enabling malaysian university students to speak english and enjoy

1

One semester of speaking tasks – an

experimental approach with low proficiency

students

Subject areas: English; pedagogical approaches; low proficiency students;

oral interaction; language acquisition; communicative language teaching

2

One semester of speaking tasks – an experimental approach with low

proficiency students

Andrew N Williams1 and Yah Awg Nik

2

1Centre for Language Studies and Generic Development, Universiti Malaysia

Kelantan, Karung Berkunci 36, Pengkalan Chepa, 16100, Kota Bharu, Kelantan

Darul Naim, Malaysia.

Tel: +609 771 7163

Fax : +609 771 7282

Mobile : +6017 286 4409

E-mail: [email protected]

2Centre for Language Studies and Generic Development, Universiti Malaysia

Kelantan, Karung Berkunci 36, Pengkalan Chepa, 16100, Kota Bharu, Kelantan

Darul Naim, Malaysia.

Tel : +609 771 7277

Fax : +609 771 7282

Mobile : +6019 980 8719 / +6019 750 7847

E-mail : [email protected]

3

One semester of speaking tasks – an experimental approach with low

proficiency students

Although there are many students who enter Malaysian universities with a high

level of proficiency in English (MUET Band 3 and above) there are many

others who struggle to express themselves in English (MUET Band 1 and 2).

What would happen if all their class time in the first semester was used for

speaking tasks? This study tested the hypothesis that undergraduate students

with low proficiency, (MUET Band 1 and 2), would make the most rapid

progress in English if the guided learning time (tutorials and lectures) in the

first semester is used entirely for speaking tasks. A class of 30 students was

taught using all 42 hours of contact time for speaking tasks, with a control class

taught „as normal.‟ Proficiency in the four language skills were measured at the

start and end of the semester.

Keywords: English; Malaysian undergraduates; low proficiency; speaking;

undergraduates.

4

Introduction

In Malaysia, children are formally taught English from the beginning of their

first year of primary school, aged seven years old. Although optional, many will

have also attended kindergarten for up to three years before primary school

where they will also have been taught English.

After six years of primary school education and five years of secondary

education, students then have the option of two years of sixth form study in

schools or one to two year long matriculation courses. During this time all

students who wish to enter public universities have to take the Malaysian

University English Test (MUET).

Thus, by the time prospective Malaysian university students take MUET they

will have been formally taught English for at least eleven years.

Although there are many students who have a good level of English, achieving

MUET Band 3 and above, there are many students who enter university with a

very low level of proficiency scoring only Band 1 or Band 2 at MUET,

hereafter referred to as „Low Proficiency‟ students. The Malaysian

Examinations Council defines the communicative ability of someone scoring

Band 1 as „Hardly able to use the language‟ and Band 2 as „Not fluent‟,

(Malaysian Examinations Council, 2006). A more detailed description of

Bands 1 and 2 is shown in Table 1.

5

TABLE 1

Descriptions of language ability for MUET Band 1 and Band 2 scores

(Malaysian Examinations Council, 2006)

Band User Communicative

ability

Comprehension Task

Performance

2 Limited

user

Not fluent;

inappropriate use

of language; very

frequent

grammatical

errors

Limited

understanding of

language and context

Limited ability to

function in the

language

1 Very

limited

user

Hardly able to use

the language

Very limited

understanding of

language and context

Very limited

ability to function

in the language

6

Despite exhibiting such limited ability in the language, it would seem likely that

during their 11 or more years of English classes these students must have

developed some sort of latent knowledge of vocabulary and grammar.

In Malaysia there is a great need for undergraduates to develop good oral

communication skills. Undergraduates often have to make presentations in

English and are expected to be able to enter into discussions in their field using

English. When they graduate they face interviews which are often carried out in

English and if they get through the interview many will have to use English in

the workplace as the language of business. At the same time those students

who try to use English to speak are often criticised by their friends for trying to

show off, or laughed at for their attempts. This creates an atmosphere where

many Malaysian students are extremely hesitant to speak English and do not

have many opportunities to practise, resulting in a low level of fluency and lack

of confidence in speaking. It would appear that the most pressing need for

undergraduate students is the need to be able to speak English at a level

sufficient to do business and yet Low Proficiency students struggle to even

express themselves in a few words.

