1

VIETNAM NATIONAL UNIVERSITY– HO CHI MINH CITY

INTERNATIONAL UNIVERSITY
DEPARTMENT OF ENGLISH

THE USE OF GRAMMATICAL

COHESIVE DEVICES IN
COLLEGE STUDENTS’ ESSAY

A thesis submitted to
The Department of English, International University,
in partial fulfillment of the requirements for the Degree of
Bachelor of Arts in English Linguistics and Literature

Student’s name: Lê Minh Toàn – ENENIU18067

Supervisor: Đặng Hoài Phương, MA.

June 2022
 1

ACKNOWLEDGEMENTS

First and foremost, I would like to -thank my Supervisor, Master of Arts Dang

Hoai Phuong, for her unwavering support and invaluable advice throughout the thesis

writing process. My thesis would not have been completed if it had not been for her

guidance.

In addition, I would want to express my sincerest gratitude to my friends and

classmates, who have continually supported and encouraged me for a long time. They

also offered helpful advice during the thesis revision process.

Finally, I cannot express my gratitude adequately to all of my family, friends, and

participants for their support and cooperation throughout the completion of this thesis.
 2

Abstract

Grammatical cohesive devices (GCDs) as a part of cohesion play an essential

role in every well-constructed and comprehensible text. This research aims to

investigate the most commonly GCDs used and the types of GCDs found in the essays

written by Vietnamese undergraduates from International University based on the

framework of Halliday and Hassan (1976). By applying quantitative research, the

students were found to be capable of using a range of GCDs in their writing, with

reference devices accounting for the majority of the total number of GCDs.

Furthermore, the study also targets to find whether the students’ uses of GCDs relate

to their writing performances. The findings revealed that the number of words and the

overall number of GCDs utilized were also found to have an insignificant relationship

with the quality of writing. In addition, several problems with the usage of reference

and conjunction were identified in the corpus. The results suggested some solutions to

improve the efficiency of teaching and learning writing skills in ESL classes.
3
 1

Introduction

Rationale of the study

Among the four skills that are involved in English language acquisition process,

writing may prove to be the most difficult skill to learn and teach (Rao, 2017). For

Vietnamese students, writing at the university level demands students to be able to

produce academic texts which are clear, brief, focused, systematic, and evidence-

based such as essays, reports, theses, research papers and others.

Therefore, regarding essays, there are numerous factors to consider in the

formation of effective ones. The two factors have significant that influence on the

general quality of written production in terms of form and meaning are cohesion and

coherence (Medve & Takač, 2013). The topic of cohesion and coherence has

attracted the attention of various researchers in the field (Palmer, 1999; Chiang, 2003;

Dastjerdi & Talebinezhad, 2006). However, cohesion is more likely to be selected for

investigation due to the unequivocal attribute of grammatical cohesive devices

(GCDs).

Since the framework of GCDs was first introduced by Halliday and Hasan in

Cohesion in English in 1976, there have been various studies on GCDs which include

refererence, conjunction, substitution and ellipsis. According to Halliday and Hasan

(1976), GCDs along with lexical cohesion form the ‘texture’ that expresses the

property of being a text. These GCDs create the connection between sentences and

between elements in sentences that then contributes to the text’s coherence. Scholars

have attempted to investigate the use of GCDs and their effects on English language

writing (Trisnaningrum et al., 2019; Bahaziq, 2016). Nevertheless, there have been

few studies in the Vietnamese setting. Therefore, this study aims at examining the
 2

instances of GCDs in the essays composed by Vietnamese undergraduates, studying

the common mistakes made by students when using GCDs, and investigating the

possible correlation between the use of GCDs and their writing quality. Then it could

probably suggest some implications for learning and teaching writing skills for EFL

learners, speakers of Vietnamese, namely college students whose apology speech act

strategy is conducted purely in Vietnamese. The results of this exploration research

would contribute to the intercultural pragmatic resources as well as a foundation for

future comparative linguistic research.

Significance of the study

Teaching and learning writing skills may ever be a part of the English

acquisition process, and GCDs as a component of cohesion are undeniably an

essential factor of writing quality. A diverse range of research efforts in GCDs by

researchers (Liu & Braine, 2005) indicate the importance of utilizing GCDs in

students' essays. However, no GCD error usage theories that exist are sufficiently

broad to cover all the aspects of students’ GCD usage mistakes. In addition, there
is
not much research that aims at studying the use of cohesion or specifically GCDs in

the Vietnamese context. This research is regarded as one of the first to investigate the

use of GCDs in Vietnamese undergraduates’ essays in order to point out the pattern of

GCD use and common problems that students encounter when using GCDs. Then,

this study offers some pedagogical implications for the teaching of English writing to

undergraduates in Vietnam.

