The Asian ESP Journal

June 2018
Volume 14, Issue 1

Chief Editor - Roger Nunn

Associate Production Editor – David Young

1
Published by ELE Publishing
(Division of the TESOL Asia Group)

TESOL Asia Group is wholly owned by SITE SKILL TRAINING Pty Ltd (Australia)

All rights reserved. No part of this book may be reproduced, stored in a retrieval system or
transmitted in any form or by any means, electronic, mechanical, photocopying or otherwise,
without the prior written permission of ELE Publishing or the Chief Editor of Asian ESP.

No unauthorized photocopying

Publisher: ELE Publishing

Managing Editor of ELE Publishing: Dr. John Adamson

Chief Editor of the Asian ESP Journal: Professor Roger Nunn

Associate Production Editor: David Young

Production Editors: Alliya Anderson; Enaam Hunaini; Joud Jabri Pickett; Ronald Jones; Sarah
Tatz; Tiffany Cammidge

ISSN. 2206-0979

2
 The Application of Wiki-Mediated Collaborative Writing as a Pedagogical
Tool to Promote ESP Learners’ Writing Performance

Masoomeh Estaji
Assistant Professor of TEFL, Allameh Tabataba’i University, Iran

Hoda Salimi
M.A. in TEFL, Allameh Tabataba’i University, Iran

Biodata

Masoomeh Estaji is Assistant Professor of TEFL at Allameh Tabataba’i University. She is

also the head of English Department and an adjunct professor of TEFL at Khatam University.
She is currently the director of Research Institute for Translation Studies at Allameh Tabataba’i
University. She has published numerous books and papers on methodology, testing, and teacher
education. Her research interests include language testing and assessment, teacher education,
and ESP.

Email: [email protected]

Hoda Salimi holds an M.A. in TEFL from Allameh Tabataba’i University. Her areas of interest
are Language Learning and Technology, ESP, and Teacher Education.

Email: [email protected]

Abstract

This study was aimed at investigating the impact of wiki-mediated courses, compared to non-
wiki courses, on the students' writing performance in ESP collaborative writing classes.
Moreover, the advantages and disadvantages of this tool in an ESP collaborative writing class
were explored from the learners’ perspective. To this end, the researcher selected 46
undergraduate students, who were doing an ESP writing course in the university. These
students were in intact groups and were randomly divided into two groups with 25 learners in
the experimental group and 21 students in the control group. The wiki-mediated program was

112
employed for the participants in the experimental group. However, the participants in the
control group enjoyed the conventional, face to face writing course. In order to collect data,
one pretest and two posttests of writing were administered. Subsequently, a semi-structured
interview was carried out with 8 participants. An Analysis of Covariance (ANCOVA) was run
and the results revealed that there was a statistically significant difference between wiki and
non-wiki users considering their writing performance (p<0.05). Furthermore, most of the ESP
students found the wiki a beneficial writing technique with great advantages while some of
them pointed out some disadvantages of the software.

Keywords: English for Specific Purposes; wiki; wiki-mediated collaborative writing; wiki
users; writing performance

1. Introduction
Since today’s world is a socially computer-mediated learning environment, technology has
become a very important means in teaching languages. Besides, technological tools are the
elements that make the English learning environment interactive and intercreative. For
instance, wikis are a set of well-known technological tools, which can help the teachers provide
a creative environment for the students so that they can have collaboration (modifying and re-
working each other’s works), peer assessment (taking advantage of peers’ comments and
feedback), and formative evaluation (providing continuous feedback) of the students (Duffy &
Bruns, 2006). According to Zorko (2009), “a wiki is a web-based publishing tool that offers
learners and teachers an easily editable virtual space for sharing information and knowledge
and it is believed to be particularly suitable for collaborative learning” (p. 645).

On the other hand, in the age of globalization, technology, information, and internet, written
English has become a critical demand for any individual who intends to follow up the vast
change in our modern world (Isa, 2012) since much information and knowledge is shared
through virtual space. However, teaching and learning writing is not an easy task for both the
teachers and learners. Writing practices in ESP classes are limited owing to the time required
to do so and that ESP learners may need more time and support while writing. Writing is also
a critical skill for all university students mainly ESP learners since they are constantly expected
to present their discipline and purpose specific research studies, written assignments, reports,
proposals both for their occupational and academic purposes.

One collaborative writing tool that deserves serious attention is the wiki, which provides a
form of collective action in which a large number of people work independently on a single

113
project. Wikis can provide the ideal collaborative environment by enabling visitors on the
internet to jointly edit content on a web page (Parker & Chao, 2007). Although many students
may be active users of Web 2.0, students do not necessarily have the experience of using wikis
or an understanding of the underlying rationale for using a wiki in an academic setting (Dohn,
2010). Wikis and their applications have been studied mostly in EFL classes; however, the use
of such devices in English for Specific Purposes (ESP) courses, especially the impact of such
devices on promoting the ESP learners’ writing has not been investigated. This study can hence
motivate ESP learners to take an active role in revising their own and their peers' writings
through wikis. Moreover, far too little attention has been paid to the students’ perceptions about
the applications of such devices in the ESP classes.

Despite the fact that wikis enable the students to add and edit texts, it is more important to
understand whether editing their own texts and the other students’ writing during the
collaborative writing process can affect their writing performance. Particularly, the number of
empirical research studies on the effect of applying wikis for promoting the collaborative
writing performance in ESP classes is very limited. In other words, the application of social
and communicative tools in learning academic language, or language skills especially in
writing has been the focus of very few research studies.

The current study is expected to enrich the body of research in the area of wiki application and
ESP academic writing. The results of the study can contribute to the learners' practices of
writing in ESP engineering fields. Moreover, the findings of the study can be a good source of
interest for the university professors and ESP language teachers and researchers to get a better
understanding of how to provide more effective, creative, and novel situation for their students
to practice writing in an academic context. The present study also intends to contribute to the
research studies of wikis in any context which are all to support the existing pedagogy and
promote writing through technology, negotiation of meaning, and student interactions. In fact,
this study adds to the small but growing body of literature (Judd, Kennedy, & Cropper, 2010;
Norton & Hathaway, 2008; Parker & Chao, 2007) by investigating the differences between
wiki-mediated and non-wiki-mediated ESP classes in terms of learners' writing performance.
In addition, the learners' perceptions of the advantages and disadvantages of the software are
explored in this study.

114
2. Literature Review

At the time that technology is penetrating the world of languages, web 2 technology or the
Social Web provides outstanding virtual educational environments. Leuf and Cunningham
(2001) introduced "wikis" as interactive websites that anyone could create, write texts, add or
edit the content without any prior knowledge or skills in editing and publishing. Wikis enable
the learners to develop texts and simultaneously edit other students’ works within a shared
document in a collaborative environment (West & West, 2009). Chu (2009) claims that wikis
are easy and simple to use. Someone who has no technological background can work it out
even with minimal training. Additionally, wikis facilitate collaborative authorship, provide a
free atmosphere in the workplace and allow no dictatorship, make the students perceive the act
of cooperation, enhance efficiency, and allow the interlinking of pages. Wikis also offer an
open editing system that makes everyone edit the page. This way, the students would be able
to see the cooperation in the digital environment and find the barriers of communication.
Moreover, the interlinking of the pages promotes knowledge sharing among users.

Collaboration is also one of the applicable methods in teaching a foreign language. Lee (2000)
and Warschauer and Healey (1998) assert that Computer Assisted Language Learning (CALL)
technologies provide the situation for the students to work collaboratively in pairs or groups,
so they can exchange their feedback and work in a team environment. This way the students
can also enhance their achievements, have access to authentic materials, use multiple web
sources in their works, and interact better in a collaborative environment. Similarly, Haring-
Smith (1994) introduced technology as the main source that could help collaborative learning
and collaborative writing. He described collaborative writing as a situation which “involves
more than one person who contributes to the creation of a text and makes sharing responsibility
more essential" (as cited in Suwantarathip & Wichadee, 2014, p. 149). In other words,
collaborative writing is a distinct process in which the participants work together and interact
in order to generate the ideas, set the structure of the text, then edit and revise it (Storch, 2011).

In recent years, there has been an increasing number of studies on collaborative writing. For
instance, Kessler, Bikowski, and Boggs (2012) carried out a study on the use of collaborative
writing in Academic Web-Based Projects. The findings of their study revealed that the use of
the discussion boards, online chats, and email communication through wikis assisted the
students in their writing when they most needed it. In 2014, Aydin and Yildiz also published a
study in which they described the role of three various meaning-focused wiki-based

115
collaborative writing tasks (argumentative, informative, and decision-making) in form and
meaning related changes and self and peer corrections. Moreover, the students’ perceptions of
using wikis in collaborative classes were explored. The results of the study showed that tasks
improved the EFL learners’ collaborative interaction in the wiki environment. However, the
argumentative tasks led to the improvement of peer corrections whereas informative tasks
improved self-corrections more. Further, writing in a collaborative environment resulted in the
learners’ concentration on the meaning rather than the form. Overall, writing through wikis
promoted the learners’ writing and they found it a positive experience. Likewise, Warschauer
(2010) elaborated on the positive impacts of the wikis on writing essays. In a study on “Tools
for Teaching Writing”, he suggested that the wiki increased the quantity of writing owing to
the ease of writing and publishing on blogs which make them an attractive means to the
students (Fellner & Apple, 2006). This software also developed confidence in the skill and
made writing more motivating to the students. Similarly, Alshumaimeri (2011) demonstrated
that the use of wikis significantly promoted the students’ writing in terms of both accuracy
(lexico-grammatical accuracy e.g. spelling, word choice, word order, punctuation), and quality
(organization, elaboration, coverage, clarity, links, and intent).

As to the learners’ perception of wikis in collaborative writing classes, Chao and Lo (2011)
found students were satisfied with wiki-based collaborative writing and that this environment
helped them write and fulfill collaborative writing tasks in a less limited time. In addition,
Marzec-Stawiarska (2015) studied the reactions of some of the students about writing through
wikis in a research entitled "Wikis and new perspectives for collaborative writing". From the
students’ point of view, writing on the wiki makes them think about the reader more. Besides,
they thought that writing on the wiki makes them use more diversified vocabularies as well as
grammatical structures. It also enables them to think about the content more while using the
wikis and they become more careful about the accuracy of their writing.

Despite the advantages found in the studies of wikis, some disadvantages, regarding the uses
of the wikis, have been also mentioned. According to Klobas and Beesley (2006), primarily,
wikis have a changing nature that makes them not be able to be published on the live webs.
Likewise, the combination of management and technology in this social software makes them
unable to effectively manage as a source of information. In addition, Adekunle and Olla (2015)
mentioned some of the other disadvantages of the wikis. They mentioned that wikis are open
systems but in some cases, this feature might not be considered as a very valuable characteristic
especially in the projects that are confidential; therefore, this software would be open to SPAM

116
and vandalism if the users do not manage it properly. According to these two scholars, we
always require an internet connection to get access to the wikis, so we can see that the print
versions of the practicing materials would be more applicable then easier for the students to
use. Congruent to these detriments, flexibility of using the wikis would make their information
disorganized (as cited in Tella, 2015).

All in all, most of the research studies on the wikis have been done in EFL classes and ESP
courses (developed based on needs analysis and designed for specific disciplines) have rarely
been taken into account. Among the existing research studies in this domain, we can refer to
Asztalos (2014) who evaluated the use of the wikis at a college in Hungary to figure out
whether or not this tool would be applicable. The results of the study showed the students have
positive perceptions of the wikis in English for Business and Finance purposes. It should be
noted that they had this perception for their general English skills in several areas and did not
solely focus on the skill of writing. In another major study, Vaičiūnienė and Užpalienė (2010)
demonstrated that integrating the wikis into teaching in ESP writing courses can help the
learners have access to authentic materials, become autonomous, and be responsible for their
own learning. However, this study proved that course books and traditional methods of
language teaching are still preferred as the first choices by the students. Further,
Kavaliauskiene (2010) compared the effects of Web Logging as well as the wikis on the
students’ writing performance in ESP classes for university students studying Social Policy.
According to their findings, the students did not find the wikis useful to improve their writing
in ESP courses.

Hence, more research is required to gain a better understanding of how wikis can influence the
process of collaborative writing in ESP courses. To this end, this study particularly intended
to address the following questions.

1. Are there any statistically significant differences between wiki and non-wikis users’
writing performance in ESP collaborative writing classes?
2. What are some of the advantages and disadvantages of using wikis for ESP
collaborative writing classes from learners' perspectives?

117
3. Research Methodology

3.1. Participants and research setting

Prior to the selection of the participants, primarily, the permission and consent of the
participants and authorities were obtained. In order to investigate the impact of applying wikis
in ESP classrooms, 60 ESP students from Iran University of Science and Technology in Tehran
were selected to participate in this study. After homogenizing the participants through an
IELTS Academic writing task 2, 46 female and male students, aged between 20 and 28 and
studying computer and mechanical engineering, were selected. These participants were in
intact groups who were randomly assigned to two groups of experimental and control with 25
participants in the first group and 21 participants in the second one. It is significant to mention
that the participants in both the experimental and control groups did 4 writings as their
assignments until the end of the course. Half of the essays were on specific topics such as
"Mainframe Computers, Minicomputers, Microcomputers as well as Computer Capabilities
and Limitations" and the other half was dedicated to Academic topics that were related to
computer and technology such as "Distance Education, Internet Advertising, and the Role of
Internet in Communication". It is worthy of note that in the control group, the participants did
not receive any treatment and the whole course was passed through conventional instruction.
Moreover, after getting the participants’ consent for further cooperation, 8 participants were
picked through convenience sampling to be interviewed in order to find out their perceptions
about the uses and efficiency of the wikis in ESP classes. Convenience sampling is a form of
non- random sampling method used for “the selection of individuals who happen to be
available for study” (Mackey & Gass, 2005, p. 122).

3.2. Materials and instruments

The researcher designed the lesson plans and established the wiki environment for six weeks
of treatment. Google Groups was the main wiki used in the research, and different groups of
four were created on Google for the experimental group by the researcher. Further, in order to
avoid any confusion or misunderstanding, the researcher taught the students how to work with
their wiki pages and post their comments and their writings. Subsequently, the students were
provided with their lesson plans as well as the dates they were supposed to post their writings
on their pages. The researcher explained that the main responsibility of the students was to
comment on each other's texts and give each other feedback. Figure 1 below shows a sample
of a wiki page.

118
Figure 1: A Sample Wiki Page Designed for the Wiki Class
The wiki pages contain two parts ―the upper section‖ and ―the lower section. The upper
section shows the author of the text, his/her name, the detailed information about the author,
and complete report about the changes in form and content that the author made. The second
section, which is the lower section, is the content part that includes the text (Mak & Coniam,
2008).

In order to assess the students’ writing performance, IELTS Academic Writing test task 2
(Cambridge ESOL, 2005) was used so as to collect the quantitative data. This test was chosen
for two reasons: Firstly, it is one of the most formal and popular methods of assessing writing
skills. Secondly, accomplishing this task requires using more words and idioms in a formal
essay; therefore, it leaves more room for making mistakes and providing feedback which can
be considered a suitable technique to evaluate students’ writing skill (Ganji, 2009). In this task,
learners are required to write an essay of 250 words in 40 minutes. At the beginning of the
course, an academic IELTS test of writing was taken to check the learners’ writing
performance at the entry level and homogenize them based on their writing scores. Likewise,
at the end of the course, the students were provided with two tests of writing; one of which
was the one administered as the pretest then another test of writing was used to ensure the level

119
of difficulty of the writing tests and examine whether the students could change their overall
writing performance. Two certified language examiners did the scoring to provide a valid
evaluation of the tests. The two examiners were also provided with a scoring rubric published
by the British Council and the University of Cambridge (ESOL Examination) in order to score
the test in a standard way. To do so, Pearson product moment correlation coefficient was run
and the results indicated that there was on the whole, high agreement between the raters
because the great majority of the indices were above 7.

In the qualitative phase of the study, a semi-structured interview was run with 8 participants,
picked through convenience sampling, to determine the advantages and disadvantages of using
the wikis in learning second language writing. It is important to note that the interview items
were designed by the researcher (see Appendix A) and three language experts reviewed them
to check their accuracy and appropriacy. To ensure consistency in the ratings (in terms of
coding of the transcripts), two raters were employed and inter-rater reliability index of 0.85
was obtained using Cohen's Kappa. Each interview lasted about 15-20 minutes and the
interviewee's responses were recorded by a Digital Voice Recorder (DVR) in order to avoid
the loss of information. Finally, all the recordings were transcribed, and then coded to be
analyzed.

3.3. Data collection procedure

The steps taken to collect the data included pilot-testing, homogenizing the participants,
administrating the writing tests as the pretest and posttests, conducting the treatment sessions,
and running a semi-structured interview. After ensuring the reliability of the research
instruments and homogenizing the participants, the researcher took a sample of IELTS
academic writing test task to examine both the control and experimental group's writing
performance. After pretest, the treatment sessions were conducted to the participants in the
experimental group. In order to achieve the objective of the study, the students were provided
with orientation sessions about wiki and how it assists them in accomplishing their writing
activities. Once the training session was over, each class was divided into four groups of five.
Then the students were given a weekly writing program. According to the program, the learners
were required to write an essay on specific subjects every week as an assignment. In the
following session, the researcher corrected all the writings and commented on the students'
errors, providing direct and indirect feedback on the areas that needed to be corrected as shown
in Appendix B.

120
Subsequently, all the writings were presented on the wiki pages one by one so that the students
of each group could view them and comment on each other's essays, modifying their own texts
as well as their peers’ writings. In particular, the students were engaged in five stages of the
writing process namely pre-writing stage for collaborative planning, drafting, peer-reviewing,
peer proofreading and editing (giving comments, editing, or creating a text together), and
individual revising and publishing. The researcher encouraged the students to return to the
history page of the wiki and review the points mentioned by the researcher and their peers. It
was also emphasized that the information on the history page could be used as a reference for
their subsequent writing. This process continued for one month until the students provided four
writings throughout the sessions.

It is worth mentioning that there was no training on the wiki for the participants in the control
group. The participants in the control group were required to take just a pretest and posttest
and they did not experience any wiki collaborative writing course. However, they were taught
through conventional methods of writing. In that conventional and face-to-face writing course,
the students were required to write four essays during the course like the practices assigned for
the participants in the experimental group. The students were provided with some writing
samples and models on the board and paper. They practiced the samples in the class and were
monitored by the teacher. Similarly, they were provided with explicit feedback from their
teacher. In particular, through mostly written feedback, the instructor wanted them to seriously
focus on all aspects of their writing, including the vocabulary, grammar, organization, content,
and mechanics. Their writings were all corrected and scored in the next session and direct
feedback was provided on the mentioned aspects. Their writing scores made them more
motivated to write better and ameliorate their writing skills (such as developing better topic
sentences, paragraphing, summarizing, evaluating, and recognizing irrelevant information).
They were also asked to rewrite their essays using the feedback they had received and hand in
their revised essay to the instructor to see if the changes were implemented.