There have been a range of studies that have shed some light on the situation of

Malaysian English learners: Mathi, Jamian and Nair (2008) evaluated the

vocabulary knowledge of students at one Malaysian university and reported that

this was „far below the university threshold level.‟ In a study into the language

7

learning style preferences of low English proficiency students in a Malaysian

university (Ahmad, 2011), all six learning styles investigated were reported as

negative learning styles by the students. It was concluded that this result was

due to the students‟ lack of interest in learning English. Thang and Alias (2006)

studied the degree of autonomy among undergraduates taking English

proficiency courses in three public universities in Malaysia. They reported that

the majority of these students preferred a teacher-centred approach to learning

and suggest that this might be „a washback effect of the “spoon-feed” system

operating in most Malaysian primary and secondary schools.‟

In one study, Malaysian low proficiency students claimed the main factor for

their low proficiency was „the element of ineffective instructional practices‟

with teachers following a „rule-oriented approach‟ and being „too syllabus and

textbook oriented.‟ (Shah, P. (1999). Perceptions of Malaysian ESL Low

Achievers About English Language Learning. Unpublished doctoral

dissertation, University of Connecticut.) However, a study by Thang and Wong

(2005) found that the majority of Malaysian ESL Instructors do actually display

a preference for current learner-centric teaching styles.

Although there have also been a number of studies on ESL teaching and

learning in Malaysia, there seems to be very few quantitative studies on

pedagogical approaches focused on developing oral communication ability.

8

In the field of second language acquisition, Long‟s Interaction Hypothesis

proposes that learners can acquire language through the breakdowns in

communication and negotiation of meaning that occur in oral interaction (Ellis,

2008). Associated with this idea is the Comprehensible Output Hypothesis put

forward by Swain that the „push‟ to produce language that others can

understand will cause the learner to evaluate and adjust their language use

(Ellis, 2008).

Based on this idea that language can be acquired through oral interaction, and

assuming these Low-Proficiency students have an inactive store of language

that could be activated, this study seeks to answer the question “What if all of

the contact hours in the first semester were used for speaking tasks alone?”

The hypothesis of this study is that for Low Proficiency Malaysian students in

their first semester of university, a pedagogical approach that uses all contact

hours for speaking tasks in an English course will be more effective than a

teaching approach that incorporates all the language skills (reading, writing,

listening and speaking), as well as including teacher-led learning of vocabulary

and grammar. The definition of „more effective‟ is that students will show a

significant improvement in their ability to speak in English in terms of fluency,

accuracy, and vocabulary use. Furthermore, they will show as much, or maybe

more, improvement in their reading, writing, and listening ability as students

whose course included these skills in the contact hours.

9

To test this hypothesis two groups of students were compared: an experimental

group was taught using all of the contact hours in the first semester for speaking

tasks. A control group was taught “as normal” – including grammar explanation

and examples; reading; writing and listening tasks. Proficiency of the students

was tested at the start (Pre-Test) and end (Post-Test) of their first semester.

10

Materials and Methods

Control and Experimental Classes

Two groups of students studying the same course were assigned to be the

control group and the experimental group. The control group had fifteen

students who obtained Band 1 in MUET, thirteen who obtained Band 2 and one

other student whose results is unknown. The experimental group had eleven

students who had obtained Band 1, eight who had obtained Band 2, three who

had obtained Band 3, one who had obtained Band 4 and the results of the other

seven are unknown. Unfortunately, the groups were not comparative by gender;

the control group was made up of 28 female students and just one male student

while the experimental group was a mixture of 14 female students and 16 male

students.

Assessment details

The students were tested in the first two weeks of the semester in all the four

language skills (Pre-Test) and then re-tested in the final week of the semester

(Post-Test).

Reading was assessed by two reading passages from a sample MUET paper and

the accompanying 15 questions for each test. The reading passages were

selected to try to ensure that the topic was of comparable level of vocabulary

for the Pre-Test and the Post-Test.