Research aims and research questions

This thesis is a detailed analysis of how Vietnamese college students employ

GCDs in their essays. The initial objective is to measure the instances of GDCs used
 3

in students' essays. Second, the study evaluates the correlation between the

deployment of GDCs and the quality of students’ essays. More precisely, the research

aims to answer three research questions.

1. What GCDs and their subcategories are employed in college students’ essays?

2. What are the common errors regarding GCD use found in college students’

essays?

3. Is there any correlation between the students’ essay scores and their use of

GCDs?

Literature Review
Cohesion, coherence and writing quality

Cohesion and coherence are the two inseparable terms in regard to the

evaluation of written discourse quality. According to Nunan (1993), cohesion refers to

how sentences are linked by cohesive devices (reference, conjunction, ellipsis,

substitution and lexical cohesion) allowing readers to perceive relationships between

sentences. In addition, coherence is the unity of the text in which all text elements are

connected to produce discourse that readers can understand.

To be specific, cohesion is defined as a semantic relation between an element

in the text and some other element that is crucial to the interpretation of it and is one

of the factor that contribute to the textuality of the text (Halliday & Hasan, 2014).

Halliday and Hasan (1976) defined text as any passage, spoken or written, that creates

a unified whole and characterized text as a semantic unit in which explicit cohesives

ties connect its parts. Cohesion of a text is dependable on lexical and grammatical

items which allow sentence sequences to be understood as connected discourse rather

than as autonomous sentences. Different parts of text are linked by explicit cohesive
 4

ties that are specified as five major classes including reference, conjunction,

substitution, elipsis, and lexical cohesion, which all have diverse subclasses. Of the

five explicit cohesive ties, reference, conjunction and lexical cohesion are more likely

to appear in written discourse whereas substitution and ellipsis are often found in

spoken discourse (Witte & Faigley, 1981).

Connor (1984) suggested that cohesion itself may not be an adequate measure

to fully illustrate the quality of the text and coherence must also be considered, not

only cohesion. Coherence is the quality of being logical, consistent and

comprehensible. Linguistically, coherence can be defined as the formal properties of

the text (Lee, 2002), and it relates to the semantic relationship within the text. One of

the features of coherence is that it links the surface text with cohesive devices

(Halliday & Hasan, 1976), which helps improve the overall quality of the text. While

cohesion in term of literature is generally accepted as being mainly related to the

structural linguistics, coherence, has been studied in fields that are apart from the

structure of the text, namely linguistics, discourse psychology, and cognition science

(Sanders & Maat, 2006). Malmkjaer (2001) claimed that “a coherent extended text is

the result of interaction between the reader’s world and the text, with the reader

making plausible interpretations.” As a result, a reader or writer is trying to make

sense of the text based on common background information outside of the text.

The topic of the correlation between cohesion, coherence and writing quality

has remained controversial. Bamberg (1983) pointed out that “coherence is generally

accepted as the essential condition in written discourse.” However, according to the

findings of Connor (1984)'s study, an ESL essay does not need to be coherent to be

cohesive. This theory is supported by Halliday and Hasan (1976), who claimed that

cohesion is the precondition for successful integration of ideas. Witte and Faigley
 5

(1981) suggested that at the most general level of analysis the density of cohesion in

higher-rated essays is higher than that of those lower-rated essays. Many studies

attempt to calculate the instances of cohesive devices to determine the impact of

cohesion on writing quality; nevertheless, the findings are mixed (Liu & Braine, 2005;

Toadithep, n.d.; Alarcon & Morales, 2011). The studies by Liu and Braine (2005)

and Toadithep (n.d.) pointed out that the writing quality correlates with the number of

lexical devices and the total number of cohesive devices used. On the contrary,

Alarcon and Morales (2011) indicated that cohesive devices are not significantly

correlated with the quality of the students’ essays.