At the end, the same writing test, which was administered at the pretest stage to the students,
was re-administered to compare the groups’ score gains and figure out whether or not their
writing performance improved. The next week, another post-test was run, in which the students
were asked to write an essay about a subject that was similarly selected from the IELTS
academic writing test task 2 (Cambridge ESOL, 2005). In addition, a semi-structured interview
was conducted, whose results paved the way to provide precise answers to the qualitative
question of the study.

121
 3.4. Data analysis

This study used a mixed methods research design, more precisely “a sequential QUAN-QUAL
type of mixed methods, denoting the sequence and dominance of the method constituents”
(Dornyei, 2007, p. 169). First the quantitative data were collected, then after analyzing the
results; the qualitative data were collected and analyzed. The data obtained from the pretest
and posttests were analyzed using descriptive statistics and Analysis of Covariance
(ANCOVA) by means of the Statistical Packages for the Social Sciences (SPSS), version 21
(IBM Corp, 2012). Prior to running the ANCOVA, the normality assumptions of the tests were
examined. Since two raters had scored these writings, the Pearson product moment correlation
coefficient was used to measure the inter-rater reliability to determine the extent to which the
raters were in agreement. Furthermore, in order to explore the merits and demerits of using the
wiki tool in the writing instruction, the researcher tabulated the learners’ responses from the
semi-structured interview, and provided the frequency and percentage through frequency count
descriptive statistics. The methods used to analyze the qualitative data were Content and
Thematic Analysis. According to Riazi (2016), content analysis is used by researchers to
analyze concepts and written and spoken texts.

4. Results
4.1. The impacts of the wikis on ESP students' writing performance

In order to test the null hypothesis to do with the first research question, correlation analysis
was employed. Since all the writing scores of the participants in both groups were rated by two
raters, all these ratings in both groups in pretest and posttests, even the writings of the
experimental group in the treatment sessions, were correlated to check inter-rater reliability.
All these correlation coefficients indicated that there was, on the whole, high agreement
between the raters because the great majority of them were above 7. For rating scales, “an
interrater reliability coefficient of 0.70 or higher is considered acceptable” (Ary, Jacobs,
Sorensen, & Walker, 2014, p. 233). Later, the average scores of the two raters were computed
and included in the main analysis.

After ensuring inter-rater reliability, the descriptive statistics of the two groups in terms of their
writing scores were computed (Tables 1 and 2). It should be noted that the experimental group
also had 5 sessions of treatment (one for getting a good orientation to wikis and 4 sessions for
practicing writing through wikis), whose descriptive statistics were also computed.

122
Table 1: Descriptive Statistics (control)
N Mean Std. Skewness Kurtosis
Deviation
Statistic Statistic Statistic Statistic Std. Statistic Std.
Error Error
Pretest 21 4.97 1.21 -.32 .50 -.30 .97
Posttest 21 5.15 1.74 -.56 .50 -.74 .97
Valid N
21
(listwise)

Table 2: Descriptive Statistics (experimental)

N Mean Std. Skewness Kurtosis
Deviation
Statistic Statistic Statistic Statistic Std. Statistic Std.
Error Error
Pretest 25 4.65 1.11 -.11 .47 -.96 .91
Session1 25 5.53 1.43 -.66 .47 .64 .91
Session2 25 5.46 1.50 -1.34 .47 2.15 .91
Session3 25 5.44 1.49 -1.09 .47 1.79 .91
Session4 25 5.79 1.35 -.72 .47 -.18 .91
Session5 25 6.20 1.17 -.13 .47 -.66 .91
Posttest 25 6.05 1.40 -.48 .47 -.16 .91
Valid N
25
(listwise)

The findings represent the writing improvement trend for the experimental group across the
pretest, treatment sessions, and the posttest. This trend indicates improvement in writing;
however, in order to see whether this improvement is significant compared with that of the
control group, the posttest of the two groups were compared via Analysis of Covariance
(ANCOVA). The reason for running ANCOVA was that, it was not clear whether the groups
were equal in terms of mean writing scores on the pretest, so the probable pretest differences
between the groups were taken into account by employing analysis of covariance (i.e. covariate
effect).

In order to run ANCOVA, several assumptions need to be checked initially. The first of these
is the assumption of normality, which was found met based on the skewness and kurtosis ratios,
which were not beyond - 1.96 (indicating normality). The next assumption is to do with the
homogeneity of variances, which was found met according to the results of Levene’s test in
Table 3 (p>.05).

123
Table 3: Levene's Test of Equality of Error Variancesa
Dependent Variable: posttest
F df1 df2 Sig.
.18 1 43 .66
Tests the null hypothesis that the error variance of the dependent variable is equal across
groups.
a. Design: Intercept + pretest + group

Table 4 below demonstrates the results of the check on the assumption of homogeneity of
regression slopes and the main ANCOVA results. The first row indicates that the assumption
of homogeneity of regression slopes is met; Interaction F(1,44) = .47, p>.05. The second row
shows that the groups were indeed different on the pretest; Pretest effect F(1,44) = 11.37,
p<.05, and finally the third row demonstrates that the groups are significantly different on the
posttest; Group F(1,44) = 6.64, p<.05, eta squared = .13 small to medium effect size.

Table 4: ANCOVA Results

Dependent Variable: posttest
Source Df Mean Square F Sig. Partial Eta
Squared
group * pretest
1 .96 .47 .49
(interaction)
Pretest (covariate) 1 22.63 11.37 .00 .21
Groups
1 13.21 6.64 .01 .13
comparison
Error 42 1.989
Total 45
Corrected Total 44
a. R Squared = .275 (Adjusted R Squared = .240)

In other words, there are statistically significant differences between wiki and non-wikis users’
writing performance in an ESP collaborative writing class. Specifically, the wiki-based group
outperformed the control group.

4.2.The merits and demerits of the wikis

Regarding the advantages and disadvantages of using the wikis in ESP collaborative writing
classes, generally, all of the students agreed that the experience of writing ESP essays using
the wiki is useful and enjoyable; none of them believed that the wikis were not appropriate.
They also believed online collaboration helped them to improve their writing skills.

124
The advantages of the Wikis mentioned by the students in the experimental group are presented
in the following extracts. The first excerpt is taken from one of the participants' transcriptions.

Participant 7: The first time I started to write by using the wiki, I found it
very difficult to do so, but after one or two writings; it helped me a lot in
improving my sentences.

According to another interviewee:

Participant 2: It was a very good experience for me to use the wiki in order
to learn how to write in English. This way, I feel that I could be able to
learn how to write; I got on the track and learned a lot of things.

Regarding another advantage of the wiki in ESP classrooms, learning the grammatical skills
was also mentioned. Below is an excerpt from one of the respondent's transcriptions.

Participant 5: Wikis corrected my grammatical mistakes and helped me

and other students to learn the parts of the speech, "Nouns", "Verbs", and
"Adjectives". Wikis helped us learn the "Structure of the sentences" and it
also aided us to know our mistakes and correct them.

As another element which is considered as a positive point about the wikis in ESP classrooms,
we could refer to cost-effectiveness. According to one of the participants:

Participant 8: Wikis were very useful for the students who lived far away
from the teacher and the classmates, so he/she did not need to pay a lot to
just go to the class or university, Wikis helped them save and learn a very
good way.

Among other assets of the wiki in ESP classrooms, time-effectiveness and accessibility are
cases in point. Below you can find the excerpts from the interviewees concerning the above
mentioned points.

Participant 2: As far as Internet is available everywhere now-a-days, we

can access it wherever we are willing to and I think this is one of the most
important positive points about Wikis.

Another interviewee was concerned about the time:

125
 Participant 3: I think that the wikis give us more time to think about our
writings and write them in an academic way.

Among other profits, one could refer to the use of wikis as an element which makes the writing
process easier for the students. The following excerpt is taken from one of the participants’
transcriptions:

Participant 6: After writing one or two writings, we found that how much
this software could be useful for us to learn how to write.

Increasing the level of creativity is another advantage mentioned by the students. Below a part
of a student's transcription is provided.

Participant 1: Wikis let us find various models in academic writing. By

following such models, we learn a lot then step by step we could be able to
create our novel way to write academic articles in our own words.

On the other hand, decreasing the level of anxiety is considered as another advantage by some
of the students. The following is an extract from a student's transcription:

Participant 5: The first time I was about to start a writing, I got very anxious
but after experiencing the wiki once or twice, I could control myself better
and then become more confident in learning how to write.

Another asset mentioned by the students is collaboration among the peers and the teacher that
helped them a lot in learning how to write. The following is taken out of a student's
transcription.

Participant 7: This collaboration was so much helpful in learning writing,

because it let me know what my mistakes are then I could compare them
with other students' texts and see what the differences are between us. This
way, I could learn how to write more correctly.

126
Figure 2 below sheds light on the advantages of Wikis in ESP classrooms from interviewees'
perspectives. by percentage.

Figure 2: Advantages of Wikis in ESP Classrooms

As shown in figure 2, accessibility, improving writing, learning the grammatical points, and
collaboration are the core advantages of the wiki which have been reiterated by 100% of the
interviewees. In addition, increasing creativity and helping in learning how to make new
sentences are the second most important components mentioned by 88% of the interviewees.
Likewise, making writing easy and cost-effectiveness were mentioned by 63% of the
interviewees. Finally, half of the participants considered decreasing anxiety, time-
effectiveness, and donating more time as the most significant advantages of the wikis.

The second part of the question focused on the factors which manifest the disadvantages of the
wiki in ESP classrooms. Table 5 below gives us a classification and percentage of these
disadvantages as mentioned by the interviewees.

127
Table 5: Disadvantages of Wikis
Disadvantages of using Wikis in ESP classrooms Percentages

1. It cannot be used instead of real and face to face interaction. 50%

2. The students could copy the writings from the Internet.
38%
3. It cannot help us like writing on the paper.
4. It makes me addicted to the computer.
5. It makes me lose my self-confidence. 38%

6. It does not have any disadvantages. 13%

7. Typing is a very difficult task. 13%

13%
13%

As shown in table 5, the most important disadvantages of wikis in ESP classrooms, as

mentioned by the interviewees (50%), was the point that wikis cannot take the place of real
and face to face interaction. Likewise, some other factors, like getting the possibility to cheat
and copy the texts from the internet as well as not being efficient for some students, were
mentioned by 38% of the interviewees. In addition, not having any disadvantages, making
students addicted to the computer, and making the students lose their self-confidence were
equally pointed out by 13% of the participants. The following excerpt is on the most important
factor, manifesting the most significant disadvantage of using the wikis.

Participant 4: I think that wikis cannot take the place of face to face and
real interaction. Writing on the paper helps me more in learning how to
write in English and I feel more comfortable with that.

5. Discussion
Based on the analysis of the results, using ANCOVA, it was revealed that there are statistically
significant differences between wiki and non-wikis users’ writing performance specifically,
the wiki-based group outperformed the control group. That is to say, the null hypothesis having
to do with the first research question was rejected. This finding is consistent with those
suggesting wikis as a perfect means to promote students’ writing performance (Kennedy, 2010;
Lin, 2005; Xiao & Lucking, 2008) and various aspects of writing such as writing accuracy
(Kuteeva, 2011) and students’ writing quality (Alshalan, 2010; Miyazoe & Anderson, 2009).
This improvement in writing scores can be as a result of writing in a collaborative environment
since through wikis the participants can freely discuss and exchange various notions, share

128
works, and provide feedback to other members (Reo, 2006). This finding is also in resonance
with that of Warschauer (2010), revealing that the wiki increases the quality of writing,
develops confidence in the writing skill, and makes writing more motivating. Neumann and
Hood (2009) observed that the wiki could assist the students in the writing process by
facilitating interaction among the students and engaging them in the classroom. Further,
Cesteros and Lacorte (2014) reported the wiki helped the learners consolidate their writing
fluency, motivating them to think of the audience more and produce a more qualified learning.

Such a significant change was really expected since the instructor/researchers provided the
students with written feedback on their writings and posted their comments on the learners’
wiki pages. Therefore, the students could easily find out their errors in writing an academic
text and enhance their writing performance. One important point is that, at the beginning, their
errors were mostly grammatical and there was no trace of cohesion and coherence in their
texts; thus, these areas were the most noteworthy parts that the instructor/researchers worked
on. As time went by, the number of grammatical mistakes decreased and their texts turned to
become more cohesive and coherent. Some of the students posted a comment and mentioned
that by having the wiki class, they learned how to generate an English academic text correctly;
they also stated that they loved this method and hoped to continue it again in the future (see
Appendix C). By the same token, in the quantitative phase of the study, a statistically
significant difference was found between wiki and non-wikis users’ overall writing
performance.

Concerning the qualitative results and research question on the advantages and disadvantages
of Wikis in ESP classrooms, some elements such as accessibility, improving writing, learning
the grammatical skills, as well as collaboration were found as the most important advantages
of the wikis in ESP classrooms. Pursuant to a study done by Chu (2009), the importance of the
factor of collaboration is supported. Chu (2009) demonstrated that the wiki facilitates
collaborative authorship, provides a free atmosphere in the class, and allows no dictatorship.
Furthermore, Shelly and Vermaat (2011) pointed out that the wiki provides secure access for
the selected group of qualified members to read the entries and write about them. Adekunle
and Olla (2015) also added that wikis provide accuracy and depth to the students; thus, these
tools have the capability to identify and correct more errors than the time they write on their
own (as cited in Tella, 2015).

129
More significantly, the results of the study are in accord with Wei, Maust, Barrick, Cuddihy,
and Spyridakis (2005), who concluded that the most significant advantage of the wiki is
providing the situation for the students to produce co-authored writings; creating a social
environment, sharing, and exchanging information. Another advantage of the wikis is that it
can be live and provide a shared space where all team members can have writing and editing
privileges. Thus, this feasibility of collaboration in a wiki can make it a powerful tool for
project management and collaborative writing. Koschmann (1999) claimed that through
collaboration, which requires social interaction between team members and dialogic processes,
the text is produced through.

The findings of this study revealed that the wiki assisted the students in learning how to form
new sentences. This finding is in compliance with the study of Marzec-Stawiarska (2015), who
observed that after applying the wiki, the students could add to their ideas, edit the texts,
amend, and expand them. They concluded that the students became more creative at the end
of the project and could write out in the way they liked.

Cost-effectiveness was mentioned as the other advantage of the wiki in ESP classrooms. This
finding is in line with the research studies conducted by Hunsinger and Senft (2014). They
suggested that creating a website is much easier and less expensive than creating a printed
pamphlet. In addition, Biella, Luther, and Pilz (2004) asserted that the wiki provides the
situation for the students to have access to less expensive materials, than borrowing books from
the electronic libraries or buying books on the net which is in resonance with the finding of the
current study. The last two advantages are time-effectiveness of the wikis and the role of such
software in reducing the anxiety of the students. Hereupon, Reio Jr and Hill-Grey (2012) found
that wikis can be useful for the learners who study English online. Further, they discussed that
time could be saved in the learning process by the facilities provided by the wikis. Regarding
the students’ anxiety, Aitken (2014) proposed that technologies such as the wiki and NiCenet
could help students reduce their anxiety because they can do their homeworks and submit them
even from their homes. The learners also believed that since they get the teachers and peers’
feedback online, they get more confidence in using the writings so that their level of anxiety
decreases.

This study also found some disadvantages in the use of the wikis. The first one to indicate is
that this tool cannot be used instead of the face to face and real interaction. In fact, none of the
studies done so far are in resonance with this finding since there are many researches whose

130
results manifest that the wikis provide face-to-face and real interaction, so the students do not
feel the need to have the traditional and real classrooms (Castaneda Vise, 2007; Mesquita,
2012). The second disadvantage that was indicated by the students was that the students can
easily copy their materials from the internet by using the wiki; therefore, the validity of the
produced materials is not ensured.

This finding is consistent with the outcome of a study done by Wheeler (2010), who recognized
that when the students use the wikis, they tend to copy and paste the texts from the internet and
Wikipedia then they put them on the wiki page (Booth, 2008; Murugesan, 2010). Besides, a
few participants emphasized that the paper-based writing could make them more successful in
writing than the wiki-based writings. This finding is in strong and sharp contrast with the
findings of other related studies. On the contrary, it has been evinced that the information
displayed in the wiki style is much more persistent than the paper-based version (Jesson &
Peacock, 2012; McKay, 2014). Regarding the last two disadvantages that the wiki makes the
students addicted to the computer and the students' difficulty in typing, no related study was
found.

Overall, the quantitative data analysis showed that applying the wikis has had a high positive
impact on the learners’ writing performance. Further, the students highlighted some of the
advantages and disadvantages of using the wikis in learning how to write an essay. All in all,
the findings revealed the beneficial effects of the wikis on the students’ writing performance
and perceptions, which were also discussed considering the previous studies done in this area.

6. Conclusions and Implications

The first question of this study investigated the differences between the experimental and
control group in terms of their writing performance. After analyzing the data, the results of
ANCOVA test evinced that the students’ writing performance on the posttest was considerably
higher than the pretest and the results of the study displayed a statistically significant difference
between the wiki-oriented courses, compared to the one that did not apply such software, and
the students' writing performance in the ESP class (p<.05). Thus, the null hypothesis
formulated based on this research question was rejected.

Regarding the qualitative phase of the study, it should be noted that the learners valued the
wiki and believed that it offered them great advantages to improve their writing skills. Almost

131
all the students mentioned that the wiki helped them learn the grammatical structures. Further,
they became far more collaborative than before since the wiki was so accessible that they found
the opportunity to write anywhere at any time. According to Duffy and Bruns (2006), the wiki
provides more time and accessibility to the students; therefore, it offers the students easier
ways to work collaboratively and consolidate their writing abilities. It is interesting to note that
they called the wiki "a careful teacher at home". Moreover, the vast majority of the learners
claimed that the wiki assisted them in learning novel idioms and expressions; it also helped
them learn how to construct new sentences. Additionally, they believed that the wiki made
more creative in writing in English. Besides, they found that the wiki was cost-effective and
made the writing process easier for them. Finally, the learners declared that the wikis helped
them decrease their anxiety because it provided them with more time and a situation, which
made them more motivated to write essays.

However, they agreed that there were some disadvantages in using the wikis. Half of the
learners believed that the wikis could not be replaced by face to face and real interaction. They
also claimed that they are more used to traditional methods and think that the main method of
teaching should revolve around the traditional ones. Further, they approbated that the wiki
environment was not very safe; it enabled them to cheat and copy some texts from internet.
They also believed that although the wiki has a lot of advantages, it could not help them
improve like writing on a paper. They were on the opinion that typing on the computer was a
difficult task that made learning through computers much harder for them. Interestingly, they
expressed that such software made them addicted to the computer. Whenever they were about
to write an essay, they felt dependent on the computer; if a computer was not there, they felt
they would not be able to write anything; therefore it lessened their self-confidence. In general,
the learners regarded the wiki as a beneficial and practical method for promoting the learners’
writing whose advantages surpassed its disadvantages.

This study contributes to writing instruction. For instance, by using wikis, the teacher and
students can discuss the written text together, exchange comments over their writings, and
follow a more interactive writing process. In other words, the teachers would be able to find
novel ways to provide feedback and pass their comments to the learners. More importantly,
they would get to know how to apply this software in their instruction so as to cover the
learners’ needs properly. Another practical implication of the findings is that it is profitable for
the learners to learn a novel method using technology such as the wiki, to be applied for writing
essays in their specified field. This technological tool would also help them improve the quality

132
of their writings by learning to use new idioms and expressions as well as many grammatical
structures.