11

For the Writing tests, students had to write a 250 word essay on a sample

MUET essay title. The essay titles were chosen so that the topics were of

equivalent level for the Pre-Test and Post-Test. The essays were marked using a

marking scheme allocating 10 marks for content, 15 marks for language and 5

marks for organisation.

Listening ability was tested using a sample IELTS listening paper with 40

questions. The paper consists of four sections: in the first students listen to a

dialogue from everyday social interaction, in the second it is a dialogue from an

academic setting, the next section is a monologue from a social setting and the

final part is a monologue from an academic setting.

In order to test Speaking, a sample IELTS speaking test was used and testing

was carried out along the lines of an IELTS speaking test. This involves a

single student meeting with the assessor. The first part of the test is a

conversation, the next part is a 2 minute monologue by the student preceded by

1 minute preparation time and the final stage is an academic discussion. The

speaking test was marked by using the IELTS guide to speaking bands (IELTS

Partners, _) and assigning a mark to each descriptor.

Teaching approaches

For the experimental group, all contact hours were used for speaking tasks. The

objective of each session was for the students to spend most of their time

12

carrying out one or more spoken dialogue tasks. However, in order to facilitate

effective dialogue, vocabulary and grammar necessary for the task were

sometimes pre-taught and sometimes the students would listen to a dialogue to

use as a model. Nevertheless, these activities were limited to providing support

to the dialogue task and were given the minimum time possible in order to keep

the dialogue the focus of each session.

The experimental group did do some writing and reading tasks and grammar

exercises for their self-study time outside of contact hours.

For students in the control group, their class time consisted of a more balanced

approach that gave time to reading, writing, listening and speaking tasks as well

as explanation of grammar and grammar exercises. Although speaking tasks

were included in their contact time, it was one component among the other

activities.

The semester consisted of 14 weeks during which the students had 3 hours of

contact time per week, a total of 42 hours. In addition they were assigned self-

study work lasting about 2 hours per week.

13

Results and Discussion

The hypothesis of this study was that for Low Proficiency students in their first

semester of university, a teaching approach that uses all contact hours for

speaking tasks in an English course will be more effective than a teaching

approach that incorporates all the language skills (reading, writing, listening

and speaking), as well as including teacher-led learning of vocabulary and

grammar. Effectiveness was defined as students showing a significant

improvement in their ability to speak in English in terms of fluency, accuracy,

and vocabulary use; as well as showing as much, or maybe more, improvement

in their reading, writing, and listening ability as students whose course included

these skills in the contact hours.

Summary of Results

The results are summarised in Table 2. There was no significant difference

between the control group and the experimental group, in terms of the change

in test scores between Pre-Test and Post-Test. This appears to show that

students taught using all their contact hours for speaking tasks, improved in

reading, writing and speaking to the same degree as students who were taught

in a class that used contact time for reading, writing, listening tasks, and

grammar exercises as well as speaking tasks. This supports the theory under test

but does not show that the experimental approach more effective than the

control approach.

14

TABLE 2

Summary of results

Reading Writing Listening Speaking

Control Group change

in score

Significant

increase

Significant

increase

Non-

Significant

decrease

No change

Experimental Group

change in score

Non-

Significant

increase

Significant

increase

Non-

Significant

increase

Significant

increase

Difference between

groups in change of

score (a)

No

significant

difference

No

significant

difference

No

significant

difference

No

significant

difference

(a) Change in score, (Post-Test – Pre-Test), for control group compared with

change in score, (Post-Test – Pre-Test), for experimental group.

15

Although a comparison of the changes in scores between the groups shows no

significant difference it should be noted that there were some interesting results

when the groups are looked at individually. The experimental group showed a

significant increase in speaking test scores while in the control group there was

no improvement. The scores of the control group for reading rose significantly

while those of the experimental group rose, but not significantly.

Results by Skill

Although there were 29 students in the control group and 30 in the experimental

group, not all the students attended all the tests and this causes the results to

show varying numbers in each group. Because the results do not conform to a

normal distribution, non-parametric analysis were used.