Grammatical cohesive devices (GCDs)

The scope of this undergraduate research focuses mainly on GCDs regarding

cohesion; therefore, lexical cohesion is not taken into account. Of the five explicit
cohesive ties, reference, substitution, ellipsis, and conjunction are the four types of
GCDs identified by Halliday and Hasan (1976) and are selected to be investigated
Reference

Reference is a situation in which one component cannot be semantically

comprehended unless it is referenced to another component in the text. Reference

may either be exophoric or endophoric (Bloor & Bloor, 2013). Exophoric reference

requires the reader to look beyond the text for interpretation of the referent based on

the knowledge of the situation. In other words, exophoric references often serve as a

link from a text to its situational context; however, as far as Halliday and Hasan

(1976) are concerned, exophoric references do not contribute to the cohesion of a text.

On the other hand, endophoric reference exists within the text and is divided into

cataphoric and anaphoric reference. Bloor and Bloor (2013) explained that cataphoric

means 'forward pointing,' which means a pronoun or demonstrative comes first in the

text and the named expression appears later. On the other hand, anaphoric is
 6

“backward pointings,” which means the named thing comes first and is followed by

the pronoun.

There are three main types of reference: personal reference, demonstrative

reference and comparative reference. Personal reference is defined as reference to

something by identifying its function or role in speech situations through the

categories of people (Halliday & Hasan, 1976). Personal reference is classified into

three types: personal pronouns, possessive determiners (adjective pronouns), and

possessive pronouns, as illustrated in Table 1.

Person Personal Possessive Possessive

Pronoun Determiner Pronouns

Speaker I, me my mine

Addressee(s) with/without other you your yours

person(s)

Speakers and other person(s) we, us our ours

Other person; male he, him his his

Other person; female she, her her hers

Other person; object they, them their theirs

Object passage of text it its its

Table 1. Types of personal reference

Demonstrative reference is a form of verbal pointing, and it depends which

demonstrative word is chosen to identify where the referent is on a scale of proximity

(Halliday & Hasan, 1976). They are divided into the circumstantial demonstratives
 7

such as here, there, now, and then and the selective nominal demonstratives such as

this, that, these, and those.

According to Jamalzadeh (2017, p. 15), comparative reference employs adverbs

and adjectives to compare components within the text. Furthermore, Halliday and

Hasan (1976) claim that by establishing a connection of contrast indicated by

adjectives and adverbs, comparative reference contributes to textual cohesion.

Comparative reference is divided into general comparison and particular comparison.

General comparison simply relates to likeness and unlikeness between two things. On

the other hand, particular comparison means comparison in terms of quantity or

quality. The examples of comparative references are illustrated in Table 2.

General comparison Identify Similarity Difference

Example same, equal, such, similar, so other,

identical, identically similarly, likewise differences,

differently

Particular comparison Quantity Quality

more, fewer, less, comparative

further, additional; adjectives and adverbs,

so- as- equally- eg: better; so- as-

+quantifier, more- less- equally- +

eg: so many comparative adjective

and adverb,

eg: equally good

Table 2. Comparative reference items

 8

The definite article the shares important similarities with other determiners and

functions as a specifying agent that identifies a particular individual or subclass within

the class designated by the noun (Halliday & Hasan, 1976). It is one of the most

common used references according to various studies (Bahaziq, 2016; Nurhidayah &

Jismulatif, 2020)

Conjunction

Conjunction refers to linking devices between sentences or clauses in a text.

Conjunction expresses the ‘logical-semantic’ relation between sentences and clauses

that
rather than between words and structures (Halliday & Hasan, 1976). In other

words, they arrange the text in a logical order that is understandable to the reader or

listener. Conjunction is divided into four types, namely additive, adversative, causal,

and temporal conjunctions.

Additive conjunctions link units that share semantic similarity. Examples of

additive conjunctions are and, furthermore, in addition, likewise, also, etc.

Example 1. People gather around and play different games to relax and strengthen

their connections.

Adversative conjunctions indicate contrasting results or opinions. This type of

conjunction is expressed by words such as however, in contrast, but, whereas, etc.

Example 2. However, it is believed that drinking too much water will damage

your health.

Causal conjunctions introduce the cause or reason of what is being stated. They

are characterized by the use of items such as so, thus, therefore, because, etc.

Example 3. Because we need food and water to survive.

Temporal conjunctions represent sequence relationships of events. The examples

are finally, then, soon, at the same time, etc.

 9

Example 4. Finally, it is important that we drink enough water everyday.