The findings of the study also provide the syllabus designers and materials developers the
opportunity to design new syllabuses and develop electronic materials to be used in the
classroom. Further, syllabus designers and materials developers can design some tasks and
activities that should specifically be done through the wikis. Another area in which the wiki
tool can be efficiently used is language testing, particularly informal language testing. Teachers
can make maximum use of the wiki in self and peer assessment of the students' writings, since
this software provides them with a history page as a checklist in order to guide the learners and
their peers to understand their language use and problems better.
Although the findings of the study revealed that writing in a collaborative environment using
the wikis is a fruitful pedagogical technique, the current study suffered from some limitations
and delimitations. Since the participants were informed about the purpose of the study and the
procedure, they could have altered their behavior in response to their awareness of being
observed. However, in order to overcome the Hawthorne effect, the researchers and instructors
constantly monitored the students’ performance and pursued a rather long-term approach.
Moreover, since the data collection and analysis for the writing skill are very demanding and
tough and the raters do not cooperate mostly when you employ an analytic scale, focusing on
writing quality, accuracy, and meaning-focused writing skills, the researchers just focused on
the overall writing performance of the ESP learners.

Last but not least, every research presents some other aspects and areas that have not been
examined yet. Research on wiki-mediated language program and its roles in ESP essay writing
is a novel subject that has not been extensively studied by the language specialists. Ergo, other
fields of studies, such as humanities and medicine along with various levels of proficiency are
required to be investigated to figure out whether or not the wiki would be considered as
beneficial in those areas. In addition, future studies can concentrate on investigating the
differences between learners of various university fields of study using the wikis, in terms of
the quality of essay writing. It would be great if future researches examine the effects of this
means on various writing features such as lexical use, accuracy, complexity, and density.

Further, examining the effect of individual differences, such as personality factors, different
learning styles, and strategies, on the use of the wiki in writing classes would be interesting to
be considered in future research studies. Another area of research is to focus on the effect of

133
using the wikis on other language skills, such as learners’ speaking, reading, and listening
skills. Finally, a very interesting avenue for research would be the analysis of cyber discourse
through the feedback and comments posted on the wiki. Reflective-thinking and critical
thinking are also very significant variables to be considered in such research studies. Overall,
the applications of wikis should be extensively inspected in various disciplines, mainly ESP
writing courses.

References

Adekunle, P. A., & Olla, G. O. (2015). Social media application and the library: An expository
discourse. In A. Tella (Ed.), Social Media Strategies for Dynamic Library Service
Development (pp. 41-71). Hershey, PA: IGI Global.

Aitken, J. E. (Ed.). (2014). Cases on communication technology for second language

acquisition and cultural learning. Hershey, PA: IGI Global.

Alshalan, R. (2010). The effects of wikis and process writing on the performance of Saudi
female EFL secondary students in writing. King Saud University, Riyadh.

Alshumaimeri, Y. (2011). The effects of wikis on foreign language students writing

performance. Procedia - Social and Behavioral Sciences, 28, 755-763.
http://dx.doi.org/10.1016/j.sbspro.2011.11.139

Ary, D., Jacobs, L. C., Sorensen, C., & Walker, D. A. (2014). Introduction to research in
education (9th ed.). Wadsworth: Cengage Learning.

Asztalos, R. (2014). The use of a wiki at a college in Hungary as a tool to enhance personal
learning. In S. Jager, L. Bradley, E. J. Meima, & S. Thouësney (Eds.), CALL Design:
Principles and Practice (pp. 18-22). Proceedings of the 2014 EUROCALL Conference,
Groningen, The Netherlands.

Aydin, Z., & Yildiz, S. (2014). Using wikis to promote collaborative EFL writing. Language
Learning & Technology, 18(1), 160-180.

Biella, D., Luther, W., & Pilz, T. (2004). A web-based system for assisted literature research.
Proceedings of the 3rd European Conference on E-Learning, Academic Conferences
Ltd., Reading, UK, 15-24.

134
Booth, D. (2008). It's critical!: Classroom strategies for promoting critical and creative
comprehension. Markham: Pembroke Publishers Ltd.

Cambridge ESOL (2005). Cambridge practice tests for IELTS 4. Student's book with answers:
Examination papers from the University of Cambridge Local Examinations Syndicate.
Cambridge: Cambridge University Press.

Castaneda Vise, D. A. (2007). The effects of wiki- and blog-technologies on the students'
performance when learning the preterite and imperfect aspects in Spanish.
Unpublished doctoral dissertation, West Virginia University-Morgantown.

Cesteros, S. P. & Lacorte, M. (2014). Teacher education. In M. Lacorte (Ed.), The Routledge
Handbook of Hispanic Applied Linguistics (pp. 117-133). New York: Routledge.

Chao, Y. C. J., & Lo, H. C. (2011). Students’ perceptions of wiki-based collaborative writing
for learners of English as a foreign language. Interactive Learning Environments, 19(4),
395-411.

Chu, S. (2009). Using wikis in academic libraries. Journal of Academic Librarianship, 35(2),
170-176.

Dohn, N. B. (2010). Teaching with Wikis and blogs: Potentials and pitfalls. Proceedings of the
7th International Conference on Networked Learning, Aalborg, Denmark, 142-150.

Dornyei, Z. (2007). Research methods in applied linguistics: Quantitative, qualitative, and

mixed methodologies. Oxford: Oxford University Press.

Duffy, P., & Bruns, A. (2006, September). The use of blogs, wikis, and RSS in education: A
conversation of possibilities. Paper presented at the Online Learning and Teaching
Conference, Brisbane, Australia.

Fellner, T., & Apple, M. (2006). Developing writing fluency and lexical complexity with blogs.
The JALT CALL Journal, 2(1), 15-26.

Ganji, M. (2009). Teacher-correction, peer-correction and self-correction: Their impacts on

Iranian students’ IELTS essay writing performance. The Journal of Asia TEFL, 6(1),
117-139.

135
Haring-Smith, T. (1994). Writing together: Collaborative learning in the writing classroom.
New York: Harper Collins.

Hunsinger, J., & Senft, T. M. (2014). The social media handbook. London: Routledge.

Isa, F. H. O. (2012). The effect of using wikis on improving palestinian 9th Graders. English
writing skills and their attitudes towards writing. Unpublished master’s thesis, The
Islamic University of Gaza, Gaza, Palestine.

Jesson, J., & Peacock, G. (2012). The Really Useful ICT Book: A practical guide to using
technology across the primary curriculum. London: Routledge.

Judd, T., Kennedy, G., & Cropper, S. (2010). Using wikis for collaborative learning: assessing
collaboration through contribution. Australasian Journal of Educational Technology,
26(3), 341-354. https://doi.org/10.14742/ajet.1079

Kavaliauskiene, G. (2010). ESP writings: Weblogs or wikis? English for Specific Purposes
World, 30(9), 112-118.

Kennedy, E. (2010). Blogs, wikis, and e-portfolios: The effectiveness of technology on actual
learning in college composition. Doctoral dissertation, George Mason University,
Virginia.

Kessler, G., Bikowski, D., & Boggs, J. (2012). Collaborative writing among second language
learners in academic web-based projects. Language Learning & Technology, 16(1),
91-109. Retrieved April 18, 2012, from
http://llt.msu.edu/issues/february2012/kesslerbikowskiboggs.pdf

Klobas, J.E. & Beesley, A. (2006). Wikis: Tools for information work and collaboration.
Oxford: Chandos Publishing.

Koschmann, T. (1999). Toward a dialogic theory of learning: Bakhtin’s contributions to

understanding learning in settings of collaboration. In C.M. Hoadley & J. Roschelle
(Eds.), Proceedings of the conference on Computer Support for Collaborative Learning
(CSCL’99) (pp. 308-313). Mahwah, Palo Alto, CA.

Kuteeva, M. (2011). Wikis and academic writing: Changing the writer-reader relationship.
English for Specific Purposes, 30(1), 44-57. https://doi.org/10.1016/j.esp.2010.04.007

136
Lee, K-W. (2000). English teachers' barriers to the use of computer-assisted language learning.
The Internet TESOL Journal, 6(12). Retrieved November 26, 2001, from
http://iteslj.org/Articles/Lee-CALLbarriers. html

Leuf, B., & Cunningham, W. (2001). The wiki way: Quick collaboration on the Web. Boston,
MA: Addison-Wesley Longman Publishing Co.

Lin, H. P. (2005). Online collaborative writing with wiki technology: A pilot study.
International conference on e-learning. Retrieved March 22, 2009, from
http://learning.edu.tw/conference/2005IceL/paper_paper/E002/E5.pdf

Mackey, A., & Gass, S. M. (2005). Second language research methodology and design. New
Jersey: Lawrence Erlbaum Associates.

Mak, B., & Coniam, D. (2008). Using wikis to enhance and develop writing skills among
secondary school students in Hong Kong. System, 36(3), 437-455.
http://dx.doi.org/10.1016/j.system.2008.02.004

Marzec-Stawiarska, M. (2015). Wikis and new perspectives for collaborative writing. In L.

Piasecka, M. Adams-Tukiendorf, & P. Wilk, (Eds.), New Media and Perennial
Problems in Foreign Language Learning and Teaching (pp. 39-55). Switzerland:
Springer International Publishing.

McKay, E. (Ed.). (2014). ePedagogy in online learning: New developments in web mediated
human computer interaction. Hershey, PA: IGI Global.

Mesquita, A. (2012). Sociological and philosophical aspects of human interaction with

technology: Advancing concepts. Hershey, PA: IGI Global.

Miyazoe, T., & Anderson, T. (2009). Learning outcomes and students’ perceptions of online
writing: Simultaneous implementation of a forum, blog, and wiki in an EFL blended
learning setting. English for Specific Purposes, 30(1), 44-57.
https://doi.org/10.1016/j.system.2010.03.006

Murugesan, S. (2010). Web X.0: A road map. In Handbook of Research on Web 2.0, 3.0, and
X.0: Technologies, Business, and Social Applications (pp. 1-11). Hershey, PA: IGI
Global.

137
Neumann, D. L., & Hood, M. (2009). The effects of using a wiki on student engagement and
learning of report writing skills in a university statistics course. Australasian Journal
of Educational Technology, 25(3), 382-398. http://dx.doi.org/10.14742/ajet.1141

Norton, P., & Hathaway, D. (2008). On its way to K-12 classrooms, Web 2.0 goes to graduate
school. Computers in Schools, 25(3), 163-180.
https://doi.org/10.1080/07380560802368116

Parker, K. R., & Chao, J. T. (2007). Wiki as a teaching tool. Interdisciplinary Journal of
Knowledge and Learning Objects, 3(1), 57-72.

Reio Jr, T. G., & Hill-Grey, K. (2012). Millennial adult learners in the 21st century:
implications for adult and community educators. In C. X. V. Wang, (Ed.). Handbook
of Research on Technologies for Improving the 21st Century Workforce: Tools for
Lifelong Learning (pp. 425-440). Hersey, PA: IGI Global.

Reo, R. (2006). Scaffolding student collaboration for group wiki projects. Chapter 4 In Mader,
S. Using Wikis in Education. Retrieved November 26, 2007, from
http://www.wikiineducation.com.

Riazi, M. (2016). The Routledge encyclopedia of research methods in Applied Linguistics.

London: Routledge.

Shelly, G., & Vermaat, M. (2011). Discovering computers and Microsoft Office 2010: A
fundamental combined approach. Boston: Cengage Learning.

Statistical Packages for the Social Sciences (Version 21) [IBM Corporation]. Armonk, NY:
USA.

Storch, N. (2011). Collaborative writing in L2 contexts: Processes, outcomes, and future

directions. Annual Review of Applied Linguistics, 31(1), 275-288.
http://dx.doi.org/10.1017/S0267190511000079

Suwantarathip, O. & Wichadee, S. (2014). The effects of collaborative writing activity using
Google Docs on students’ writing abilities. The Turkish Online Journal of Educational
Technology- TOJET, 13(2), 148-156.

138
Tella, A. (2015). Social media strategies for dynamic library service development. Hershey,
PA, USA: IGI Global.

Vaičiūnienė, V., & Užpalienė, D. (2010). Authentic resources in technology-based ESP

learning. Studies about Languages, 17, 94-98.

Warschauer, M. (2010). Invited commentary: New tools for teaching writing. Language
Learning & Technology, 14(1), 3-8.

Warschauer, M., & Healey, D. (1998). Computers and language learning: An overview.
Language Teaching, 31(2), 57-71. Retrieved November 19, 2001, from
http://www.gse.uci.edu/markw/overview.html

Wei, C., Maust, B., Barrick, J., Cuddihy, E., & Spyridakis, J. H. (2005). Wikis for supporting
distributed collaborative writing. Proceedings of the Annual Conference of the Society
for Technical Communication, Arlington, VA, USA, 204-209.

West, J. A., & West, M. L. (2009). Using wikis for online collaboration: The power of the
read-write web. San Francisco, CA: Jossey-Bass. https://doi.org/10.5860/choice.46-
6341

Wheeler, S. (2010). Open content, open learning 2.0: Using wikis and blogs in higher
education. In U. Ehlers, and D. Schneckenberg (Eds.), Changing Cultures in Higher
Education: Moving Ahead to Future Learning (pp. 103-114). Berlin: Springer-Verlag.

Xiao, Y., & Lucking, R. (2008). The impact of two types of peer assessment on students'
performance and satisfaction within a Wiki environment. Internet and Higher
Education, 11(3-4), 186-193. https://doi.org/10.1016/j.iheduc.2008.06.005

Zorko, V. (2009). Factors affecting the way students collaborate in a wiki for English language
learning. Australasian Journal of Educational Technology, 25(5), 645-665.
http://dx.doi.org/10.14742/ajet.1113

139
Appendix A
Interview Questions

1. Briefly explain your experience in using wikis for writing assignments.

2. Do you think your English writing skills have improved through online collaboration with your peers?
Why?
3. Do you think that the collaboration in the wikis helps you learn better how to write?
4. How much do you think exchanging the comments via the Internet could be useful in the process of
learning how to write?
5. Can you name some of the advantages of using wikis? Or tell me the most valuable and interesting part
of wiki writing and why.
6. Can you name some of the disadvantages of using wikis? Or describe the most difficult part of the wiki
writing and why.
7. How does the use of the wiki help you achieve your writing goals?
a. Do you think that wikis improve your creativity and writing knowledge?
b. Has wiki made the writing process easier for you?
c. Does wiki help you organize your ideas for writing?
d. Does wiki help you write faster in a shorter amount of time?
e. Do you think that the history page used to review the changes you had had in writing helps you
improve your writing skills?
f. Do you think that the discussion page plays a positive role in posting your comments and
communicating with the others?
g. Do you think that editing information in the wiki is easy?
h. Does wiki lessen the level of your anxiety while writing?
i. Does wiki help you learn from your own mistakes?
8. Overall, tell me how the experience of using wikis as a collaborative writing tool changed your
perspectives on L2 writing.
9. Do you have any suggestion for the future use of wiki technology?

Appendix B
A Sample of Researcher's Feedback on a Student’s Writing

140
Appendix C

Students' Emails to the Researcher

Our writings have been improved a lot in your class.

Many Thanks and Good Job ☺

141
 Investigating Saudi Medical Students’ Attitudes Towards English-Arabic
Code-Switching in Classroom Instruction

Mohammed Qurait Alenezi

College of Art, Law and Education, University of Tasmania, Australia

Dr Paul G Kebble

National Institute of Education, Singapore, and an adjunct academic with the University of
Tasmania

Biodata

Mohammed Qurait Alenezi is a Ph.D research scholar at the Faculty of Education, University
of Tasmania, Australia. His research thesis focuses on the ways and strategies (conscious and
unconscious) used by Saudi Arabian youths while constructing their different identities online
as well as for being (im)polite. The present paper is a part of this research study. He has
attended various international conferences and presented papers on different aspects of the
research which he is currently carrying out. His major areas of research interest include-
identity construction, online communication, language and Saudi Arabian society.

Email: [email protected]

Dr. Paul G. Kebble has over 30 years of international English language teaching, teacher
training and academic research experience and has lived and worked in eight countries:
Australia, Barbados, Brunei Darussalam, Fiji, Malaysia ,Portugal, U.A.E. and U.K. His
research generally focuses on language learning and communication within an online
environment, culture and language learning and teaching, and the support and assessment of
English for general and academic purposes. Paul has recently had 2 apps published with Apple,
the ‘IELTS predictor’ and ‘How good is my English’, both available as free downloads.

Email: [email protected]

142
Abstract

The present study aimed to investigate a group of monolingual Arabic-speaking Saudi

Arabian medical students and their attitudes towards the medium of instruction and code-
switching within their classroom. A quantitative analysis of research data collected from 189
students (male and female) through the use of questionnaires, revealed there was a strong
preference for using code-switching over a monolingual medium of instruction. Although the
majority of the students strongly agreed that using one language had certain benefits, they
suggested code-switching was more desirable for course content comprehensibility. The data
showed students disagreed that code-switching led to any confusion in teaching and learning
and the students described their increased respect for the lecturer if s/he condoned or practised
code-switching themselves. These findings suggest managers of learning and teaching should
consider revising their monolingual language policies to embrace the positives within learning
that code-switching can provide.

Keywords: code-switching, medium of instruction, monolingual classrooms, medical

students, Saudi Arabian learners, EFL

1. Introduction

Medium of instruction plays vital role in effective imparting of subject knowledge and thereby
achieving set teaching and learning goals. Thus, any decision regarding the medium of
instruction can have a direct impact on whole teaching and learning process and on all those
related with these processes including teachers, learners, policy makes and quality controllers.
This is the case for all subject-specific learning, and medical sciences equally, if not more so
than others. Analysis of student feedback is essential for improving learning and teaching in
context, and for developing the linguistic medium for achieving desired outcomes.

The use of English as a medium of instruction in many Gulf-countries, where it is not the main
language, can sometimes be problematic both for the teachers and students. In such bilingual
classrooms, code-switching is an often-observed phenomenon as invariably all learning and
teaching participants are able to fluently use the first language, Arabic. As the phenomenon of
code-switching is common within bilingual and multilingual societies, research on the topic
(Giancaspro, 2015; MacSwan, 2016; Ribot, Hoff &Burridge2017) has also been increasing.
Particularly in recent years, the phenomenon of code-switching has become the centre of
research interest and has generated extensive arguments and counterarguments regarding the
practice of code-switching inside bilingual classrooms.

143
With this background, the present study investigated the language position and attitudes of
Saudi Arabian students at the medical college of Northern Border University, Saudi Arabia,
towards code-switching in classroom teaching involving both Arabic and English. The study
revolved around the central research question: What is the students’ attitude towards using one
or two languages in code-switching used in the medical courses at Northern Border University,
Saudi Arabia? Studying students’ attitudes towards medium is important as medium used for
instruction can affect whole process of teaching and learning and the set aims of the whole
instruction programme.