Reading

For students in the control group there was a significant increase in the Post-

Test score (Mdn=6.00) compared to the Pre-Test score (Mdn=5.00), Z=-2.73,

p<0.05, r=0.64. (Tables 3, 4, 5)

Students in the experimental group also increased their score in the Post-Test

(Mdn=6.00) but this was not a significant increase from the Pre-Test score

(Mdn=5.00), Z=-0.88, ns, r=0.18. (Tables 6, 7, 8).

16

Comparing the increases in score between the groups, there was not a

significant difference between the increases in the control group ( Mdn=1.50 )

and the experimental group (Mdn=1.00 ), Z=-1.08, ns, r=0.17. (Tables 9, 10)

17

TABLE 3

Reading Results: Control Group Descriptive Statistics

N Mean

Std.

Deviation Minimum Maximum

Pre-Test score 18 4.61 1.720 0 8

Post-Test score 18 5.89 1.676 3 9

18

TABLE 4

Reading Results: Control Group change in score between Pre-Test and Post-

Test; Wilcoxon Signed Ranks Test ranks

N Mean Rank Sum of Ranks

Post-Test score - Pre-Test score

Negative Ranks 2(a) 4.75 9.50

Positive Ranks 12(b) 7.96 95.50

Ties 4(c)

Total 18

a Post-Test score < Pre-Test score

b Post-Test score > Pre-Test score

c Post-Test score = Pre-Test score

19

TABLE 5

Reading Results: Control Group change in score between Pre-Test and Post-

Test; Wilcoxon Signed Ranks Test Statistics

Post-Test score -

Pre-Test score

Z -2.733(a)

Asymp. Sig. (2-tailed) .006

Exact Sig. (2-tailed) .005

Point Probability .001

a Based on negative ranks.

20

TABLE 6

Reading Results: Experimental Group Descriptive Statistics

N Mean

Std.

Deviation Minimum Maximum

Pre-Test score 23 5.22 1.506 2 8

Post-Test score 23 5.78 2.194 1 10

21

TABLE 7

Reading Results: Experimental Group change in score between Pre-Test and

Post-Test; Wilcoxon Signed Ranks Test Ranks

N

Mean

Rank

Sum of

Ranks

Post-Test score - Pre-Test score Negative

Ranks

9(a) 10.11 91.00

Positive

Ranks

12(b) 11.67 140.00

Ties 2(c)

Total 23




22

TABLE 8

Reading Results: Experimental Group change in score between Pre -Test and

Post- Test; Wilcoxon Signed Ranks Test Statistics

Post-Test score

- Pre-Test score

Z -.876(a)





23

TABLE 9

Reading Results: Comparison of change in score for Control Group with that of

Experimental Group; Mann-Whitney Test Ranks

Group N Mean

Rank

Sum of

Ranks

Post-Test score – Pre-

Test score

Control 18 23.25 418.50

Experimental 23 19.24 442.50

Total 41

24

TABLE 10

Reading Results: Comparison of change in score for Control Group with that of

Experimental Group; Mann-Whitney Test Statistics

Post-Test score–

Pre-Test score

Mann-Whitney U 166.500

Wilcoxon W 442.500

Z -1.081




25

Writing

The Post-Test writing scores for the control group (Mdn=13.00) were

significantly higher than the results for the Pre-Test score (Mdn=12.00), Z=-

3.67, p<0.05, r=0.78. (Tables 11, 12, 13)

For the experimental group, the Post-Test scores (Mdn=12.00) were also

significantly higher than the Pre-Test scores (Mdn=11.00), Z=-2.62, p<0.05,

r=0.51. (Tables 14, 15, 16).

Comparing the increases in score between the groups, there was not a

significant difference between the increases in the control group (Mdn=3.00)

and the experimental group (Mdn=2.00), Z=-1.31, ns, r=0.19. (Tables 17, 18).

26

TABLE 11

Writing Results: Control Group Descriptive Statistics

N Mean Std.