Substitution

Substitution is defined as a replacement of one item by another in order to avoid

duplication. The main difference between substitution and reference is that
substitution is associated with the wording relationship whereas reference is
concerned with the relationship between meaning. It is claimed that substitution
occurs more frequently in conversation than in written discourse (Witte & Faigley,
1981). There are three types of substitution: nominal, verbal, and clausal. Nominal
substitution is the replacement of a noun or a nominal group with another noun; one,
ones and same are the elements of nominal substitution. In the following example,
one is the substitution of ‘the bottle of milk.’
Example 5. The bottle of milk has expired, so my mother will buy a new one.

Substituting a verb or a verbal group with another verb is the process of verbal

substitution. It is expressed by do, as illustrated in the following example.

Example 6. A: I hate eating fish.

B: Many people do.

Clausal substitution is substituting clauses by so or not.

Example 7. A: Everyone believes that she is guilty.

B: Yes, but supposing not. What will happen?

Ellipsis
Halliday and Hasan (1976) suggests that ellipsis shares a similar process to

substitution and is to substitute one item with nothing. They also claim that “ellipsis

is simply substitution by zero.” Another similar trait between these two is that ellipsis

is also observed more often in conversation rather than written discourse (Witte &

Faigley, 1981). When ellipsis occurs, the item that is removed from the text still can

be perceived by readers due to the anaphoric relationship. Like substitution, ellipsis is

also divided into three types: nominal, verbal, and clausal.

In nominal ellipsis, the noun is omitted.

 10

Example 8. I accept the first request form, but reject the other two (forms).

Verbal ellipsis involves the deletion of the verb.

Example 9. A: Have you been swimming?

B: Yes, I have (been swimming).

Causal ellipsis happens when the clause is eliminated.

Example 10. A: How did they get in?

B: I will show you how (they get in).

GCD errors

The study by Nasser (2017) points out some mistakes regarding the use of

GCDs made by Yemeni EFL learners. In terms of reference, the errors are divided

into three categories: omission of reference, overuse of reference, misuse or

inappropriate use of reference. In addition to the reference errors, the study also

suggested the errors regarding conjunction share the same pattern: omission of

conjunction, overuse of conjunction, misuse or inappropriate use of conjunction. The

author illustrated that due to the difficulty of substitution and ellipsis, students tend to

avoid using them. Therefore, the number of substitution and ellipsis errors were the
is
least as students’ mastery of these categories low.
 11

The GCD errors are also investigated by Liu and Braine (2005). The result of the
study clarifies some common mistakes when using GCDs in Chinese undergraduates’
writings. In terms of reference, one problem is the shift of pronouns by which students
shift pronouns within clauses, particularly from the singular to the plural or from the
first person to the second person pronoun. These are examples of pronoun shifting.

Example 11. When a child cry that means they want something from the parents.

Example 12. Playing sports is very good for your health so we should play sports
everyday.

The second reference error proposed by Liu and Braine (2005) is the omission or
misuse of the definite article. The students sometimes added or removed the definite
article due to the confusion between indefinite or definite articles. Below are the
examples of this error.

Example 13. The video games are becoming an essetial part of their life.

Example 14. People who live in (the) US have the right to refuse vaccination.

The third error related to the underuse of comparatives and overuse of the phrase
‘more and more.’ The study suggested that student have difficulty using other
comparative expressions beside ‘more and more’ as comparative reference only
accounts for 6.1% of the total number of the sub-categories of reference.

Previous studies

There have been various studies focusing mainly on the subject of GCDs used in
written discourse. The majority of those studies followed the pattern of qualifying
GCDs to see the effect of cohesion on the written discourse quality; they might share
the same research question which is to examine the most frequently used GCDs.
However, each study was distinctive in terms of methods, subjects and findings.

The study of Abdurrahman (2013) investigated the use of GCDs in students’ thesis
writing. This research aimed to find out the types of GCDs students mostly used and
how GCDs created cohesive discourse. In this study, a descriptive case study
approach was applied to investigate 10 thesis writings composed by students of the
English Education Study Program in Tanjungpura University, Indonesia. The findings
suggested that the majority of GCDs used were reference and conjunction, with
82.25% and 17.12%, respectively. The result also indicated that students tended to use
singular pronouns to refer to plural objects and plural pronouns to refer to singular
objects.
 12

Another study by Bahaziq (2017) defined and described the cohesive devices based
on the framework of Halliday and Hasan (1976). The study investigated the essay
from Michigan English Language Assessment Battery (MELAB), which is a
standardized examination of English as a foreign language to indicate the necessity of
using GCDs. The analysis also revealed that the most GCDs used were reference and
conjunction, while there was no evidence of substitution. Even though the data
analysis denoted the importance of using GCDs to produce high quality writing, the
essays still need modifications to achieve a higher level of cohesion.