2. Literature Review

2.1 Code-switching and code-mixing

There are many definitions of code-switching offered by various scholars and there has yet to
be consensus, particularly as it is implemented differently in different contexts. These codes
can be two distinct languages, or dialects derived from the same base language. For example,
Gumperz (1982) explained that code-switching was “the juxtaposition within the same speech
exchange of passages of speech belonging to two different grammatical systems or
subsystems” (p.59), while Cook (2001) described the process as requiring two agents able to
use both codes when code-switching was moving “from one language to the other in mid-
speech when both speakers know the same languages” (p.83). Numan and Carter (2001)
defined code-switching as “a phenomenon of switching from one language to another in the
same discourse” (p. 275), whilst it was defined by Myers-Scotton (2006) as “the use of two
language varieties in the same conversation”(p. 239).

In addition to code-switching, “code-mixing” is a term that was used within the learning
environment explored in this study, and therefore requires further elucidation in order to avoid
confusion. Bokamba (1989, p.278) stated that code-mixing is “the embedding of various
linguistic units such as affixes (bound morphemes), words (unbound morphemes), phrases and
clauses from two distinct grammatical systems or sub systems within the same sentence and
speech event”. This implies that code-mixing arises in the intra-sentential level as opposed to
code-switching which arises in the inter-sentential level. Myers-Scotton (1993) added that
code-switching occurs when bilingual people alternate between two languages during one
interaction with another bilingual person, while code-mixing is the use of words, affixes,
phrases and clauses from more than one language within the same sentence. However,
according to Muysken (2000), code-switching is a term that is used for cases in which the two

144
codes maintain their monolingual features, while code-mixing is used for cases where there is
some convergence between the two languages. In the learning context of this research, spoken
Arabic can often be interspersed with technical and scientific phrases emanating from their
English usage (although sometimes Latin-based), with the occasional use of prefixes as
modifiers. For this study, code-switching is defined as the ability to shift from Arabic to English
within a conversation, discussion or utterance, and when both learners are extensive users of
both languages. This also extends into the context of the language classroom, and includes the
alternate use of the first language and the target language as means of communication and
clarification by the lecturer, when the need arises.

2.2 Language position in the Kingdom of Saudi Arabia

Language position or positioning relates to how each language is perceived within the societies
that utilise them (Abdi, 2011&Kayi-Aydar2014). For example, the use of English in French is
often derided and has negative social connotations, whereas in UK English, being able to use
French phrases (code-mixing) is seen as having a superior command of language and an
indicator of quality education. In the case of Saudi Arabian society, there are different factors
which can be attributed to bilingualism wherein the prestige forms the highest. That is, the
Saudi middle, upper and ruling classes choose English vocabulary and expressions in their
daily interaction and prefer to use English in every facet of their lives. Professionals make it a
matter of personal prestige to use English extensively in their interaction with people, primarily
to show others that they are fluent in the language and obviously well-educated. To meet the
need of the situation is another reason for the proliferation of bilingualism in Saudi Arabia. It
is important, however, to note here that many bilingual Saudis often utilise code-switching as
they need to fill their conversation gaps in Arabic with English words, or vice versa,
particularly for vocabulary that is absent either in Saudi Arabic, or in English.

2.3 Attitudes towards code-switching in learning and teaching

Attitudes to code-switching in the classroom are subject to the level of perception of its use as
good practices by the lecturer or the institution, with research having identified various
attitudes towards demonstration of this communication behaviour (Kachru, 1978; Grojean,
1982; Gumperz, 1982;Keh and Stoessel, 2016;Balam and Prada Pérez, 2017). These attitudes
vary from a belief that there is no intrinsic linguistic benefit from accepting or utilising its

145
practice, to a belief that code-switching can enhance and accelerate language comprehension
and hence learning.

2.3.1 Code-switching enhancing learning and teaching

Baker (2006) listed twelve main purposes of code-switching, including being used to
emphasise a particular point, to substitute a word in place of an unknown word in the target
language, to express a concept that has no equivalent in the culture of the other language, to
reinforce a request, to clarify a point, to express identity and communicate friendship, to ease
tension and inject humour into a conversation. In some bilingual situations, code-switching can
also occur when certain topics are introduced, providing a shortcut to the understanding of a
concept or lexical item which, in the medical setting of this research, appeared to be prevalent.
Skiba (1997) stated that code-switching should be regarded as an opportunity for language
development, since it serves as a signal to tell the listener the need for providing samples from
another language. Although language switching may sound disruptive for the listener at times,
if the listener understands the meaning of the code-switching language, then s/he is provided
with an opportunity for language learning and development. Learning and developing activities
will take place as translation into the second language is provided. This, in turn, will allow for
a reduced amount of switching and less subsequent interference as time progresses. These
principles may also be applied in the second language classroom.

Duran (1994) suggested that code-switching should be seen from the perspectives of
interlinguistic phenomena specific to bilingual people, and that the relationship between code-
switching and interlanguage should be taken into consideration. The notion of interlanguage is
associated with the earlier developmental period of bilingual people, while code-switching is
regarded as occurring during the later phase of bilingual acquisition. More specifically, Duran
(1994) recognized interlanguage as the language constructed before the ideal forms of the target
language is achieved, while code-switching may occur during or after the interlanguage stage.
Duran (1994) also stated that in order to perform code-switching, competence in both
languages is required, even if the speakers have not fully achieved fluency in both languages.
She even further explained that code-switching must serve important functions for the language
learners/users, in order for it to happen naturally in the scheme of bilingualism (Duran, 1994).

However, to use or not to use code-switching inside the classroom is contentious among
research scholars. Those supporting its use in learning vouch for its usefulness as a teaching
tool. For example, Rollnick and Rutherford (1996) believed that code-switching assists learners

146
in exploring their ideas. In the context of a science classroom, they perceived code-switching
as a tool used by learners to reveal their alternative thoughts on the subject matter being taught.
Code-switching to the students’ mother-tongue also enables them to employ sense-making
resources (Amin, 2009). Hornberger (2005) also argues that bilingual or multilingual learning
is enhanced when students are permitted to use their resources in their existing language skills
in one or two languages as opposed to being confined to monolingual instructional behaviour
and practices. The ability to move from one language to another is not only quite “normal” but
also often “highly desirable among learners” and, no matter how it might be disruptive to the
listener during a conversation, it still provides an opportunity for language development (Cook,
2001,p. 105).

There are a number of studies which have investigated positive attitudes of both educators and
learners towards the use of code-switching inside the classroom. A study by Ariffin and Husin
(2011) found bilingual instructors functioning with learners with the same two languages
frequently code-switched and code-mixed between these languages in the classroom in order
to facilitate what was believed to be more effective teaching and learning, in particular as a
shortcut to clearer interpretations and understanding. Jingxia (2010) also reported that teachers’
code-switching between Chinese and English existed in the English as a Foreign Language
(EFL) classrooms of Chinese universities. All of the teachers and students reported that there
was switching to Chinese “sometimes” or “occasionally” in their English-language classes,
both consciously and unconsciously. However, most of the teachers (80%) and students (66%)
held a positive view with regard to teachers’ code-switching to the first language (L1).

Regarding learners’ positive attitudes towards code-switching, Macaro’s (1997) earlier study
reported that most students expected their bilingual teachers to speak the L1 sometimes to
facilitate their understanding, with many also indicating they could not learn if they could not
understand their teacher. Additionally, research conducted by Arthur and Martin (2006), with
students in Malaysia having high English proficiency, found that code-switching
communicative behaviour could be off-putting and does not help in improving linguistic
competencies in English. Rolin-Ianziti and Brownlie (2002) also reported that the use of the
learners’ first language was conducive to the correct understanding of the newly input target
language for the students. Similarly, in research conducted in Ormoc city in the Philippines, on
students’ attitudes towards English and Fil-English code-switching, Durano (2009) reported
that the respondents had a positive attitude towards English and Fil-English code-switching.

147
In further research, Alenezi (2010) investigated students’ language activities in the context of
Arabic and English code-switching which was being used as the medium of instruction of a
science subject in the Human Development for Occupational Therapy course at the Allied
Health Science College in Kuwait University. The findings of the study showed students’
strong preference toward a specific medium of instruction that is Arabic-English code-
switching. In another study from the Saudi Arabian context, Alenezi (2016) reported that Saudi
Arabian English for Specific Purposes (ESP) students at Northern Border University agreed
that ‘code switching was more desirable and believed that it made the course easy to
understand’ (p. 154). Thus, the findings of these studies show the positive attitudes of learners
from different bilingual contexts towards code-switching.

2.3.2 Code-switching as a detractor to learning and teaching

Those with a belief that code-switching can have a negative impact on learning relate the
process to negative language habits (Shin, 2005). In this respect, Martin (2005) asserts that the
practice of code-switching in many teaching situations is either condemned as unprofessional
or considered to be an institutionally unrequired practice. In many cases, the practice of code-
switching in teaching is attributed to the bilingual teacher’s lack of competence in the English
language and possibly lack of classroom management skills in preventing it happening between
students. A study conducted by Payawal-Gabriel and Reyes-Otero (2006) reported that the
practice of code-switching by Filipino mathematics teachers in their classroom instructions had
a negative impact on learning. The findings revealed that teachers’ code-switching confused
students and thus affected their lesson comprehension. Through early research conducted at the
University of California, Los Angeles, Duff and Polio (1990) suggestedteachers using their
first language (Spanish) were depriving students of many opportunities to be exposed to and
deal with the target language (English) and concluded that in the process of learning the second
language (L2), L1 should not be used.

In accordance with previous views, Halliwell and Jones (1991) claim that in order to learn L2
successfully, learners should take the risk to practice using L2 in both speaking and
understanding L2. They believe that using L2 as a normal mean of communication is possible.
To support their claim, Halliwell and Jones (1991) proposed that learners can understand the
content of the message even though they do not fully understand the exact meaning of words
or structures. Furthermore, in support of this theory, Macdonald (1993) believes that the focus
on L2 can activate the conscious and unconscious learning of L2. The use of L2 can enhance

148
the students’ communication skills, thus giving learners the confidence and challenge to
communicate in L2 even with limited knowledge of L2. Macdonald (1993) even advises
teachers that, “If you get stuck in the middle of a lesson, try to communicate your message by
some other means, such as mime or demonstration” (p. 23).

The opinions and methods of avoiding switching between L1 and L2 in class are based on two
assumptions: that L1 and L2 learning processes are similar, and the previous L1 learning
experience and process should not be linked to L2 (Cook, 2001). Cook (2001) believes that L2
should be developed with no reference to L1 due to language compartmentalisation, and the
reasoning behind this recommendation is to avoid L1 interference. One example of language
interference would be an L1 negative transfer that would result in an L2 language error. The
next section of this paper will review selected arguments against these assumptions and discuss
how code-switching can be implemented in the English as a Second Language (ESL) and EFL
classrooms.

2.3.3 Functions of code-switching in the classroom

Mattson and Burenhult (1999) describe four functions of code-switching which may be
beneficial in L2 learning environments, whether performed consciously or subconsciously.
These code-switching functions are as follows: topic switch, where the teacher shifts from L2
to L1 to explain a particular grammatical item so that the new content and meaning can be
directly and easily transferred; affective functions, where code-switching serves as expression
of emotions and is used to build solidarity and relationships with the students, whether
consciously or subconsciously; socialising functions, when the teachers signals connectedness
and solidarity by using L1; and repetitive functions, where clarifications of meaning are offered
using L1 after L2 has been used.

3. Research Methodology

This is a quantitative study with a survey approach. In order to measure the students’ language
attitudes towards code-switching, a questionnaire was administered to 230 Saudi Arabian
medical students studying at the faculty of medicine in the Northern Border University of
Kingdom of Saudi Arabia out of whom 194 returned completed questionnaires. The
questionnaires were administered by the researcher himself after visiting the research site and
requesting the participants to complete it for the study purpose. The process was voluntary.
Five of these questionnaires were excluded from the analysis as they contained too many

149
unanswered questions. Therefore a total of 189 questionnaires were analysed. All the formal
permissions were obtained from the university administrations to conduct the study and
participation in study was voluntary for the students. The data collected were analysed using
SPSS to determine the quantitative results gained from the questionnaire. According to
Creswell (2008), the results from the quantitative research method have the potential to be
generalised to larger populations if an appropriate sampling design had been utilised; if the
questionnaire was properly constructed, it could have high measurement reliability, and high
construct validity if proper controls were implemented.

3.1 Participants

The participants in this study were medical students at the Northern Border University in Saudi
Arabia. The total number of students who participated in the study was 230 (127 male and 103
female students) with ages ranging from 18 to 23 years and they were studying in different
years of medical education. The details are tabulated below:

Table 1: The Details of the Participants

Academic The Faculty of Medicine, Northern Border University, Kingdom of Saudi
Institute Arabia
Gender/Year Male Female Age Total
Second Year 50 30 18-20 80
Third Year 42 38 20-22 80
Fourth Year 35 35 22-23 70
Total 127 103 - 230

3.2 Data collection instrument

The main instrument for data collection was the questionnaire (see Appendix A). Cohen,
Manion, Morrison, and Morrison (2007) explain that a questionnaire is a highly effective
instrument for the collection of information, especially as it can be administered in the absence
of the researcher. In the present study, the questionnaire was adapted from Olugbara’s (2008)
study to measure students’ attitudes towards code-switching. The questionnaire was divided
into two major sections. The first section collected the demographic information of the
students, such as gender, language taught in previous school and language use as a medium of
instruction in the class. Section two of the questionnaire contained 10 items asking for specific

150
information in relation to attitudes and understanding of code-switching within their learning
environment. Other than the demographic questions, all of the survey questions used the Likert
scale format, in which a four-point scale (strongly disagree-disagree-agree-strongly agree) was
used.10 items were finalised as they covered the aims of the study and were enough to get
research related data. The questionnaires were distributed on the campus by the researcher and
collected by hand on the same day.

4. Data Analysis

Data analysis concerns analysing the data collected from participants through the instrument.
The researchers used the digital Statistical Package for Social Sciences (SPSS) version 16 to
analyse the quantitative data that had been extracted from the completed questionnaires.
Descriptive statistics were used to explore, summarise and describe the data. In this regard,
Pallant (2007) states that descriptive statistics are aimed at depicting the different attributes of
data, verifying any violation of the principal assumptions for the statistical methods to be used
in the study, and addressing particular research questions. In this study, the descriptive statistics
were undertaken using central tendency and variation statistics such as frequency, means,
ranges, and standard deviation.

4.1 Pilot study

A pilot test was conducted on a randomly chosen group of 10 students (both male and females)
of faculty of medicine in order to increase the accuracy and consistency of the data collection
methodology and its measurements, as recommended by Cohen, Manion, Morrison and
Morrison (2007).The test was done in the month of December, 2015, prior to the distribution
of the questionnaires to the participants of the study. The test was performed with the purpose
of assessing test face validity, i.e. did the questions appear to those tested to be suitable for the
task, as well as conducting an initial reliability analysis. Each of these is discussed in the
following sections.

151
Table 2: Cronbach’s Alpha Analysis

Cronbach’s
Variable Number of Items
Alpha
Attitude 1-2-3-4 0.798
L1 and L2 5-6-7 0.871
Teacher’s
8-9-10 0.801
image
Results 11-12-13 0.750
Code-
5 to 13 0.819
switching

4.2 Validity and Reliability

A panel of four academic peers at Northern Border University examined the instrument and
provided feedback. They were chosen based on their teaching experience and higher
qualification in the area of the research study. The valuable suggestion included making the
language of the questions simpler for better understanding of the respondents as they were
medical students; reducing the number of questions to 10 so that the respondents did not feel
burdened to answer; inclusion of the statement that there is no right or wrong answer and they
should express their views honestly. The revised and final pilot testing assessment of face
validity revealed the questions did not cause problems for students in terms of language and
clarity. The results of this pilot study (shown in Table 2) revealed that the reliability, as a
measure of stability and consistency in which the instrument measures the concept and assesses
the integrity of the measurement instrument (Sekaran & Bougie, 2010), was acceptable. All
items for Attitude (alpha = 0.798) and Code-Switching (alpha = 0.819) met the conditions set
for consistency and reliability.

Table 3: Cronbach’s Alpha: Scale Reliability [Pilot Test]

Overall
Number Cronbach’s
Variable Factor Cronbach’s
of Items Alpha
Alpha
Positive attitude
Attitude towards code- 3 0.798 0.798
switching
Code- First language
1 0.781 0.819
Switching and second

152
 language
confusion
Teacher’s
3 0.801
Images
Results 3 0.756

4.3 Findings
This study employed an initial sample of 230 students from Northern Border University, of
whom 194 returned completed questionnaires. Five of these questionnaires were excluded from
the analysis as they contained too many unanswered questions. Therefore a total of 189
questionnaires were analysed and, from the responses, the following recapitulation frequency
table was constructed (Table 3), presented as a table of percentages.

Table 4: Students’ Attitudes Towards Code-Switching

SD DA A SA Ratio %
No Statement M Dis - Ad
(%) (%) (%) (%)
Attitude
Teaching the course 21.7 -78.3
1 only in one language is 9.5 12.2 49.2 29.1 2.9788
beneficial to me.
Teaching the course in 0 -100
2 Arabic and English is 0 0 69.8 30.2 3.3016
desirable to me.
Teaching the course in 0 -100
Arabic and English
3 0 0 69.3 30.7 3.3069
makes it easy for me to
understand.
It confuses me when 92.6 -7.4
course instructor
4 teaches in Arabic and 22.2 70.4 7.4 0 1.8519
English at the same
class period.
Opinion of the instructor based on the use of languages in teaching
I respect an instructor 0 - 100
5 more when teaching in 0 0 50.8 49.2 3.4921
Arabic and English.
I respect an instructor 17 - 83
6 more when teaching in 7.4 9.6 46 37 3.1270
Arabic.
I respect an instructor 15.3 - 84.7
7 more when teaching in 7.4 7.9 45.5 39.2 3.1640
English.
Opinion on academic results based on language used in the
classroom
Teaching the course 18 - 82
only in Arabic increases
8 0 18 43.4 38.6 3.2063
my chances of passing
the exams.
Teaching the course 16.9 - 83.1
only in English
9 2.6 14.3 41.8 41.3 3.2169
increases my chances of
passing the exams.
Teaching the course in 9 - 91
10 0 9 43.9 47.1 3.3810
Arabic and English

153
 increases my chances of
passing the exams.
Note: SD (strongly disagree); DA (disagree); A (agree); SA (strongly agree);
M (mean)

This study aimed to explore Arabic-speaking medical students’ attitudes towards the use of
code-switching between Arabic and English in learning and teaching. Overall, the results of
the questionnaire showed positive attitudes towards the use of code-switching by the majority
of students, consistently indicated through all elements of the collected data (Table 3), with
one anomaly. Item one in the table clearly indicates that these students believed teaching the
course in only one language is beneficial, however, which language is not specified. The
researchers speculated this response might indicate that the respondents believed the use of
English only in learning and teaching would be beneficial, particularly as the questionnaire
itself was conducted in English.