Deviation

Minimum Maximum

Pre-Test score 23 11.83 3.460 3 19

Post-Test score 23 14.17 2.949 9 20

27

TABLE 12

Writing Results: Control Group change in score between Pre-Test and Post-


N Mean

Rank

Sum of

Ranks

Post-Test score – Pre-Test

score

Negative

Ranks

4(a) 3.50 14.00

Positive

Ranks

18(b) 13.28 239.00

Ties 1(c)

Total 23

a Post-Test Score < Pre-Test Score

b Post-Test Score > Pre-Test Score

c Post-Test Score = Pre-Test Score

28

TABLE 13

Writing Results: Control Group change in score between Pre -Test and Post-


Post-Test Score –

Pre-Test Score

Z -3.673(a)





29

TABLE 14

Writing Results: Experimental Group Descriptive Statistics

N Mean Std.

Deviation

Minimum Maximum

Pre-Test Score 27 11.41 3.411 4 17

Post-Test Score 27 12.85 2.476 8 18

30

TABLE 15

Writing Results: Experimental Group change in score between Pre -Test and

Post- Test; Wilcoxon Signed Ranks Test Ranks

N Mean

Rank

Sum of

Ranks

Post-Test Score -

Pre Score

Negative

Ranks

8(a) 6.56 52.50

Positive

Ranks

15(b) 14.90 223.50

Ties 4(c)

Total 27

a Post-Test Score < Pre Score

b Post-Test Score > Pre Score

c Post-Test Score = Pre Score

31

TABLE 16

Writing Results: Experimental Group change in score between Pre -Test and


Post-Test Score –

Pre-Test Score

Z -2.622(a)





32

TABLE 17

Writing Results: Comparison of change in score for Control Group with that of

Experimental Group; Mann-Whitney Test Ranks

Group N Mean

Rank

Sum of

Ranks

Post-Test – Pre-

Test

Control 23 28.39 653.00


Total 50

33

TABLE 18

Writing Results: Comparison of change in score for Control Group with that of

Experimental Group; Mann-Whitney Test Statistics

Post-Test –

Pre-Test


Wilcoxon W 622.000

Z -1.306




34

Listening

The average listening score for the control group was lower in the Post-Test

(Mdn=14.00) than in the Pre-Test (Mdn=16.00) but this was not a significant

difference, Z=-0.35, ns, r=0.08. (Tables 19, 20, 21)

For the experimental group there was an increase in the listening score from the

Pre-Test (Mdn=16.50) to the Post-Test(Mdn=15.00) but this was not a

significant increase, Z=-0.73, ns, r=0.16. Tables 22, 23, 24)

Comparing the changes in score between the two groups, the average increase

in score in the experimental group (Mdn=-0.50) was not significantly different

to the decrease in score in the control group (Mdn=0.00), Z=-0.73, ns,

r=0.12.(Tables 25, 26).

35

TABLE 19

Listening Results: Control Group Descriptive Statistics

N Mean Std.

Deviation

Minimum Maximum

Pre-Test score 19 16.5263 5.08121 7.00 26.00

Post-Test score 19 16.0000 5.62731 9.00 28.00

36

TABLE 20

Listening Results: Control Group change in score between Pre -Test and Post-


N Mean

Rank

Sum of

Ranks

Post-Test score –

Pre-test

Negative

Ranks

9(a) 10.39 93.50

Positive

Ranks

9(b) 8.61 77.50

Ties 1(c)

Total 19




37

TABLE 21

Listening Results: Control Group change in score between Pre -Test and Post-


Post-Test score –

Pre-Test score

Z -.351(a)




a Based on positive ranks.

38

TABLE 22

Listening Results: Experimental Group Descriptive Statistics

N Mean Std.