The study from Trisnaningrum et al. (2019) employed the framework of Halliday and
Hasan (1976) to investigate GCDs used and to find out the kind of GCD
specifications in college students’ academic writing essays. The study was conducted
with 42 college students who participated in an online writing course. The descriptive
and analytical methods were applied by giving an assignment to assess GCDs used in
their academic essays. The findings of the study suggested that 1048 GCDs were used
and reference was the highest frequently used with 53.53%. The data indicated that
students were more familiar with reference and conjunction use rather than
substitution and ellipsis. The researchers also suggested that inappropriate use of
GCDs was the result of the lack of GCDs in terms of comprehension, knowledge and
ability in writing.

Nasser (2017)’s study aimed to identify the different categories and the most frequent
category of errors in the use of GCDs. The subjects of his study were twenty-four
argumentative essays composed by students of the third level in the Department of
English at the Faculty of Education-Saber, Aden University, Yemen. The study
employed a qualitative approach to identify the categories of errors, whereas a
quantitative approach was used to diagnose the most frequent category of errors made
in the essays. The findings stated that the most frequent errors of GCDs were as
follows: reference, conjunction, substitution, and ellipsis.

Another significant study of the correlation between cohesive features and writing
quality was carried out by Liu and Braine (2005). The purpose of the research is to
investigate the use of cohesive devices in 50 argumentative compositions created by
Chinese undergraduate non-English majors by using Halliday and Hasan(1976)’s
taxonomy and framework. The researchers identified and counted the cohesive
devices used in each composition. After that, by using SPSS, they were capable of
calculating the frequency, mean and standard deviation of the cohesive devices in
each category. Then, the correlation between the frequency of the use of cohesive
devices and the writing quality was also examined. The study results illustrated that
the writing quality significantly correlates with the number of used cohesive devices.
Methodology

Research design
 13

The current study employed mixed approaches that combine the quantitative

research and qualitative research to answer the three research questions. With regard

to quantitative method, every instances of each GCDs were collected and categorized

based on the framework of Halliday and Hassan (1976). In addition, the researcher

applied the Pearson’s Correlation Coefficient which is a measure of the strength of the

linear relationship between two variables (Lane, 2003, p. 170), in this study the two

variables are the number of GCDs and the scores of the essays. Pearson's r has a range

of -1 to 1. A correlation of -1 indicates a perfect negative linear relationship, 0

indicates no linear relationship, and 1 indicates a perfect positive linear relationship

(Lane, 2003, p. 175). In other words, Pearson’s Correlation Coefficient was used to

calculate the correlation between the use of GCDs and the writing quality. In term of

qualitative research, it was applied to investigate the common GCDs errors made by

non-English major in their argumentative essays. Thereupon, these error GCDs were

categorized and analyzed.

 14

Subjects

This study examined 60 essays written by the students of Writing AE1 and

Writing AE2 classes in International University - VNU HCM. Writing AE1 and

Writing AE2 courses are required in order to prepare students with adequate English

writing skills before entering their majors with English as the primary language of

instruction. The participants came from different departments ranging from

Information Technology, Business Administration and Biotechnology. Students at this

level are expected to be at B2 to C1 level, according to the Common European

Framework of Reference for Languages (CEFR) - a framework for evaluating

language proficiency. In other words, students attending these writing courses are

required to have good command of English and are able to understand detailed

reasoning as well as generally handle complex language well.

To complete these courses, the students must be capable of producing an

argumentative essay with proper organization which consists of 3 clear parts:

introduction, body and conclusion. Furthermore, the students also have to

acknowledge the opposing points and rebuttals as well as provide their own

arguments. Finally, grammar and language use is another important factor that the

students must satisfy when composing their essays.

Data collection

At the beginning, a consent form was provided to get the students’ approval

for the use of their in-class assignment essays. After that, 60 essays composed by the

students from Writing AE1 and Writing AE2 classes were gathered. The topics of the

investigated essays may vary, but the required knowledge of writing skills remains the

same. It is demanded that the students finished their essays within a time limit of 60

minutes.
 15

Data analysis

In order to answer the first question, all the GCDs in the essays were collected,

identified, counted and categorized manually. Then, each GCD in terms of reference,

conjunction, substitution and ellipsis detected were classified into smaller

specifications. By using the SPSS program, the frequency, mean and standard

deviation of the word number, score and GCDs were calculated.