Through further examination of the presented results, items 2, 3 and 5 (Table 3) indicated the
greatest accord. Items 2 and 3 were closely aligned and the researchers suggested respondents
were stating that code-switching in learning and teaching was beneficial because it enhanced
understanding. Item 5 perhaps indicated deeper respect through both cultural and professional
understandings, a lecturer who is confidently able to move between languages demonstrates
deep understanding of the topic and in relation to the cultural setting of the students. Items 4
and 10 also showed a very high accord, with item 4 obviously in reverse whereby the
respondents indicated a lack of confusion when engaging in code-switching. Item 10 related
directly to academic success, with respondents believing code-switching enhances examination
success, through a deeper understanding of the learning materials provided through code-
switching, it is postulated. Items 6, 7, 8 and 9 also showed strong agreement ratios, however,
in the cases of items 6 and 7, the indication was that much respect was shown to the lecturer,
whether using Arabic or English, and this would seem to be at odds with item 5. However, it
was hypothesised that students within the learning culture of Saudi Arabia demonstrated much
respect for their lecturers, regardless of the medium of instruction. It was also believed by the
researchers that the questions themselves could have been better worded to investigate more
deeply the subject of respect and language usage. Finally, items 8 and 9 again showed perhaps
that the respondents appreciated the learning process and equated success to whichever
language of instruction is utilised, however, in comparison to item 10, it was seen the latter,
code-switching, was believed to be the most beneficial medium of instruction.

154
4.4 Discussion

The results from this research, corroborating the real benefits in learning and teaching of
moving between languages, correspond well with the findings of a number of studies
previously discussed (Cook, 2001; Jingxia, 2010; Rolin-Ianziti& Brownlie, 2002; Macaro,
1997; Durano, 2009; Alenezi, 2010). However, the findings of this study did not justify those
of Payawal-Gabriel and Reyes-Otero (2006) who reported that the practice of code-switching
by mathematics teachers in their instructions negatively affected learning. Payawal-Gabriel and
Reyes-Otero’s (2006) findings revealed that teachers’ code-switching confused students and
thus affected their lesson understanding. It is postulated by the researchers of the present study,
based on their professional experience that mathematical language, both lexically and
syntactically, is quite specific, and direct translations might lead to confusion and not clear
comprehension.

When comparing the students’ attitudes towards using one language (either Arabic or English)
in teaching through Arabic-English code-switching, the findings of this study indicated a
preference for using code-switching rather than using one language as a medium of instruction.
Although the majority of the students strongly agreed that using one language is still beneficial
to them, they found it more desirable and believed it makes the course material more easily
understood if code-switching is utilised.

5. Conclusion

This study aimed to describe a group of Saudi medical students’ attitudes towards the practice
of code-switching for instruction purposes inside the medical-course classrooms. The
participants’ attitude towards code-switching as a medium of instruction was affirmative. They
discussed it as having a positive impact on academic performance, and was seen as being an
influential teaching tool to facilitate learning and increase student comprehension and hence,
academic success. From the findings obtained, it can be concluded that the Saudi Arabian
medical students at the Northern Border University, Saudi Arabia had positive attitudes
towards the practice of code-switching between Arabic and English for instruction purposes as
was seen to facilitate understanding and help them in passing exams. Although overall
reverence for lecturers was evident, the students also demonstrated greater respect for the
lecturers who practised code-switching during the class. This can be put further into context by
considering that English is taught as a foreign language in Saudi Arabia and it is not easy for
many students to comprehend difficult concepts in medical science purely in English without

155
any supporting explanation in their mother tongue, Arabic. Therefore, it is recommended that
educational decision makers and managers at programs, department, colleges, and university
levels in Saudi Arabia should consider revising their language policy in order to incorporate
code-switching in the planning of syllabi.

At present, medical departments are applying English-only as the language of teaching and
learning, and this is demonstrated in the language policy held at the Northern Border University
(as per the 2009–2010 hand book). Whilst English is the medium of textbooks, assignments,
and examinations, in reality, Arabic-English code-switching is the dominant medium of
classroom communication. Also, educators and teachers of medical subjects at colleges where
English is used as a foreign language might want to consider the students’ language preferences
and attitudes towards medium of instruction. This research also suggests teachers and lecturers
should be encouraged to make adequate use of code-switching in classrooms when explaining
concepts to students so that the students will be able to actively participate in classroom lessons
and hence increase achievement.

The present study included the participants only from the Northern Border University
and only from the faculty of medicine. More studies are required with more participants
from different universities and from different parts of Saudi Arabia to get more diverse
and more reliable results.

References

Abdi, K. (2011). ‘She Really Only Speaks English’: Positioning, Language Ideology, and
Heritage Language Learners. Canadian Modern Language Review. 67(2), pp.
161–190.

Alenezi, A. (2010). Students’ language attitude towards using code-switching as a medium of

instruction in the college of health sciences: An exploratory study. ARECLS7, 1–22.

Alenezi, M. (2016). Gender and students’ attitude toward code-switching: A correlational study
with reference to Saudi Arabian medical students at Northern Border University.
International Journal of English Language & Translation Studies.4 (3), 154–166.

Amin, T. (2009). Language of instruction and science education in the Arab world: Toward a
situated research agenda. In S. BouJaoude& Z. R. Dagher (Eds), The World of Science

156
 Education Volume 3: Arab States (pp. 61–82) Rotterdam, The Netherlands:Sense
Publishers.

Ariffin, K. &Husin, M. S. 2011. Code-switching and code-mixing of English and Bahasa

Malaysia in content-based classrooms: Frequency and attitudes. The Linguistics
Journal 5(1), 220–247.

Arthur, J. & Martin, P. (2006). Accomplishing lessons in postcolonial classrooms:

Comparative perspectives from Botswana and Brunei Darussalam. Comparative
Education, 42(2), 177–202.

Baker, C. (2006). Foundations of bilingual education and bilingualism (4thEd.).Multilingual

Matters Ltd.

Balam, O. and Prada Pérez, A. (2017). Attitudes toward Spanish and Code-Switching in Belize:
Stigmatization and Innovation in the Spanish Classroom. Journal of Language, Identity
& Education. Vol. 16 (1|). 17-31

Bokamba, E. (1989).Code-mixing, language variation, and linguistic theory: Evidence from

Bantu languages. Lingua76, 21–62.

Cohen, L., Manion, L., Morrison, K., & Morrison, K. (2007). Research methods in education
(6th ed.).Psychology Press.

Cook, V. (2001). Second language learning and language teaching (3rded.).London: Arnold.

Creswell, J. (2008). Educational research: Planning, conducting and evaluating quantitative

and qualitative research (3rded.). Lincoln University of Nebraska, Nebraska: Pearson
Prentice Hall.

Duff, P. & Polio, C. (1990). How much foreign language is there in the foreign language
classroom? Modern Language Journal. 74,154–166.

Duran, L. (1994). Toward a Better Understanding of Code Switching and Interlanguage

InBilinguality: Implications for Bilingual Instruction. The Journal of Educational Issues
of Language Minority Students, V14. P69-88.

157
Durano, F. (2009). Attitudes towards English and Fil-English code-switching amongst high
school students in Ormoc city, Philippines (student thesis).Malmö University.

Giancaspro D. (2015). Code-switching at the auxiliary VP boundary: A comparison of heritage

speakers and L2 learners. Linguistic Approaches to Bilingualism, 5(3), 380–408.

Grosjean, F. (1982). Life with Two Languages: An Introduction to Bilingualism. Cambridge,

Mass: Harvard University Press.

Gumperz J. (1982). Discourse strategies. Cambridge: Cambridge University Press.

Halliwell, S. & Jones, B. (1991). On target. London: Centre for Information on Language
Teaching.

Hornberger, N. (2005). Opening and filling up implementational and ideological spaces in

heritage language education. Modern Language Journal 89, 605–609.

Jingxia, L. (2010). Teachers’ code-switching to the L1 in EFL classroom. The Open Applied
Linguistics Journal3, 10–23.

Kachru, B. (1978). Code-mixing as a communicative strategy in India. In J. Alatis (ed.).

International dimension of bilingual education (pp. 25–52). Washington, DC:
Georgetown University Press.

Kayi-Aydar, H. (2014). Social Positioning, Participation, and Second Language

Learning: Talkative Students in an Academic ESL Classroom. TESOL Quarterly.
48 (04), Pp. 686–714.

Keh, M. and Stoessel, S. (2016). How First Is First? Revisiting Language Maintenance and
Shift and the Meaning of L1/L2 in Three Case Studies. International Multilingual
Research Journal. Vol: 11(02). 101-114

Macaro E. (1997). Target language, collaborative learning and autonomy.Clevedon:

Multilingual Matters.

Macdonald, C. (1993). Using the target language. Cheltenham: Mary Glasgow.

MacSwan J. (2016). Code switching and the timing of lexical insertion. Linguistic Approaches
to Bilingualism, 6(6), 782–787

158
Martin, P. (2005). Safe language practices in two rural schools in Malaysia. Tensions between
policy and practice. In A.M. Lin &P.W. Martin (Eds.), Decolonisation, globalization:
Language-in-education policy and practice (pp. 74–97).Clevedon, UK: Multilingual
Matters.

Mattsson, A &Burenhult-Mattsson, N. (1999). Code-switching in second language teaching of

French. Working Papers 47: 59-72.

Muysken, P. 2000. Bilingual speech: A typology of code-mixing. Cambridge: Cambridge

University Press.

Myers-Scotton, C. (1993). Social motivations for code switching. Oxford: Clarendon Press.

Myers-Scotton, C. (2006). Multiple voices: An introduction to bilingualism. London:

Blackwell.

Northern Border University. (2009). Northern Border University Handbook.Arar: Northern

Border University Press.

Numan, D. and Carter, D. (2001). Teaching English to Speakers of Other Languages.

Cambridge: Cambridge University Press

Olugbara, C. (2008). The effects of IsiZulu/English code switching as a medium of instruction

on students’ performance in and attitudes towards biology (MA dissertation
thesis).University of Zululand.

Pallant, J. (2007). SPSS survival manual, 3rd edition. Philadelphia, PA: Open Press University,
McGraw-Hill

Payawal-Gabriel, J. & Reyes-Otero, M. (2006). Comprehensible input strategies and

pedagogical moves using Filipino/English as medium of instruction in secondary
mathematics. Education Quarterly, 83–97.

Ribot, K. M., Hoff, E. and Burridge, A. (2017). Language Use Contributes to Expressive
Language Growth: Evidence From Bilingual Children. Child Dev.
doi:10.1111/cdev.12770

159
Rolin-Ianziti, J. &Brownlie, S. (2002). Teacher use of learners’ native language in the foreign
language classroom. Canadian Modern Language Review58, 402–426.

Rollnick, M. & Rutherford, M. (1996). The use of mother tongue and English in the learning
and expression of science concepts: A classroom-based study. International Journal of
Science Education 18 (1), 91–103.

Sekaran, U., & Bougie, R. (2010). Research Methods. For Business: A Skill Building
Approach (5th ed.). West. Sussex, UK: John Wiley & Sons Ltd.

Shin, S. (2005). Developing in two languages: Korean children in America. Clevedon, UK:
Multilingual Matters.

Skiba, R. (1997). Code Switching as a Countenance of Language Interference. The Internet

TESL Journal. 3,10<ttp://iteslj.org/Articles/Skiba-CodeSwitching.html>

APPENDICES
Appendix A: Questionnaire
Introduction
This questionnaire is designed to find out your honest views about the language of teaching at your current course.
Please respond to all the questions below carefully and honestly. This is not a test and there are no right or wrong
answers. Your responses will be kept strictly confidential, and will only be used for the purpose of this study.
Your answers will not prejudice you in any way.
Section A: Biographical information:
Please, answer the following questions.
1. What is your gender?
( ) Female.
( ) Male.
2. In what language(s) have you been mostly taught in your previous schooling?
( ) Arabic.
( ) English.
( ) English and Arabic.
3. What language(s) do you use in communicating with your classmates, teachers, and staff at the college?
…………………………………………………………………………………………
…………………………………………………………………………………………

Section B: Learners honest views about the teaching language.

Please read each of the following statements very carefully and tick the answer which best describes your degree
of agreement or disagreement. The following abbreviations are used: SA - Strongly Agree; AG - Agree; DA
Disagree; SD - Strongly Disagree.

160
 SD DA AG SA
NO ITEMS DESCRIPTION
(1) (2) (3) (4)
Teaching the course only in one language is
1
beneficial to me.
Teaching the course in Arabic and English
2
is desirable to me.
Teaching the course in Arabic and English
3
makes it easy for me to understand.
It confuses me when course instructor
4 teaches in Arabic and English at the same
class period.
I respect instructor more when teaching in
5
Arabic and English.
I respect instructor more when teaching in
6
Arabic.
I respect instructor more when teaching in
7
English.
Teaching the course in Arabic increases my
8
chances of passing the exams.
Teaching the course in English increases
9
my chances of passing the exams.
Teaching the course in Arabic and English
10
increases my chances of passing the exams.

161
 Customized EAP Program for Novice Researchers in Engineering: Focus
on Progress in Use of Cohesive Devices

Soo-Hyun Koo & Min-Chang Sung

Seoul National University, Korea

Biodata

Soo-Hyun Koo is a doctoral student (ABD) at the Department of Foreign Language Education,
Seoul National University, Korea. She has been teaching English in colleges in Seoul and its
vicinity. Her research interests include speech perception, speech production, and EAP
speaking and writing.

Email: [email protected]

Min-Chang Sung holds a Ph.D. in English Language Education from Seoul National
University, Korea. He is teaching in colleges in Seoul and its vicinity. His research interests
include academic writing, construction grammar, corpus linguistics, and grammar teaching.

Email: [email protected]

Abstract

This study reports on the development and evaluation of an academic writing course
customized to postgraduate EFL learners at a university in Korea. The curriculum was tailored
to the learners based on the results of learner writing diagnosis, which identified an urgent
demand for textual cohesion. Therefore, the instruction focused on cohesive devices such as
conjunctions, references, and lexical cohesion. The improvement in the use of cohesive devices
was assessed by analyzing types and functions of cohesive devices in pre- and post-writings.
The results found that the instruction helped the learners expand their repertoire of cohesive
devices and avoid ambiguous references. The learners, however, had trouble understanding the
discourse features of the definite article the and thus made many errors even after the
instruction. These findings may imply that, while a diagnostic analysis of learner writings is a
prerequisite for customizing academic writing courses to target learners, a multi-faceted

162
pedagogical approach which examines linguistic, pragmatic, and cognitive aspects is requested
to address problems that learners of academic writing are facing.

Keywords: academic writing; customization; learner writing analysis; cohesion; consciousness

raising

1. Introduction

English for academic purposes (EAP) is a major domain of English for specific purposes (ESP)
that focuses on the specific communicative needs and practices of academic experts (Hyland
& Shaw, 2016; Johns & Dudley-Evans, 1991). Provided that the core principle of ESP is
“tailoring instruction to specific rather than general learning purposes” (Hyland & Hamp-
Lyons, 2002, p. 2), EAP researchers and instructors have placed special importance on
academic writing competence, which is most needed for postgraduate students — a major
beneficiary of EAP programs — to excel in their research and academic careers. A variety of
components in academic writing, such as vocabulary, register, and style, have been intensively
investigated, and the results have helped EAP practitioners develop finely customized
instruction (e.g., Hu, 2007; Xudong, Cheng, Varaprasad, & Leng, 2010).

Although many of these customization-oriented approaches have been found to be effective, a

marked gap exists with regard to the exhaustive investigation of academic texts written by EAP
learners, which would provide valuable information for the development of a customized
academic writing program and the measurement of substantial gain or qualitative
improvements. Instead of analyzing genuine features of EAP learners’ writings, EAP
instruction studies have relied on indirect measurements of academic writing competence, such
as standardized English test scores, academic status, or band scores. When these indirect
methods are used, it is very difficult to identify specific difficulties of a certain learner group
and estimate learning gains for individual instructional focus (Leung, Lewkowicz, & Jenkins,
2016).

Accordingly, the purpose of the present study is to develop an EAP writing course for Korean
postgraduate students based on the analysis of their writings. This customization process has
led the instructor to implement a dual instruction mode (i.e., lecture plus conference) and adopt
the consciousness-raising approach for teaching a variety of cohesive devices (i.e., conjunction,
referencing, and shell noun). In addition, classroom materials have been tailored to the target
learners as texts directly extracted from their writings or academic journal papers in their

163
discipline were often used as samples in the class. The effectiveness of this customized
instruction is measured through a comparative analysis of pre- and post-instruction writings.

2. Previous Studies

2.1 EAP writing instruction

The primary goal of EAP programs is to help novice researchers acquire academic literacy,
which is distinguished from general literacy and broadly defined as the linguistic and social
competence to comprehend and articulate academic texts (Hyland, 2013; Wang, 2017).
Academic literacy includes a complex set of language skills or cross-cultural insights that are
required for success in academic communities, yet EAP education and research have focused
primarily on academic writing because postgraduate students — the main beneficiary of EAP
programs — are in great need of writing skills. First, they are often asked to write academically
as course requirements (Woo, 2015). In addition, postgraduate students, as novice scholars,
need to acquire strong academic writing competence to present research designs and findings
in a logical, comprehensive, and persuasive manner (Flowerdew, 1999a). Moreover, academic
writing skills are essential for publishing research papers and monographs, which play a critical
role in the academic job market and research projects (Ganobcsik-Williams, 2004). Noting the
significance of academic writing, researchers, educators, and policy makers in higher education
have endeavored to provide a variety of EAP writing courses in their institutions (Cargill, 1996;
Casanave & Hubbard, 1992; Storch & Tapper, 2009).

These EAP writing courses have primarily been guided by two different approaches: the
general approach and the specific approach. The general approach understands EAP as “a
common core of universal skills or language forms” (Hyland & Hamp-Lyons, 2002, p. 5),
thereby making EAP courses across different regions and institutions more uniform and
generic. In contrast, the specific approach gears the instruction towards the specific needs of
certain academic communities and frequently follows their nomenclatures such as ‘research
article’ and ‘lab report’ (Johns, 2003).

Although the general approach has notable logistic advantages such as finding capable
instructors and administrating college-level interdepartmental programs, the specific approach
has been argued to be better in accommodating the genuine characteristics of target learners
(e.g., stages in academic pursuit, majors, and English writing proficiency) and tailoring the
instruction and material to accurately mirror their educational needs. This is a very important
advantage of the specific approach because the success of an EAP writing course is heavily

164
dependent on the degree to which the course addresses the problems and learning needs
peculiar to the target students; therefore, many studies have noted that the specific approach is
a more appropriate framework for EAP writing courses (Benesch, 1996; Flowerdew &
Peacock, 2001; Hinkel, 2003; Reid, 2001).

