm vụ tới độ chính xác và độ trôi chảy trong kỹ năng nói tiếng Anh của ba học viên. Bên cạnh
đó, nghiên cứu cũng xem xét sự thay đổi trong tâm lý khi thực hiện kỹ năng nói. Phương pháp chủ
đạo là nghiên cứu định tính, với sự bổ trợ của một vài phân tích định lượng. Kết quả nghiên cứu cho
thấy, nhìn chung cả ba học viên có tiến bộ về độ chính xác, trong khi độ trôi chảy cho kết quả không
giống nhau. Đồng thời, sự hồi hộp trong khi nói của ba học viên cũng giảm. Trong lần thực hiện nói
thứ hai, tất cả các học viên đều có điều chỉnh về nội dung, điều này mở ra hướng nghiên cứu mới cho
tác giả trong tương lai.
Từ khóa: độ chính xác, độ trôi chảy, lặp lại nhiệm vụ, dạy kỹ năng nói tiếng Anh, Học viện Phòng

không – Không quân
Ngày nhận bài: 23/12/2018; ngày sửa chữa: 31/01/2019; ngày duyệt đăng: 28/4/2019

KHOA HỌC NGOẠI NGỮ QUÂN SỰ

No. 19 (5/2019)

37