To answer the second research question, the correctness and incorrectness of

the GCD use were assessed. The instances of GCDs use in the essays were put into

comparison with the score given to them marked by the researcher and an advanced

English speaker based on the provided rubrics from International University - VNU

HCM. To be specific, the researcher applied the Pearson correlation coefficient on the

frequency of used GCDs, the frequency of each GCD subcategory, and the essay

scores for the purpose of observing whether there is a correlation between students’

essay scores and their use of GCDs. Pearson's r can range from -1 to 1, a correlation

of -1 implies a perfect negative linear relationship, a correlation of 0 indicates no

linear link, and a correlation of 1 suggests a perfect positive linear relationship.

To answer the third research question, qualitative approach was applied. First,

the researcher reviewed each composition, after that the common GCD errors existed

in the corpus were identified. Based on the previous studies by Liu and Braine (2005),

Nasher (2017), these errors were then classified and assigned to their categories.

Moreover, the researcher was able to describe other type of GCD errors that were not

mentioned in those studies.

Findings and discussion

Findings

The corpus and evaluation results

 16

As can be seen in Table 3, the corpus comprised 60 essays composed by

Vietnamese undergraduates. On average, each essay consisted of 375 words and

ranges from 161 to 641 words. It can be seen that the texts vary in terms of length.

The standard deviation is 94.16 indicates the number of words in each essay vary

significantly. The size of the corpus is 22,516 words.

Variable N Mean StDev Sum Max Min

Word number 6 375.27 94.16 22516 641 161

Table 3. Description of the corpus

Table 4 illustrates the overall result of the score given to the essays. 60

compositions were gathered and rated by the researcher and an advanced level

English user based on the marking criteria of International University’s

Argumentative Essay Rubrics. The mean score of 60 essays is 74, most of the score

ranges from 65 to 83 with the standard deviation of 9.32. Overall, International

University students’ writing skill is above average level with only one essay gets the

score below 50.

Variable N Mean StDev Max Min

Score 60 73.68 9.32 94 49

Table 4. Description of the scores

 17

GCDs found in the essays

As shown in Table 5, among 22,516 words of the corpus, 4466 words are GCDs.

Reference is the most GCD used with 2965 instances account for 65.4% of total

number of GCD, followed by conjunction with 1438 instances (33.1%). On the other

hand, the instances of substitution and ellipsis are very few with only 40 (0.89%) and

23 (0.51%) occurrences which suggest that students may not be familiar with the use

of them. The order of GCDs used are illustrated in the mean of each device number

per each essay, with nearly 50 instances of reference, followed by 24 of conjunction

items, while substitution and ellipsis devices were very low with 0.66 and 0.38

respectively. The value standard deviation of reference is 17.8 which indicates that the

number of this item is notably diversified from essay to essay, whereas conjunction

placed second with the value standard deviation of 7.72 and there is no significant

change in the instances of substitution (1.09) and ellipsis (0.61) as they are barely

used by students.

Furthermore, the ranking of GDCs’ frequency is also consistent with that of

previous study by Trisnaningrum et al. (2019) in which reference was the most use

with 53.53%, followed by conjunction with 45.80%, substitution with 0.67% and no

instance of ellipsis found.

Type of GCDs Referenc Conjunction Substitution Ellipsis Total

Frequency 2965 1438 40 23 4466

 18

Mean/essay 49.41 24 0.66 0.38 74.38

Standard Deviation 17.7 7.72 1.09 0.61 21.79

Percentage based on 65.4% 33.1% 0.89% 0.51% 100%

total GCDs

Table 5. GCDs used

The use of reference

From Table 6, it can be seen that among the four subcategories of reference, the

definite article ‘the’ has the highest proportion of use with 1249 occurrences

constituting 65.4% of total references. Personal references place second with 1054

instances (35.54%), followed by demonstrative references (15.4%) and comparative

references (6.84%). Averagely, there are 20.81 definite articles used by students in

each essay and its standard deviation value is 12.01 which means the range of ‘the’

used in each composition is from 9 to 32 occurrences. The mean frequency of

personal reference per essay is 17.5 and the standard deviation value is 8.84 suggests

that the number of personal references are wide spread out around the mean. On the

other hand, the averages of demonstrative and comparative items are 7.65 and 3.38,

with the standard deviation of 3.93 and 3.43 signify that they are clustered tightly

around the mean. In terms of total GCDs used, the percentage of personal items

(23.6%) and definite article (27.96%) account for more than half of the entire GCDs.