Therefore, many EAP researchers have adopted the specific approach to develop EAP writing
courses, and much effort has been exerted to customize the courses to specific learning needs.
One form of effort is genre-based method, which, focuses on teaching lexical and grammatical
resources for specific academic genres (e.g., Cheng, 2008; Cunningham, 2017; Hyland, 2004).
Storch and Tapper (2009) developed such EAP writing program where international
postgraduates students learned academic linguistic conventions and applied the knowledge to
composing fundamental academic genres such as summaries, critical reviews and research
proposals. Another form of customization effort is to design EAP writing courses to be more
process-oriented. When an EAP writing instructor follows the writing processes of their
students, the instructor is able to identify both common and individual difficulties in composing
a certain type of text and provide timely and dialogic feedback by using multiple modes of
draft discussion, such as student annotation, teacher scaffolding, peer reviewing and editing,
and writing conferences (Badger & White, 2000). Lastly, many studies on customized EAP
writing programs have shown the effectiveness of raising learners’ awareness of academic
writing practices (Harmer, 2004; Silva & Brice, 2004; Swales, 1990). This consciousness-
raising method often requests EAP students to analyze texts from their own disciplines and
explore multiple features including textual patterns, authors’ intentions, rhetorical functions,
and expectations from the discipline, both holistically and implicitly. This exploratory
experience is known to promote both sensitivity and autonomy among the students (Hyland,
2013).

Despite these meaningful findings, salient limits exist when customizing EAP writing courses
to the target student. In many studies, course development and material design were based on
instructors’ personal insights and prior experiences, consequently rendering the EAP writing
courses implicitly teacher-centered, rather than learner-centered. As a result, students become
less satisfied with and have little reliance on EAP writing courses. For instance, Jin (2015)
found, in her study of lexical verb use by three L1 Chinese novice researchers in Engineering,
that all the participants questioned the effectiveness of university writing courses, evaluating
them as providing “broad introduction to academic writing in the science discipline” (p. 25).

165
The participants, instead, put more trust towards alternative sources such as their supervisors
or peer students in the lab.

Even when students’ levels and learning needs were taken into consideration for course
development, little attention has been given to analyzing their writing samples and identifying
recurring problems. Instead, previous studies have heavily relied on indirect measurements of
academic writing competence, such as general/holistic evaluation and learners’ reports
(Flowerdew, 1999b; Gosden, 1996; Hu, 2007; Storch &Tapper, 2009; Xudong et al., 2010).
These general evaluations, however, are inappropriate for addressing complex and multi-
dimensional difficulties that L2 English-speaking learners, who are the major beneficiary of
EAP writing programs, face while producing academic writings.

Hinkel (2003) reveals that L2 learners’ academic writings “frequently rely on a limited lexical
and syntactic repertoire that results in vague and less sophisticated prose relative to that of [L1
learners]” (p. 276). In addition, L2 English writers encounter cross-cultural dissimilarities in
academic conventions (Leki & Carson, 1997) and have trouble abiding western academic
writing practices, such as academic integrity (Currie, 1998; Hu, 2007; Sowden, 2005) and genre
conventions (Bitchener & Basturkmen, 2006). Even after the EAP writing instruction, L2
English writers show a slow development rate (e.g., Knoch, Rouhshad, Oon, & Storch, 2015;
Read & Hays, 2003). Although some areas, such as vocabulary, rhetorical expressions, and text
structure, displayed measurable development, areas revolving around linguistic skills like
fluency and grammatical accuracy showed little or no improvement (Shaw & Liu, 1998; Storch
& Tapper, 2009; Xudong et al., 2010). Noting these unique and diverse difficulties that L2
learners have, due attention needs to be paid to designing EAP writing programs based on more
comprehensive and diagnostic information of participating learners (Johns & Price-Machado,
2001).

Another limitation in addressing specific difficulties that EAP learners have is found in the
measurement of instruction effects. Many studies employed a test–retest design, where writing
development is gauged by using composition/band scores (e.g., Hu, 2007; Sasaki, 2007, 2009,
2011; Shaw & Liu, 1998; Storch & Tapper, 2009; Woo, 2015; Xudong et al., 2010). The
problem with band scores — regardless of their scope — is that they tend to coarsely mix up
several features of writing under a single rubric (Purpura, 2004). Considering the complex
nature of writing competence, simple increases or decreases in band scores offer limited
insights into which components of academic writing competence have improved, making it
difficult to measure how effective each instructional component is (Goulden, 1994). To date,

166
few studies on EAP writing instruction have attempted to analyze pre- and post-instruction
learner writings based on the instruction contents. If such in-depth analyses, which are already
common in the body of research on genuine features of academic texts such as logical moves,
rhetorical conventions, and metadiscourse markers (e.g., Hyland, 2005), had been more
frequently implemented when evaluating EAP writing courses, EAP professionals could have
better understood important issues in EAP writing pedagogy, such as specific difficulties for
L2 English writers, effective teaching and learning techniques, and developmental sequences
of EAP writing competence.

This study addresses these research gaps — a) absence of EAP course design based on the in-
depth analysis of learner performance and b) indirect measurement of improvements in target
learners’ academic literacy — by analyzing the participants’ academic writings and applying
the results to development as well as the assessment of an EAP writing program. To be more
specific, the present study identifies text cohesion as one of the most problematic areas found
in the writings that novice researchers in the civil engineering produced, introduces the
development of a customized curriculum on text cohesion for the target learners, and presents
their improvements measured by analyzing structural and functional features of each and every
cohesive marker in their pre- and post-instruction writings.

2.2 Cohesion in academic writing

Structures of written texts have been analyzed in terms of two important constructs: coherence
and cohesion. Although text coherence and text cohesion contribute to each other in a variety
of ways (e.g., Lorenz, 1999), the former is distinguished from the latter by the degree of
concreteness and the range of scope. In a broad definition, text coherence is related to
discourse-level unity where “the elements of the message are seen to be connected, with or
without overt linguistic connections between those elements” (Brown & Yule, 1983, p.224)
and defined as “the degree to which concepts and relations that underlie the surface text are
mutually relevant” (Kern, 2000, p. 80). On the other hand, text cohesion is more concerned
with linguistic markers (Brown & Yule, 1983), which refer to “surface-level patterns like
referential pronouns and demonstratives” (Tardy & Swale, 2009, p. 568).

Text cohesion has been of great relevance in the EAP discipline, as a single academic paper
involves a number of cohesive domains ‘‘where the INTERPRETATION of some element in the
discourse is dependent on that of another’’ (Halliday & Hasan, 1976, p. 4). Building these
interpretational links that organize a lengthy text in a connective manner, cohesive devices

167
guide the reader to arrive at a certain interpretation intended by the author, “pointing out topic
shifts, signaling sequences, cross referencing, connecting ideas [and] previewing material”
(Hyland & Tse, 2004, p. 158). Therefore, the process of writing academic papers requires the
author to unceasingly consider “the range of possibilities that exist for linking something with
what has gone before” (Halliday & Hasan, 1976, p. 10) and to determine whether and which
cohesive devices to use.

According to the pioneering work by Halliday and Hasan (1976), cohesion is achieved by five
major types of cohesive devices — conjunction, reference, lexical cohesion, substitution, and
ellipsis, yet many EAP studies have been conducted on three of them, namely, conjunction
(e.g., Lee, 2004; Lorenz, 1999), reference (e.g., Oh, 2009; Petch-Tyson, 2000), and lexical
cohesion (Aktas & Cortes, 2008; Gray, 2010). Conjunctions create cohesion by “[expressing]
certain meanings which presupposes the presence of other components in the discourse”
(Halliday & Hasan, 1976, p. 226). An exhaustive list of these discourse structuring meanings
is provided in Halliday and Matthiessen (2014, p. 612). The authors discretely and semantically
categorized the meanings of conjunctions into three major semantic types — i.e., elaborating,
extending, and enhancing — which are further branched into nine sub-types: appositive and
clarifying; additive, adversative, and varying; matter, manner, spatio-temporal, and causal-
conditional. Alongside this functional semantic classification, which focuses on conjunctive
meanings, structural classification has been also influential wherein conjunctions are
categorized by structural properties of connected components (Celce-Murcia & Larsen-
Freeman, 1999), namely, coordinate conjunctions (e.g., and), subordinate conjunctions (e.g.,
when), correlative conjunctions (e.g., either … or …), and conjunctive adverbs (e.g., however).
Both types of classification have been considered in the present study because they provide a
comprehensive basis of conjunctions on which a variety of difficulties that L2 learners
experience can be addressed.

Another contributor to textual cohesion is reference devices, which link to a preceding or

following notion, with one such device being textual references (Halliday & Hasan, 1976).
When the author wants to mention once again an element that has been previously mentioned
and thus has a given-information status, there are three possible types of textual reference —
namely, personal pronouns (e.g., it), demonstratives (e.g., this/that), and the definite article the
(Reid, 1992; Swales & Feak, 2004). The choice of a particular reference pertains to several
important aspects in academic writings. For example, the choice between proximal (i.e., this
and these) and distal demonstratives (i.e., that and those) is often related to the writer’s two-

168
fold evaluation of the referent: “1) the relative amount of information that the [writer] presumes
the [reader] to have with respect to the referent and 2) the relative importance of the referent
itself to the [writer]” (Strauss, 1993, p. 404). Such evaluative and interactive functions of
textual references guide readers to give selective attention to novel and important notions and
ultimately “grasp the significance of particular information in the way the writer intends”
(Hyland, 2005, p. 76).

Along with conjunctions and references, lexical cohesion — i.e., text continuity established by
the choice of certain words — has been a central issue in EAP research. In particular, the focal
attention has been placed on shell nouns, which “enclose or anticipate the meaning of the
preceding or succeeding discourse” (Aktas & Cortes, 2008, p. 3). The following excerpt depicts
how the shell noun “This process” encloses the information in the preceding sentence.

(1) A t-test is applied to test for a difference in the mean degree of risk
reduction. This process is conducted separately for each risk proxy.
(Adopted from Aktas & Cortes, 2008, p. 11)

The importance of shell nouns has been acknowledged in many EAP studies (e.g., Aktas &
Cortes, 2008; Petch-Tyson, 2000; Schmidt, 2000). Since academic writings often contain
complex information developed over a stretch of discourse compared to other written genres,
it is essential for the writer to “[condense] the content of previous propositions into a nominal
phrase which can then be included in new propositions, thus creating a rhetorical effect of
brick-building” (Petch-Tyson, 2000, pp. 45-46). When this process of “encapsulating or
packaging” (Francis, 1994, p. 86) is absent or too general, textual cohesion becomes obscured,
as in the following “sloppy use of this” (Petch-Tyson, 2000, pp. 61-62):

(2) A lot of people are enrolled in a political party because that party is
“the one who could help them”, it is always difficult to know exactly
who believes in their party and who expects to profit by it. This is also
made clear very early to the people and many political parties gain a
lot of votes thanks to that situation. (French-speaking learner)
(3) Consider scientists lacking any source of imagination whatsoever.
This consideration is completely out of the question. (Swedish-
speaking learner)
The aforementioned frame of cohesive devices has been extended to the analysis of academic
texts written by L2 learners of English. This line of research has identified idiosyncratic

169
features in their academic writings, including the limited repertoire of cohesive devices
(Hinkel, 2001; Lu, Li, & Ottewell, 2016); the overuse of certain devices (Green, Christopher,
& Kam Mei, 2000; Shaw, 2009); and semantic and syntactic deviations from native norms
(Gardezi & Nesi, 2009; Lee, 2004; Oh, 2009). As will be noted later, the participants in the
present study displayed similar problems. The present study, however, aims to further this line
of research by applying the diagnosis into an actual EAP program and examining the
improvements in the use of cohesive devices via the comparative analysis of pre- and post-
instructional writings.

3. METHOD

3.1 Participants

An EAP writing course, entitled Technical Writing Course for Engineering Graduate Students,
was developed for graduate students in the department of civil engineering. The class consisted
of four MA students and two Ph.D. students, but the present study focuses only on the four MA
students for the following reasons. First, the MA students paid special attention to the design
of the instruction; therefore, the EAP writing course was more customized to the MA students.
Second, the MA students had received little or no EAP writing instruction before and thus had
a clean slate when learning the instructional contents. Therefore, whatever improvement was
observed during and after the course, the instruction was most likely a deducible cause for that
progress. Finally, the four MA students showed homogenous demographics, i.e., L1 Korean
speakers in their 20s.

3.2 Instruction

3.2.1 Preparatory stage: learner writing analysis

The course objective was to help the students enhance academic writing competence and
produce a research paper that most suits the professional norm of the civil engineering
academia. To this aim, the instructor first analyzed the students’ written samples to categorize
frequent errors and identify the deficits in their academic writing competence. The writing
samples were three- to five-page research paper which reported their recent research or
provided an overview of their research areas. Students handed in these works in February, one
month before the course started. As the writings were redrafted by the instructor, it was
observed that many corrections occurred around textual cohesion. While every student had

170
little trouble writing individual grammatical sentences in English, they displayed biased over-
or underuse of certain cohesive markers:

 Overuse of certain conjunctions (e.g., and, but, so, therefore)

 Overuse of sentence-initial coordinate conjunctions (e.g., And the
researcher … )
 Underuse of sentence-medial conjunctions (e.g., The results, however,
revealed …)
 Underuse of demonstrative pronouns and shell nouns (e.g., this
approach, this method)
These unnatural usages of cohesive devices are detrimental to academic writings in the target
discipline, i.e. civil engineering. Hyland (2013, p. 106) identified that a major component of
academic writing in science and technology is “activity-based skills” such as describing
procedures, defining objects, and planning solutions. This convention applies to academic
writings in civil engineering. For example, research in civil engineering often suggests a
mathematical model that makes the best fit to solve identified problems. This modeling is an
intricate process, involving frequent references to previously mentioned procedures,
identification of each symbol in equation, and validation rationale. These activities can be
successfully performed only when cohesive devices are effectively employed, as shown in the
following except from a work of civil engineering experts:
The state of Es,i is defined as Es,i = 1 if {Es,Pa(i) = 1 ∩ {Ci = 1} = 0
Otherwise where Es,Pa(i) defines the state of the SPE node that is parent to Es,i;
Es,i=1 indicates that the node is in the survival state and Es,i=0 indicates its
failure (we use this Boolean notation throughout this paper). Thus, for a series
system, the BN formulation takes the form shown in Fig.8a. The state of node
Es,1 is equal to the state of node C1. Es,2 is in the survival state only if Es,1 is in
the survival state and C2 is in the survival state. This pattern continues such
that Es,Nc is in the survival state only if both Es,Nc–1 and CNc are in the survival
state. Consequently, the state of Es,Nc describes the state of the SPS (i.e., it
indicates whether all components in the MLS have survived) and, therefore,
that of the series system. The state of node Ssys is equal to the state of Es,Nc in
Fig.8a (Bensi, Der Kiureghian, Straub, 2013, p. 204).

This one paragraph-long excerpt contains a number of cohesive devices such as conjunctions
(and, thus. only if, both X and Y, consequently, therefore), references (the, its, this, that), and

171
other cohesive devices (semi colon, i.e.). In addition, other expressions (e.g., is parent to, is
equal to, indicates, in Fig.8a) also contribute textual cohesion. Given this important role of
cohesive devices in civil engineering, a variety of cohesive features were instructed under a
specific course module and teaching methods.

3.2.2 Course organization: alternation of two instructional modes

During the instruction period, two different teaching modes were complementarily used: a bi-
weekly lecture session and a monthly individual conference. A single phase consisted of two
lecture sessions and one individual conference in-between. Such a dual mode was intended to
conduct formative evaluation of the students’ progress in a repetitive and personal manner and
(re)design the forthcoming class to be more customized to their unique problems and
improvements. The major objective of the lecture mode was to teach academic conventions for
cohesion while individual conference sessions aimed to provide dialogic and timely reaction
to students’ emerging strengths and weaknesses.

This dual mode revolved around the monthly written assignment that the learners handed in
throughout the semester. It was a three- to five-page semi-research paper on major topics in
their discipline such as annual bridge maintenance prioritization, lifeline networks against
natural hazards, and Bayesian parameter estimation. The instructor analyzed the assignment to
identify cohesion-related errors that the learners commonly and frequently produced. Some
notable errors were presented as materials for group- or whole-class discussion in a lecture
session, in attempt to train students to notice the problem and collaboratively develop possible
solutions. During the conference sessions, the instructor provided more customized feedback
to each learner by explicitly identifying the actual errors that the learner repeatedly produced
in the assignment drafts and highlighting relevant EAP knowledge or skills that had been taught
in the lecture sessions. This helped the learners to become aware what their problems are and
how to resolve them.

3.2.3 Teaching method: consciousness-raising approach

The aforementioned problems that the learners had in academic writing — i.e., overuse or
underuse of certain cohesive markers — are, in fact, common even among advanced L2
learners of English (Crewe, 1990). This problem has been frequently attributed to lack of
awareness: not only are L2 learners unaware that there are multiple, competing ways to express
a particular cohesive meaning, but they are also insensitive to genre- or domain-specific
features in selecting the most appropriate one. Therefore, the consciousness-raising approach

172
(Swales, 1990) was employed as the underlying principle of the instruction, and a variety of
consciousness-raising tasks were developed from the written assignments and journal article
samples.

These consciousness-raising tasks were expected to “promote learner awareness and

autonomy” (Hyland, 2013, p. 107) and to induce learners to explore many cohesive devices
under certain discourse contexts. This effect seems significant to the teaching of textual
cohesion in academic writings, given the following two. First, academic writing is a “non-
linear, exploratory, and generative process whereby writers discover and reformulate their
ideas” (Zamel, 1983, p. 165), which demands the author’s conscious and constant efforts to
improve the text. Second, L2 learners unconsciously, and thus repeatedly, make similar types
of cohesion-related errors (Yoon, 2006). Only when they become aware and conscious of
appropriate selection among a number of cohesion markers, the repeated errors can be avoided.

Consciousness-raising was guided through the two phases: awareness of problems and
awareness of solutions. First, the learners were provided a text which had been written by one
of them. Then, they were told that the text had a problem in cohesion and asked to identify
what the problem is. Exploring diverse aspects of textual cohesion and referring to scaffolding
cues (e.g., types and locations of errors), the learners usually succeeded in figuring out the
problem. Second, the learners were requested to correct the identified problem. In this task,
they were encouraged to examine various alternatives for enhancing textual cohesion, which
was intended to increase their awareness of multiple ways to organize ideas and concepts with
the propositional meanings maintained. Figure 1 provides an example of consciousness-raising
activities.

[Exercise] Change the function of conjunction: from additive to spatio-

temporal.
We need to introduce some set operation rules such as union and intersections,
and also need to understand the operations of probabilities in dependent and
independent situations.
- Step 1. Split the sentence into two.
- Step 2. Take out also from the sentence.
- Step 3. Insert a more appropriate conjunction in the right location.

173
Figure 1: Consciousness-raising Activity on Conjunctive Function

The sample sentence, which was extracted from a student’s assignment, has several errors in
textual cohesion. The instructor asked the learners to detect and correct the errors in-class.
Presenting examples from the written assignments that the students handed in, instead of those
from ESL/EFL writing textbooks, rendered course contents more relatable and engaging to the
students and led them to pay a focal attention to the consciousness-raising tasks.