Deviation

Minimum Maximum

Pre-Test score 20 15.2000 4.39617 5.00 21.00

Post-Test score 20 16.2500 5.12861 9.00 28.00

39

TABLE 23

Listening Results: Experimental Group change in score between Pre -Test and

Post- Test; Wilcoxon Signed Ranks Test ranks

N Mean

Rank

Sum of

Ranks

Post-Test score –

Pre-Test score

Negative

Ranks

10(a) 7.70 77.00

Positive

Ranks

9(b) 12.56 113.00

Ties 1(c)

Total 20




40

TABLE 24

Listening Results: Experimental Group change in score between Pre -Test and


Post-Test score

– Pre-Test score

Z -.726(a)





41

TABLE 25

Listening Results: Comparison of change in score for Control Group with that

of Experimental Group; Mann-Whitney Test Ranks

Group N Mean

Rank

Sum of

Ranks

Post-Test score –

Pre-Test score

Control

Group

19 18.63 354.00

Experimental

Group

20 21.30 426.00

Total 39

42

TABLE 26

Listening Results: Comparison of change in score for Control Group with that

of Experimental Group; Mann-Whitney Test Statistics

Post-Test

score – Pre-

Test score


Wilcoxon W 354.000

Z -.733


Exact Sig. [2*(1-tailed Sig.)] .478(a)



a Not corrected for ties.

43

Speaking

There was no change in the average score for speaking in the control group

from the Pre-Test (Mdn=3.00) to the Post-Test (Mdn=3.00), Z=-0.05, ns,

r=0.01.

However, there was a significant increase in the speaking score in the

experimental group from the Pre-Test (Mdn=3.00) to the Post-Test

(Mdn=3.00), Z=-2.82, p<0.05, r=0.52.

But comparing the average change between the control group (Mdn=0.00) and

the experimental group (Mdn=5.00) this was not a statistically significant

difference, Z=-1.81, ns, r=0.26.

44

TABLE 27

Speaking Results: Control Group Descriptive Statistics

N Mean Std.

Deviation

Minimum Maximum

Pre-Test Score 19 3.2105 1.34697 1.50 5.50

Post-Test Score 19 3.2105 1.17042 1.50 5.50

45

TABLE 31

Speaking Results: Control Group change in score between Pre -Test and Post-


N Mean

Rank

Sum of

Ranks

Post-Test Score -

Pre-Test Score

Negative

Ranks

4(a) 8.13 32.50

Positive

Ranks

7(b) 4.79 33.50

Ties 8(c)

Total 19




46

TABLE 32

Speaking Results: Control Group change in score between Pre -Test and Post-


Post-Test Score -

Pre-Test Score

Z -.045(a)


Exact Sig. (2-tailed) 1.000



47

TABLE 33

Speaking Results: Experimental Group Descriptive Statistics

N Mean Std.

Deviation

Minimu

m

Maxim

um

Pre-Test Score 29 2.8276 .85853 1.50 5.50

Post-Test Score 29 3.2069 .67503 2.00 4.50

48

TABLE 34

Speaking Results: Experimental Group change in score between Pre -Test and

Post- Test; Wilcoxon Signed Ranks Test Ranks

N Mean

Rank

Sum of

Ranks

Post-Test Score -

Pre-Test Score

Negative

Ranks

4(a) 10.50 42.00

Positive

Ranks

18(b) 11.72 211.00

Ties 7(c)

Total 29




49

TABLE 35

Speaking Results: Experimental Group change in score between Pre -Test and


Post-Test

Score - Pre-

Test Score

Z -2.819(a)





50

TABLE 37

Speaking Results: Comparison of change in score for Control Group with that

of Experimental Group; Mann-Whitney Test Ranks

Group N Mean

Rank

Sum of

Ranks

Post-Test score –

Pre-Test score

Control 19 20.13 382.50


Total 48

51

TABLE 38

Speaking Results: Comparison of change in score for Control Group with that

of Experimental Group; Mann-Whitney Test Statistics

Post-Test score

– Pre-Test score


Wilcoxon W 382.500

Z -1.811




52

One factor that may have affected the results is the choice of tests used to

measure student ability. The changes in language ability over one semester

were represented by only very small changes on the MUET and IELTS scales,

making marking difficult. Many students only improved by half a band or one

band on the MUET and IELTS scales and many did not show any

improvement. Alternative standard tests, that measure lower levels of language

ability and therefore smaller changes, could have provided more conclusive

results.

There are a number of extraneous factors that could have affected the results.