Demonstrative and comparative devices attribute 10.2% and 4.54% to the total GCDs.
 19

However, the result of this study is contradictory with that of Liu & Brane (2005),

the previous study shown that personal references (60%) had the highest percentage

of use, followed by definite article (29.3%) and comparatives (6.1%), demonstratives

(4.6%) was the least reference used. Moreover, in the studies on cohesive devices by

Zhang (2000) and Trisnaningrum et al. (2019), the use of definite article ‘the’ in terms

of reference was excluded which made personal references the most reference

subcategory used.

The compositions' predominant use of personal reference in the corpus are ‘I’,

‘my’, ‘they’, ‘their’, ‘we’, ‘it’, ‘its’. The students tend to use ‘I’ and ‘my’ to express

their personal perspective, such phrases like ‘I believe’, ‘I think’, ‘my opinion’ appear

a lot in the corpus. Third person pronouns such as: they or theirs are also utilized by

students so that their writings are more authoritative and not too objective. This

finding suggests that students are well aware of the use of pronouns when composing

argumentative essays. In terms of demonstrative reference, ‘this’, ‘that’, ‘these’,

‘there’ are the most common devices used by student, whereas instances of

circumstantial demonstratives like ‘here’, ‘now’, ‘then’ are barely found in the

corpus. Among the comparatives, comparative adjectives and adverbs are used a lot

namely ‘more’, ‘better’, ‘harder’, ‘more and more’.

Type of Personal Demonstrative Comparative Definite Total

references article ‘the’ references

Frequency 1054 459 203 1249 2965

Mean/essay 17.5 7.65 3.38 20.81 49.91

 20

Standard 8.84 3.93 3.43 12.01 17.7

Deviation

Percentage based 35.54% 15.4% 6.84% 42.12% 100%

on total references

Percentage based 23.6% 10.2% 4.54% 27.96% 65.4%

on total GCDs

Table 6. References used

As shown in Table 7, with 815 instances, additive devices account for the largest

percentage of conjunction used (56.67%), followed by adversative devices with 306

instances (21.27%), causal devices with 181 instances (12.58%) and temporal devices

with 136 instances (3.04%). The average number of additive items appear in each

essay is 13.58, that of demonstrative is 5.1, causal is 3.01 and temporal is 2.26. The

standard deviation values for these items are 6.15, 2.54, 2.15, 1.42, reveal that each

conjunction item firmly vary around the mean. With regard to the total number of

GCDs, additive account for 18,24%, followed by adversative with 6.85%, causal with

4.05% and with 3.04%.

This finding is inconsistent with that of the study by Liu & Braine (2005) as the

number of causal devices ranked second (19.5%) among five sub-categories and more

frequency of temporal devices (18.6%) were found than adversative devices (15.6%)

in the corpus.

In terms of additive conjunction, ‘and’ appears in every essays as it is the most

basic connecter to link phrases, clauses and sentences. Beside ‘and’, ‘also’, ‘or’,

‘furthermore’ and ‘moreover’ are also employed a lot by students in their essays. The

adversative items with the highest used rate are ‘but’ and ‘however’. The instances of
 21

‘but’ and ‘however’ can be found in almost 60 compositions just like ‘and’. On the

other hand, others adversative items namely ‘nevertheless’, ‘despite’ or ‘yet’ are not

very common in the corpus which indicate that students may not be competent

enough to specify the shift of meaning. As to causal conjunction, ‘because’,

‘therefore’ and ‘as the result’ are used the most and students make use of them

correctly. This implies that regarding causal items, students have good understanding

of them and are confident to use them in their essays. Among temporal devices,

‘first’, ‘firstly’, ‘second’, ‘secondly’, ‘finally’, ‘to sum up’, ‘in conclusion’ account

for the highest instances. The typical pattern of temporal items used in most of the

essay begins with ‘first’ or ‘firstly’ to demonstrate the first point of the argument,

follow by ‘second’ or ‘secondly’ to add more ideas to the argument and ‘in

conclusion’ to summarize all the illustrated points.