3.2.4 Instructional contents

The instructor provided four lecture sessions on cohesive devices under three lecture themes,
namely smooth transition, inter-paragraph transition, and old information reappearing
technique (See Table 1).
Table 1: Curriculum Organization and Topic Coverage Overview
Date Theme Lesson content
 Structural category of conjunctions
Mar Smooth (coordinate, subordinate, correlative, and adverbial)
26th transition 1  Functional category of conjunctions
(adversative, causal, additive, manner, etc.)
 Connection level of conjunctions
Apr Smooth  Frequently occurring errors regarding connection levels
9th transition 2  Conjunction location (initial, medial, and final)
 Conjunction function: usage practice
 Definition of new and old information
Apr Inter-paragraph
 Examples of smooth transition between new and old
30th transition
information
 Shell noun: definition and types
Old
May  [this/that+ shell noun] combination: usage patterns
information
14th  Pronoun: types and usage patterns
reappearance
 Peer review exercise

The first lesson on cohesion, conducted on March 26, focused primarily on introducing the
structures and functions of conjunctions. These two subtopics were taught to address the
problem of limited conjunction repertoire. The instructor offered a range of alternative options
to the students, illustrating detailed usage patterns including location, connection level, and
function. This structural-functional approach classifies conjunctions into four groups:
coordinate conjunctions, subordinate conjunctions, correlative conjunctions, and conjunctive
adverbs (Celce-Murcia & Larsen-Freeman, 1999; Hacker, Sommers, Jehn, Rosenzweig, & Van
Horn, 2004). For each conjunction type, its structural properties and examples were provided
as a referral resource that the learners can draw out to escape the narrow loop of repertoires
while drafting or editing their papers (See Table 2).

174
After conducting two subsequent lecture series on conjunction, the instructional focus was
moved onto referring device usage, which is directly related to tackling the underuse of the
demonstrative pronoun and shell noun (e.g., this approach and this method).

Table 2: Structural-functional Classification of Conjunctions

Type Coordinate Subordinate Correlative Conjunctive
conjunctions conjunctions conjunctions adverbs
 Used between  Used in
two independent  Used between a  Two features must compound,
ideas within the main idea and a appear in a sentence complex,
sentence (S1 and subordinate idea structure compound-
Structural- S2) within the sentence  Used in complex complex, and
functional  Used in (S1 and S'1) and compound- simple sentences
property compound  Used in complex complex sentences (i.e., having a
sentences (i.e., sentences (i.e., (i.e., having two single
having two having subordinate independent independent
independent clauses) complex clauses) clause)
clauses)
Example and, but, or, nor, after, although, as, either (a) or (b), accordingly, also,
for, so, yet as if, because, neither (a) nor (b), anyway, besides,
before, even both (a) and (b), certainly,
though, if, in order not only (a) but also (b) consequently, …
that, …

3.3 Data collection

The researchers collected two essays from each of the participating learners over the course of
the semester — one before the instruction and the other after the instruction — to determine
the instructional effects on the learners’ use of cohesive devices. The former was a three-page
paper explaining their field of interest to non-majors, but the latter was a 13- to 17-page
research paper that each student wrote on the research project conducted in their laboratory.
Eight essays in total were collected from the four students and subjected to analysis. The pre-
instruction written sample were 3,903 words, while the post-instruction written sample was
11,113 words.

It should be acknowledged that the pre-instruction and the post-instruction written sample did
not exactly correspond in genre. Such discrepancy seems inevitable because the learners did
not have L2 academic literacy to handle a 10-page research paper in the beginning stage of the
instruction. As complementary measures for this discrepancy, individual interviews between
the instructor and the learners were conducted to delve deeper into the intention of the novice
writers in the selection of particular cohesive devices. The interview was semi-structured

175
according to the results of the writing analysis. The instructor focused on evident improvements
and/or lingering problems, asking questions such as:

(a) Were there any conjunctions you tried to avoid using in the post-instruction essays?
(b) On what basis did you choose to insert the cohesive device?

3.4. Data coding and analysis

Cohesive devices in the learners’ writings were manually coded under the following criteria,
which were adapted from Halliday and Hasan (1976) and Halliday and Matthiessen (2014):

 Expressions used as cohesive devices: e.g., the, him, and however

 The type of cohesive devices
- Conjunctions: coordinate (e.g., and), correlative (e.g., not only … but also
…),
subordinate (e.g., when), and conjunctive adverbs (e.g., however)
- References: definite article the, demonstratives (e.g., this and that), and
personal pronouns (e.g., he and it)
- Shell nouns: nouns which lexicalize their immediate context (e.g.,
interregional income disparities … have surged only in the past two
centuries. This trend … [Gray, 2010, p. 170])
 Error description: only for inappropriate use of cohesive devices

For conjunctions, additional coding was conducted to note their location and function, which
was highly relevant to the lessons on conjunctions:

 Location: sentence-initial (e.g., However, the man revealed that…),

sentence-medial (e.g., finds the weakest structure, and analyzes), and
sentence-final (e.g., I like him though.)
 Function: appositive (e.g., in other words), clarifying (e.g., in short),
additive (e.g., moreover), adversative (e.g., but), varying (e.g.,
instead),
matter (e.g., in that respect), manner (e.g., likewise),
spatio-temporal (e.g., finally), and causal-conditional (e.g., therefore)

Both of the researchers in the present study participated in the coding process to ensure that
every writing sample went through the inter-rater analysis. When the two rates differed in

176
categorizing certain cohesion devices, they discussed the issue to arrive at a mutual agreement.
This cross-coding procedure identified 311 non-error and 117 error instances of cohesive
devices in the pre-writings and 735 non-error and 167 error instances in the post-writings.

For a comprehensive and multi-dimensional analysis, the coding data were input into MS Excel
(ver. 2013) spreadsheets and analyzed primarily via the pivot table function, which allowed the
researchers to easily compute absolute frequencies and percentages of different types and
subtypes of cohesive devices. The relative frequency of each cohesive device was then
computed in terms of parts per thousand (ppt) by using the total number of words in the pre-
writings (N=3,909) and the post-writings (N=11,113). This normalizing process allowed the
researchers to examine which (usages of) cohesive devices the learners more or less frequently
produced in the post-writings.

4. Findings

As previously mentioned about the diagnostic phase, the researchers identified the following
cohesion-related problems in the pre-instruction writings:

 Overuse of certain conjunctions

 Overuse of sentence-initial coordinate conjunctions
 Underuse of sentence-medial conjunctions
 Underuse of demonstrative pronouns and shell nouns

In what follows, it is examined whether and to what extent the instruction, which was
developed to address cohesive devices in academic writing, was effective.

4.1 Conjunctions

The comparative analysis of non-error conjunctions between the pre- and post-instruction
written accounts revealed that the learners used conjunctions less frequently in the post-writing.
The relative frequency of conjunctions in the pre-writings was 40.23 ppt, compared to 25.56
ppt in the post-writings (see Table 3). The Chi-square test conducted based on the total word
tokens also confirmed this gap as a significant decrease in the learners’ use of conjunctions:
χ2(df1)= 21.656, p < .001, odds ratio = 1.60.

177
Table 3: Frequency Distribution across Subtypes of Conjunctions in Pre- and Post-
writings
Pre-writing (N=3,909) Post-writing (N=11,113).
Type
Abs. Rel. Percentage Abs. Rel. Percentage
CRD. 53 13.58 33.8% 60 5.40 21.1%
SUB. 45 11.53 28.7% 112 10.08 39.4%
CAD. 58 14.86 36.9% 111 9.99 39.1%
CRR. 1 0.26 0.6% 1 0.09 0.4%
Total 157 40.23 100% 284 25.56 100%
Note. Abs = Absolute frequency; Rel = Relative frequency; CRD = CooRDinate conjunction;
SUB = SUBordinate conjunction; CAD = Conjuntive ADverb; CRR = CoRRelative conjunction.

For subtypes of conjunctions, significant decreases were observed with coordinate

conjunctions (from 13.58 to 5.40 ppt: χ2[df1]= 25.788, p < .001, odds ratio = 2.53) and
conjunctive adverbs (from 14.86 to 9.99 ppt: χ2[df1]= 6.113, p < .05, odds ratio = 1.49), while
the frequency changes of the other types of conjunctions were found to be insignificant (for
subordinate conjunctions, p = .45; for correlative conjunctions, p = .44).

Regarding the significant change in the frequencies of coordinate conjunctions, the most
noteworthy pattern was the drastic decrease of sentence-initial coordinates (see Table 4). The
relative frequency of sentence-initial coordinates in the pre-writings was 4.09, while that in the
post-writings was 0.45 (χ2[df1]= 27.495, p < .001, odds ratio = 9.13).

Table 4: Frequencies of Sentence-initial and Medial Conjunctions in Pre- and Post-

writings
Pre-writing Post-writing
Location Initial Medial Initial Medial
TYPE Abs. Rel. Abs. Rel. Abs. Rel. Abs. Rel.
CRD. 16 4.09 37 9.47 5 0.45 55 4.95
SUB. 14 3.58 31 7.93 39 3.51 73 6.57
CAD. 47 12.02 11 2.81 71 6.39 40 3.60
CRR. 0 0.00 1 0.26 1 0.09 0 0.00
Total 77 19.70 80 20.47 116 10.44 168 15.12

In the pre-writing, the learners frequently placed coordinate conjunction (e.g., so, but) at the
initial position of a sentence, as in (4) and (5):

(4) … is inevitable. So we introduce one of a method … (Pre-writing; Learner A)

(5) … has calculated. But if it becomes the problems … (Pre-writing; Learner C)

178
This usage pattern is hardly the best lexical choice, especially for the genre of academic writing,
as these conjunctions list different ideas in a simple linear fashion, without coherently
specifying the logical relationship among the ideas. However, after the instructional treatment,
this problem was fairly solved: the relative frequency of initial coordinates significantly
decreased. The use of academically less appropriate coordinate conjunctions seldom appeared,
and most of the coordinate conjunctions in the post-writings were sentence-medial, as in (6)
and (7):

(6) … not as same as my assumption so one can may say … (Post-writing; Learner
A)
(7) … stable declining tendency as it should be, but the declining speed was not
always uniform (Post-writing; Learner C)

This finding was corroborated by the interview data. The following excerpts demonstrate that
the learners have raised awareness towards their less effective use of initial coordinates:

(8) Learner A: Since the instructor told me that I used this kind of expression (i.e.,
initial coordinate) too often, I tried not to use it when writing articles.
(9) Learner D: After taking this course, I tried to avoid using conjunction and start
with simple sentences. [Both translated by the researchers]

Both (8) and (9) indicate that the instructor’s constant feedback helped the learners notice the
erroneous usage of conjunctions and become aware of the appropriate solution. Consequently,
the learners tried a variety of conjunctions (or other cohesive markers) when connecting clauses
and sentences.

Thanks to their increased awareness and follow-up efforts to consciously substitute the non-
academic coordinates with alternative expressions, the type frequency of conjunctive adverbs
significantly increased from 26 to 42, as 28 conjunctive adverbs that were not used in the pre-
writings started to appear in the post-writings. In sum, the instruction led the learners to use
coordinate conjunctions less frequently whereas their repertoire of conjunctive adverbs seemed
to increase. This substitution pattern was found to be more evident for certain functions. Table
5 shows absolute frequencies and percentages of coordinates, subordinates, and conjunctive
adverbs for the five most common functions — namely, causal-conditional, spatio-temporal,
additive, adversative, and appositive.

179
Table 5: Absolute Frequencies and Percentages of Subtypes of Conjunctions for Five
Functions
Pre-writing Post-writing
Function CRD SUB CAD Sum CRD SUB CAD Sum
16 26 13 55 9 42 21 72
Causal-conditional
(29.1) (47.3) (23.6) (100) (12.5) (58.3) (29.2) (100)
4 10 10 24 6 38 28 72
Spatio-temporal
(16.7) (41.7) (41.7) (100) (8.3) (52.8) (38.9) (100)
31 0 8 39 35 0 18 54*
Additive
(79.5) (0) (20.5) (100) (64.8) (0) (33.3) (100)
2 6 7 15 10 25 26 61
Adversative
(13.3) (40.0) (46.7) (100) (16.4) (41.0) (42.6) (100)
0 0 15 15 1 0 5 6
Appositive
(0) (0) (100) (100) (16.7) (0) (83.3) (100)
53 42 53 148 61 105 98 265
Total
(35.8) (28.4) (35.8) (100) (23.0) (39.6) (37.0) (100)
Note. The first number in each slot is the absolute frequency, while the number in the below parentheses
is its percentage to the total frequency of the function.
* Since there was only one case where a correlative conjunction used for these functions, that subtype
was excluded from the table for clear demonstration, though included in the sum of the additive
conjunctions in the post-writings.

It is noteworthy that the most frequent function, i.e., causal-conditional, made greater use of
subordinate conjunctions (from 47.3% to 58.3%) and less use of coordinate conjunctions (from
29.1% to 12.5%) in the post-writings. A similar pattern was observed for the second most
frequent function, i.e., spatio-temporal: subordinate conjunctions increased (from 41.7% to
52.8%), but coordinate conjunctions decreased (from 16.7% to 8.3%). For example, the
absolute frequencies of the causal-conditional subordinates since and if increased by 6 and 7
tokens respectively, while that of the causal-conditional coordinate so decreased by 10 tokens.
Likewise, the absolute frequency of the spatio-temporal subordinate when increased by 18
tokens, while the sole spatio-temporal coordinate (i.e., and) did not show such drastic increase,
with its absolute frequency increasing only by two tokens. This finding reveals that the learners
used a more diverse repertoire of conjunctions which display very specific functions rather than
relied on generic meanings of coordinate conjunctions that broadly describe the relationship
between phrases or sentences. This may indicate that the learners began to reflect their
blossoming awareness toward a variety of inter-sentence relationships, beyond simple and
visible ones.

180
4.2 References

The comparative analysis of references, including the definite article the, personal pronouns
(e.g., it), and demonstratives (e.g., this), identified that the definite article was used much
more frequently in the post-writings. The relative frequency increased from 17.91 to 25.56
(see Table 6). The Chi-square test confirmed this increase as significant: χ2(df1)= 7.351, p <
.01, odds ratio = 0.70.

Table 6: Absolute and Relative Frequencies of References

Pre-writing Post-writing
Type
Abs. Rel. % Abs. Rel. %
Definite Article 70 17.91 51.9 284 25.56 73.0
Personal Pronouns 37 9.47 27.4 45 4.05 11.6
Demonstratives 28 7.16 20.7 60 5.40 15.4
Total 135 34.54 100 389 35.01 100

Another interesting finding was that, as the use of the definite article increased, the use of the
other types of references decreased. The relative frequency of personal pronouns decreased
from 9.47 to 4.05 (χ2[df1]= 15.625, p < .001, odds ratio = 2.35), and that of demonstratives
decreased, although not significantly so, from 7.16 to 5.40 (χ2[df1]= 1.545, p = 0.213, odds
ratio = 1.33). These results indicate that, when referring to a previously mentioned concept,
the learners become more reliant on the definite article rather than personal pronouns and
demonstratives. This change should be viewed as appropriate because the pre-writings
showed many problematic cases where the reference of a personal pronoun or a demonstrative
was not clear:

(10) To improve prediction and inference, first you have to know and study the
variety of methods to make predictive model. There are already lots of tools
however this subject is start with the advent of computers, in other words it is a
recently developed area (Pre-writing; Learner A)
(11) And then from the failure probabilities of the systems like pier, deck and
bearing joint etc., we can finally draw the probability of the whole bridge. This
is the basic concept of probability based analysis and it is foundation rule
throughout various application of reliability engineering (Pre-writing; Learner C)

In (10-11), it is unclear which entity the underlined references, it and this, refer to. In contrast,
the construction of the + NP does not evoke such a problem, as the noun phrase clarifies which

181
constituent is being referred to and provides a clearer connection than personal pronouns or
demonstratives. However, the heavy reliance on the definite article brought about other side
effects: the learners produced a number of errors regarding the use of the in the post-writings
(see Table 7). When compared to the other types of references, which had few errors in the
post-writings, the frequency of erroneous the was still considerable.

Table 7: Absolute Frequencies of Legitimate and Erroneous Uses of References

Pre-writing Post-writing
Type
Non-error Error Non-error Error
Definite Article 70 (17.91) 66 (16.88) 284 (25.56) 140 (12.60)
Personal Pronouns 37 (9.47) 9 (2.30) 45 (4.05) 4 (0.36)
Demonstratives 28 (7.16) 6 (1.53) 60 (5.40) 4 (0.36)
Note. The first number in each slot is the absolute frequency, while the number in the next parentheses
is its relative frequency.

The prevalence of the-related errors might suggest that instruction on cohesive devices was
of little effect in helping learners use the definite article in a grammatically correct way. Most
errors with the definite article were related to adding an unnecessary the to a concept that
appeared for the first time, as in (12), and to a concept in which the referent is unclear, as in
(13):

(12) Another difficulty arises from the multiple conflicting objectives in optimization
problems. (Post-writing; Learner B)
(13) Using Bayesian parameter approach, however is not necessary to use to express
the probabilistic model considering the given condition at the research. (Post-
writing; Learner A)

This problem was not really unexpected, as a number of studies on second language learners
of English have presented the definite article the as one of the most difficult units in English
(Dulay, Burt, & Krashen, 1982; Liu & Gleason, 2002). In fact, the definite article the is not
easy either to learn or to teach. First, this linguistic unit is not restricted to a single semantic
and/or pragmatic feature. According to Hawkins (2015), the definite article the is associated
with at least six representative usage types: (a) anaphoric use (e.g., I bought a book. I want to
give the book to my mom); (b) associative anaphoric use (e.g., Kate was driving a car, but the
engine suddenly broke down); (c) immediate situation use (e.g., Pass me the bucket, please);
(d) larger situation use (e.g., the town clerk [in a local newspaper]); (e) noun phrases with
explanatory modifiers (e.g., Bill was amazed by the fact that there is so much life on earth);

182
and (f) noun phrases with non-explanatory modifiers (e.g., We share the same secrets). As the
examples show, only the first two usage types — anaphoric and associative anaphoric use —
are relevant to textual cohesion, whereas the other four usage types are far less relevant.
Therefore, the instruction on textual cohesion alone might have played a very limited role for
teaching the definite article the.

Another difficulty for using the definite article the is found in applying the concept of
definiteness or identifiability to (virtual) writer-reader interaction. The decision as to whether
or not to use the definite article in front of a noun phrase should be based on both the writer’s
and readers’ ability to identify what the noun phrase indicates. If the writer wants to use the, it
should be ensured that readers are also able to identify the referent of the noun phrase.
However, the awareness of readers seemed to be lacking when the learners used the definite
article, as shown in the following responses to an interview question on the use of the definite
article:

(14) Learner C: I use the when mentioning a word that has been mentioned once
beforehand, but I do not put an awful amount of deep thought
besides that.
(15) Learner D: I use the when I feel that the expression seems original and specific,
and when I re-mention a term. [Both translated by the researchers]

This phenomenon highlights an important aspect of academic writing: academic writing is a

highly interactive process where the writer should be aware of readers’ existence. The
instruction in the present study, however, was limited in addressing this discourse feature of
academic writing since it focused on providing linguistic resources that the learners can refer
to when writing academic texts. This instructional limitation may suggest the need for a new
dimension of customization: when teaching discourse-oriented linguistic units such as a and
the, EAP writing courses should go beyond teaching generic academic norms and involve
learners in interactive processes of writing by examining whether and to what extent students
are discourse-oriented.