The groups were not balanced according to gender so it could be that the

difference in the results is explained by the fact that the experimental group had

just over 50% male students while the control group had only one male student.

Secondly, the effects observed could be due to the comparative ineffectiveness

of teaching in the contact hours for the control group. To rule out this factor the

study would have to be repeated with a number of different teachers

participating.

The thesis of this study was based on the assumption that Low Proficiency

students have some latent knowledge of vocabulary and grammar. However, in

carrying out this study it was observed that the students in the experimental

group often did not have a sufficient grasp of vocabulary in order to complete

the speaking tasks. Many did not appear to know even some basic high

53

frequency words. This lack of vocabulary was the biggest impediment to them

completing the speaking tasks successfully.

Another factor to be considered is the long-term effects of this experimental

approach. The students who were taught using contact hours for speaking tasks

showed no significant difference in their ability compared to those students who

had been taught in the control group, but this was only over one semester, there

may be longer term effects observed in following semesters. The students in the

control group may not have showed a significant difference in their reading,

writing and listening ability compared with those in the experimental group, but

the first semester may have laid a foundation that would enable them to make

more rapid progress than the experimental group in subsequent semesters.

Equally, for the experimental group, the effects of doing so much speaking may

give them a significant boost of confidence that could make their development

of reading, writing and listening skills in subsequent semesters more rapid than

those in the control group.

In summarising research on communicative classrooms Ellis (2008) points out

that while this approach is effective in developing fluency it can result in a

limited development of linguistic and sociolinguistic proficiency. Although the

approach under investigation might be effective for the first semester, it could

be detrimental to development of accurate syntax if extended beyond this

period.

54

Conclusion

Although the results are inconclusive, they do suggest that the approach of

using all contact hours for speaking is at least as effective as more traditional

approaches. However, various extraneous factors could have affected the results

and the longer-term effects on the students‟ language learning in subsequent

semesters has not been studied.

In order to more accurately assess if this method is a practical solution to

helping Low Proficiency students, future studies should be carried out with a

range of classes and a variety of teachers. Assessment materials more suited to

Low Proficiency students would give a clearer view of the effectiveness of this

approach.

Although inconclusive, the significant improvement in speaking results and the

fact that the students improved equally in other areas as those taught in a more

traditional approach, is enough to suggest that this may be a promising

approach to helping Low Proficiency students that merits further study.

55

References

Ahmad, A. (2011). Language Learning Style Preferences of Low English

Proficiency (LEP) Students in Tertiary Education. Malaysian Journal of ELT

Research , 7 (2), 33-62.

Ellis, R. (2008). The Study of Second Language Acquisition (pp. 252-260,

p827). Oxford: Oxford University Press.

IELTS Partners. (_). IELTS Speaking band descriptors (public version).

Retrieved March 23, 2008, from IELTS:

http://www.ielts.org/PDF/UOBDs_SpeakingFinal.pdf

Malaysian Examinations Council. (2006). Examinations: MUET Regulations,

Test Specifications, Test Format and Sample Questions. Retrieved August 3,

2011, from Malaysian Examinations Council: http://mpm.edu.my

Mathai, E. J. (2008). Assessing Malaysian University Students' English

Vocabulary Knowledge. In W. S. Khanittanan (Ed.), SEALSXIV: papers from

the 14th meeting of the Southeast Asian Linguistics Society (2004), Volume 1

(pp. 219-237). Canberra: Pacific Linguistics.

Razali, N. (1992). ESL in Malaysia: Looking Beyond the Classroom. The

English Teacher , XXI.

56

Thang, S. and Wong, F. (2005). Teaching Styles of Malaysian ESL Instructors:

An Investigation into Current Practices and Implications to English Language

Teaching (ELT). Journal of Language Teaching, Linguistics and Literature ,

10.

Thang, S. and Alias, A. (2006). Investigating readiness for autonomy: A

comparison of Malaysian ESL undergraduates of three public universities.

Reflections on English Language Teaching , 6 (1), 1-18.

enabling malaysian university students to speak english and enjoy

Documents