4.3 Shell nouns

It should be noted that references are used not only for previously mentioned nominal
elements, but also for clausal or beyond-clausal contents. This function was frequently
expressed in forms of “definite article/demonstratives + shell noun” (e.g., this assumption).

183
Table 8 shows the absolute and relative frequency of shell nouns preceded by either the
definite article the or a demonstrative (e.g., this and these).

Table 8: Frequencies of Shell Nouns for Its Preceding References

Pre-writing Post-writing
Type Abs. Rel. Abs. Rel.
the + Shell Noun 7 1.79 24 2.16
Demonstrative + Shell Noun 10 2.56 30 2.70
Others + Shell Noun 1 0.26 0 0.00
Total 18 4.60 54 4.86

Although the small increase in the relative token frequencies of shell nouns, from 4.60 to 4.86,
was found to be insignificant (χ2[df1]= 0.0392, p = .843, odds ratio = 0.95), a drastic change
was observed in the type frequency of shell nouns. The type frequency of shell nouns
increased from 14 to 45, as the post-writings were much lengthier than the pre-writings. The
list of shell nouns in student samples are shown below:

 Shell nouns extracted from pre-writings: analyses, area, criteria, decision, impact,
kind, mechanism, method, model, modeling, phenomenon, process, suspicion,
technologies
 Shell nouns extracted from post-writings: context, algorithm, analysis, application,
approach, assumption, candidates, case, chronic ailment, comment, concept,
declining speed, discrepancy, drawback, flexibility, fluctuation, form, idea,
inaccuracy, incorporation, issue, issues, measure, method, methodology,
mode value, model selection, observations, point of view, poor assumption,
prior description, problem, procedure, process, results, sense, spectrum,
suggested constraint, system-failure-sequence, tendency, tragedy, value,
variance, work

In addition to the quantitative change, a qualitative change was observed in the use of shell
nouns. The underlined shell nouns in the list above (e.g., drawback and tragedy) display
evaluative and interpretive connotations and thus carry the writer’s subjective characterization
of the preceding notion, as shown in the following excerpts.

(14) However such equivalent limit-state surface tend to be highly non-linear which
makes FORM or SORM ineffective. To avoid this problem, in the original

184
 paper of Lee and Song, authors introduced a dummy index. (Post-writing;
Learner C)
(15) However, even advanced models may show lack of accuracy no matter how
complicated procedures and how advanced techniques they may use. To
overcome these drawbacks, research efforts have been made to develop
probabilistic shear strength models (Post-writing; Learner A)

This qualitative change may indicate that the learners began to express their authorship as
their repertoire of academic devices expanded. This may also mean that, after the customized
instruction, the learners appeared to be more competent in academic writing — enough so to
go beyond the mode of observation or description and step toward the mode of evaluation or
validation.

5. Conclusion

The aim of this study was to develop an EAP writing program wherein the teaching of cohesive
devices is customized to the context of L2 English-speaking postgraduates in civil engineering.
The customization process was based on an in-depth analysis of pre-instruction writings. The
results revealed marked improvements in the post-instruction writings, such as an expanded
repertoire of cohesive devices, diverse structural patterns, and unambiguous references.
Learners became more sensible in selecting the cohesive devices most appropriate for the
context. This improvement allowed them to exert their authorship and make a meaningful
transition from observer to evaluator. These findings highlight an important role of in-depth
writing analysis that provides diagnostic information of the target learners and guides the
instructor to design a tailor-made EAP writing course. When rendering an EAP writing course
focused on specific problems that a target learner group shares, it is common to supply
appropriate linguistic and rhetorical resources that the learners can use to remedy their
problems. However, this supply-based approach is often inappropriate, especially when
teaching L2 English writers an EAP component related to a variety of linguistic systems. For
example, it was argued that the learners in the present study had trouble learning the definite
article the because its usage pattern is multi-dimensional. Accordingly, EAP writing programs
should be based on a more comprehensive understanding of linguistic properties and
interlanguage systems. The present study also suggests that consciousness-raising is an
effective teaching method for addressing specific problems among learners. Step-wise
guidance using excerpts from learner essays was found to enhance the students’ awareness of

185
both problems and solutions. Nevertheless, additional set of effort should be made to help them
aware of readers.

The present study, however, has several underlying concerns that need to be addressed in future
studies. First, the limited number of the participants — although it allowed the researchers to
conduct a thorough investigation — made it very difficult to generalize the findings to other
teaching and learning contexts. The results would have been more reliable and generalizable if
a greater number of participants from different academic disciples had been included. Second,
an emic perspective (i.e., the perspective of insiders) was not fully addressed as both
researchers are professionals in applied linguistics, not civil engineering. It is advisable for
further research to invite experienced scholars from a particular field to assess research papers
written by novice writers in the same field. Finally, it would be interesting to track the long-
term development of academic writing competence. Provided that improving EAP writing
skills and acquiring a membership in certain academia are often life-long processes, the
longitudinal investigation of the relationship between experiences of learning and practicing
EAP writing skills, on the one hand, and EAP writing competence and academic achievements,
on the other hand, would make a meaningful contribution to the academic writing research.

References

Aktas, R., & Cortes, V. (2008). Shell nouns as cohesive devices in published and ESL student
writing. Journal of English for Academic Purposes, 7(1), 3–14.
doi:10.1016/j.jeap.2008.02.002

Badger, R., & White, G. (2000). A process genre approach to teaching writing. ELT Journal,
54(2), 153–160. doi:10.1093/elt/54.2.153

Benesch, S. (1996). Needs analysis and curriculum development in EAP: An example of a

critical approach. TESOL Quarterly, 30(4), 723–738. doi:10.2307/3587931

Bensi, M., Der Kiureghian, A., & Straub, D. (2013). Efficient Bayesian network modeling of
systems. Reliability Engineering & System Safety, 112, 200–213.
doi:10.1016/j.ress.2012.11.017

Bitchener, J., & Basturkmen, H. (2006). Perceptions of the difficulties of postgraduate L2

thesis students writing the discussion section. Journal of English for Academic
Purposes, 5(1), 4–18. doi:10.1016/j.jeap.2005.10.002

Brown, G., & Yule, G. (1983). Discourse analysis. Cambridge: Cambridge University Press.

186
Cargill, M. (1996). An integrated bridging program for international postgraduate students.
Higher Education Research and Development, 15(2), 177–188.
doi:10.1080/0729436960150204

Casanave, C. P., & Hubbard, P. (1992). The writing assignments and writing problems of
doctoral students: Faculty perceptions, pedagogical issues, and needed research.
English for Specific Purposes, 11(1), 33–49. doi:10.1016/0889-4906(92)90005-U

Celce-Murcia, M., & Larsen-Freeman, D. (1999). The grammar book: An ESL/EFL teacher’s
course. Boston, MA: Heinle & Heinle.

Cheng, A. (2008). Analyzing genre exemplars in preparation for writing: The case of an L2
graduate student in the ESP genre-based instructional framework of academic literacy.
Applied linguistics, 29(1), 50–71. doi:10.1093/applin/amm021

Crewe, W. J. (1990). The illogic of logical connectives. ELT Journal, 44(4), 316–325.
doi:10.1093/elt/44.4.316

Cunningham, K. J. (2017). A phraseological exploration of recent mathematics research

articles through key phrase frames. Journal of English for Academic Purposes, 25,
71–83.

Currie, P. (1998). Staying out of trouble: Apparent plagiarism and academic survival. Journal
of Second Language Writing, 7(1), 1–18. doi:10.1016/S1060-3743(98)90003-0

Dulay, H., Burt, M., & Krashen, S. (1982). Language two. Oxford: Oxford University Press.

Flowerdew, J. (1999a). Writing for scholarly publication in English: The case of Hong Kong.
Journal of Second Language Writing, 8(2), 123–145. doi:10.1016/S1060-
3743(99)80125-8

Flowerdew, J. (1999b). Problems in writing for scholarly publication in English: The case of
Hong Kong. Journal of Second Language Writing, 8(3), 243–264.
doi:10.1016/S1060-3743(99)80116-7

Flowerdew, J., & Peacock, M. (2001). Issues in EAP: A preliminary perspective. In J.

Flowerdew, & M. Peacock (eds.), Research perspectives on English for academic
purposes (pp. 8–24). Cambridge: CUP. doi:10.1017/CBO9781139524766

187
Francis, G. (1994). Labelling discourse: An aspect of nominal-group lexical cohesion. In M.
Coulthard (ed.), Advances in written text analysis (pp. 83–101). London: Routledge.

Ganobcsik-Williams, L. (2004). A report on the teaching of academic writing in UK Higher

Education. London: Royal Literary Fund.

Gardezi, A. A., & Nesi, H. (2009). Variation in the writing of economics of students in Britain
and Pakistan: the case of conjunctive ties. In M. Charles, D. Pecorari, & S. Hunston
(eds.), Academic writing: At the interface of corpus and discourse (pp. 236–250).
London: Continuum.

Gosden, H. (1996). Verbal reports of Japanese novices’ research writing practices in English.
Journal of Second Language Writing, 5(2), 109–128. doi:10.1016/S1060-
3743(96)90021-1

Goulden, N. R. (1994). Relationship of analytic and holistic methods to raters’ scores for
speeches. The Journal of Research and Development in Education, 27, 73–82.

Gray, B. (2010). On the use of demonstrative pronouns and determiners as cohesive devices:
A focus on sentence-initial this/these in academic prose. Journal of English for
Academic Purposes, 9, 167–183. doi:10.1016/j.jeap.2009.11.003

Green, C. F., Christopher, E. R., & Kam Mei, J. L. K. (2000). The incidence and effects on
coherence of marked themes in interlanguage texts: A corpus-based enquiry. English
for Specific Purposes, 19, 99–113. doi:10.1016/S0889-4906(00)00015-6

Hacker, D., Sommers, N. I., Jehn, T. R., Rosenzweig, J., & Van Horn, M. C. (2004). Rules for
writers. Boston: Bedford/St. Martin’s.

Halliday, M. A. K., & Hasan, R. (1976). Cohesion in English. London: Longman.

Halliday, M. A. K., & Matthiessen, C. (2014). Holliday’s introduction to functional grammar.

New York, NY: Routledge.

Harmer, J. (2004). How to teach writing. Essex: Pearson Education.

Hawkins, J. A. (2015). Definiteness and indefiniteness: A study in reference and

grammaticality prediction. New York, NY: Routledge.

Hinkel, E. (2001). Matters of cohesion in L2 academic texts. Applied Language Learning, 12,
111–132.

188
Hinkel, E. (2003). Simplicity without elegance: Features of sentences in L2 and L1 academic
texts. TESOL Quarterly, 37, 275–301. doi:10.2307/3588505

Hu, G. (2007). Developing an EAP writing course for Chinese ESL students. RELC Journal,
38(1), 67–86. doi:10.1177/0033688206076160

Hyland, K. (2004). Disciplinary interactions: Metadiscourse in L2 postgraduate writing.

Journal of Second Language Writing, 13, 133–151. doi:10.1016/j.jslw.2004.02.001

Hyland, K. (2005). Metadiscourse. New York, NY: Continuum.

Hyland, K. (2013). ESP and writing. In B. Paltridge, & S. Starfield (eds.), The handbook of
English for specific purposes (pp. 95–113). Oxford: Wiley-Blackwell.

Hyland, K., & Hamp-Lyons, L. (2002). EAP: Issues and directions. Journal of English for
Academic Purposes, 1(1), 1–12. doi:10.1016/S1475-1585(02)00002-4

Hyland, K., & Shaw, P. (eds.). (2016). The Routledge handbook of English for academic
purposes. New York, NY: Routledge.

Hyland, K., & Tse, P. (2004). Metadiscouse in academic writing: A reappraisal. Applied
Linguistics, 25(2), 156–177.

Jin, B. (2015). Exploring the development of lexical verbs in academic writing: A multiple-
case study of three Chinese novice researchers. The Asian ESP Journal, 2(1), 7–38.

Johns, A. M. (2003). Genre and ESL/EFL composition instruction. In B. Kroll (ed.), Exploring
the dynamics of second language writing (pp. 195–217). Cambridge, UK: Cambridge
University Press. doi:10.1017/CBO9781139524810.014

Johns, A. M., & Dudley-Evans, T. (1991). English for specific purposes: International in
scope, specific in purpose. TESOL Quarterly, 26, 297–314. doi:10.2307/3587465

Johns, A. M., & Price-Machado, D. (2001). English for specific purposes: Tailoring courses
to student needs and to the outside world. In M. Celce-Murcia (ed.), Teaching English
as a second or foreign language (pp 43–54). Boston: Heinle & Heinle.
doi:10.1016/S0889-4906(00)00018-1

Kern, R. (2000). Literacy and language teaching. Oxford: Oxford University Press.

189
Knoch, U., Rouhshad, A., Oon, S. P., & Storch, N. (2015). What happens to ESL students’
writing after three years of study at an English medium university? Journal of Second
Language Writing, 28, 39–52. doi:10.1016/j.jslw.2015.02.005

Lee, E. J. (2004). A corpus-based analysis of the Korean EFL learners’ use of conjunctive
adverbials. English Teaching, 59(4), 283–301.

Leki, I., & Carson, J. (1997). Completely different worlds: EAP and the writing experiences
of ESL students in university courses. TESOL Quarterly, 31(1), 39–69.
doi:10.2307/3587974

Leung, C., Lewkowicz, J., & Jenkins, J. (2016). English for academic purposes: A need for
remodelling. Englishes in Practice, 3(3), 55–73.

Liu, D., & Gleason, J. L. (2002). Acquisition of the article the by nonnative speakers of
English. Studies in Second Language Acquisition, 24(1), 1–26.
doi:10.1017/S0272263102001018

Lorenz, G. (1999). Learning to cohere: Causal links in native vs. non-native argumentative
writing. In W. Bublitz, U. Lenk, & E. Ventola (eds.), Coherence in spoken and written
discourse: How to create it and how to describe it (pp. 55–76). Amsterdam: John
Benjamns. doi:10.1075/pbns.63.07lor

Lu, Z., Li, L., & Ottewell, K. (2016). Rhetorical diversity and the implications for teaching
academic English. The Asian Journal of Applied Linguistics, 3(1), 101–113.

Oh, S. Y. (2009). Korean college students’ use of English demonstratives in argumentative

essays. English Teaching, 64(1), 51–78. doi:10.15858/engtea.64.1.200903.51

Petch-Tyson, S. (2000). Demonstrative expressions in argumentative discourse: A computer

corpus-based comparison of non-native and native English. In S. Botley (ed.), Corpus-
based and computational approaches to discourse anaphora (pp. 43–64).
Philadelphia: John Benjamins Publishing Company. doi:10.1075/scl.3.02pet

Purpura, J. (2004). Assessing grammar. Cambridge, UK: Cambridge University Press.

doi:10.1017/CBO9780511733086

Read, J., & Hays, B. (2003). The impact of the IELTS test on preparation for academic study
in New Zealand. In IELTS Research Reports V (pp. 237–262). Canberra: IELTS
Australia.

190
Reid, J. (1992). A computer text analysis of four cohesion devices in English discourse by
native and nonnative writers. Journal of Second Language Writing, 1(2), 79–107.
doi:10.1016/1060-3743(92)90010-M

Reid, J. (2001). Academic EAP writing and curriculum design: what do we need to know? In
T. Silva & P. K. Matsuda (eds.), On second language writing (pp. 143–160). Mahwah,
NJ: Lawrence Erlbaum.

Sasaki, M. (2007). Effects of study‐abroad experiences on EFL writers: A multiple‐data

analysis. The Modern Language Journal, 91(4), 602–620. doi:10.1111/j.1540-
4781.2007.00625.x

Sasaki, M. (2009). Changes in English as a foreign language students’ writing over 3.5 years:
A sociocognitive account. In R. Manchón (ed.), Writing in foreign language contexts:
Learning, teaching, and research (Vol. 43) (pp. 49–76). Bristol: Multilingual Matters.

Sasaki, M. (2011). Effects of varying lengths of study-abroad experiences on Japanese EFL

students’ L2 writing ability and motivation: A longitudinal study. TESOL Quarterly,
45(1), 81–105. doi:10.5054/tq.2011.240861

Schmidt, H. (2000). English abstract nouns as conceptual shells: From corpus to cognition.
Berlin: Mouton de Gruyter. doi:10.1515/9783110808704

Shaw, P. (2009). Linking adverbials in student and professional writing in literary studies:
What makes writing mature? In M. Charles, D. Pecorari, & S. Hunston (eds.),
Academic writing: At the interface of corpus and discourse (pp. 213–235). London:
Continuum.

Shaw, P., & Liu, E. T.-K. (1998). What develops in the development of second language
writing. Applied Linguistics, 19(2), 225–254. doi:10.1093/applin/19.2.225

Silva, T., & Brice, C. (2004). Research in teaching writing. Annual Review of Applied
Linguistics, 24, 70–106. doi:10.1017/s0267190504000042

Sowden, C. (2005). Plagiarism and the culture of multilingual students in higher education
abroad. ELT Journal, 59(3), 226–233. doi:10.1093/elt/cci042

Storch, N., & Tapper, J. (2009). The impact of an EAP course on postgraduate writing. Journal
of English for Academic Purposes, 8(3), 207–223. doi:10.1016/j.jeap.2009.03.001

191
Strauss, S. (1993). Why ‘this’ and ‘that’ are not complete without ‘it’. In K. Beals, G. Cooke,
D. Kathman, S. Kita, K. McCullogh, & D. Testen (eds.), Papers from the 29th
Regional Meeting of the Chicago Linguistic Society (pp. 403–417). Stanford: CSLI.

Swales, J. M. (1990). Genre analysis: English in academic and research settings. Cambridge:
Cambridge University Press.

Swales, J. M., & Feak, C. B. (2004). Academic writing for graduate students: Essential tasks
and skills (2nd edition). Ann Arbor, MI: University of Michigan Press.

Tardy, C., & Swale, J. (2009). Form, text organization, genre, coherence, and cohesion. In C.
Bazerman (ed.), Handbook of research on writing: History, society, school,
individual, text (pp. 565–581). New York, NY: Routledge.

Wang, W. (2017). Learner characteristics in an EAP thesis-writing class: Looking into

students’ responses to genre-based instruction and pedagogical tasks. English for
Specific Purposes, 47, 52–60.

Woo, J. I. (2015). Effects of grammar instruction in context with writing practice on students’
manuscript writing in an EAP course. Korean Journal of English Language and
Linguistics, 15(2), 149–180. doi:10.15738/kjell.15.2.201506.149

Xudong, D., Cheng, L. K., Varaprasad, C., & Leng, L. M. (2010). Academic writing
development of ESL/EFL graduate students in NUS. Reflections on English Language
Teaching, 9(2), 119–138.

Yoon, H. (2006). A corpus-based analysis of connectors in Korean students’ essay writing.

Korean Journal of Applied Linguistics, 22(2), 159–178.

Zamel, V. (1983). The composing processes of advanced ESL students: six case studies.
TESOL Quarterly, 17, 165–187. doi:10.2307/3586647

192