Type of Additive Adversativ Causal Temporal Total

conjunction e conjunction

Frequency 815 306 181 136 1438

Mean/essay 13.58 5.1 3.01 2.26 23.96

Standard 6.15 2.54 2.15 1.42 7.72

Deviation

Percentage 56.67% 21.27% 12.58% 9.45% 100%

based on total

conjunction

Percentage 18.24% 6.85% 4.05% 3.04% 33.1%

based on total
 22

GCDs

Table 7. Conjunctions used

The use of substitution

From Table 8, it can be seen that nominal substitution account for the highest

percentage of substitution used with 33 instances (82.5%), followed by 4 clausal

substitution devices (15%) and verbal substitution with 1 instances (2.5%). The

instance of substitution in each essay is very low which indicated by the mean number

of them with 0.55 instance of nominal substitution, 0.05 instance of verbal

substitution and 0.06 of clausal substitution. The low standard deviation of these

devices, 0.99, 0.12, 0.35 sequentially indicate the instance of these items are close to

the mean number. Regarding the total GCDs, substitution items only account for a

small percent with 0.73% for nominal substitution and 0.13% for clausal substitution,

whereas verbal substitution is the least used item with 0.02%.

With regard to nominal substitution, the most common words used to replace

nouns are ‘one’, ‘one’s’, ‘ones’ and others. The only occurrence of verbal substitution

found in the corpus is ‘don’t’, whereas ‘not’ are used more often to replace a clause.

The finding indicates that even though students are not familiar with substitution in

writing essays, they show in depth understanding of these GCDs as no errors of

substitution are found in the corpus.

Type of substitution Nominal Verbal Clausal Total

Frequency 33 1 6 40
 23

Mean/essay 0.55 0.05 0.06 0.66

Standard Deviation 0.99 0.12 0.35 1.09

Percentage based on total 82.5% 2.5% 15% 100%

substitution

Percentage based on total GCDs 0.73% 0.02% 0.13% 0.89%

Table 8. Substitution used

The use of ellipsis

Table 9 presents the number and percentage of ellipsis used by students in 60

essays. Nominal ellipsis account for the highest rate use with 15 occurrences (15%),

clausal placed second with 6 times employed by the students which accounts for

26.08% and verbal ellipsis has the least percentage of use with only 1 instance

(4.34%). The average number of ellipsis items are very low with 0.25 nominal

ellipsis, 0.01 verbal ellipsis and 0.1 clausal verbal ellipsis found per each essay. Their

standard deviations are 0.61, 0.57 and 0.12 respectively which signify they are closely

grouped around the mean. On behalf of total GCDs, they also contribute for small

percentages, nominal ellipsis account for 0.73%, verbal ellipsis is 0.02 and clausal

ellipsis 0.13%.

As can be seen from the result that with 23 instances of ellipsis which is

significantly small in number, this indicates that students are aware that ellipsis might

be really not appropriate and formal enough to use in academic essays. The finding is
 24

related to the study by Witte & Faigley (1981) which suggests that ellipsis and

conjunction are more likely to be used in spoken discourse rather than written. In

terms of usage, nominal ellipsis is often used by omitting nouns and pronouns such as

‘people ‘or ‘they’. The only instance of verbal ellipsis is used wrongly. Clausal

ellipsis is also employed by eliminating the ‘wh’ phrase to shorten the sentences.

Lack of talent is just an excuse for laziness because as (what have been)

demonstrated above.

However, each Vietnamese person and the government are currently making

major contributions to epidemic prevention and treatment for those (who have been/

are/ had been) afflicted.

However, vaccination is always a controversial topic, so getting vaccinated should

depend on individual agreement, (it should) not (be) a demand.

Type of ellipsis Nominal Verbal Clausal Total

Frequency 16 1 5 23

Mean/essay 0.25 0.01 0.1 0.38

Standard Deviation 0.61 0.57 0.12 0.3

Percentage based on total ellipsis 65.21% 4.34% 26.08% 9.45%

Percentage based on total GCDs 0.33% 0.02% 0.13% 0.51%

Table 9. Ellipsis used

The correlation between GCDs use and students’ writing quality

Overall, students are familiar with GCDs and capable of using a variety of them in

their essays. However, to answer the second research question: whether there is a

correlation between the use of GCDs and the quality of writing composed by students,
 25

various factors and the scores are assessed. In particular, Pearson’s correlation

coefficients are performed among the number of words and writing scores, the

number of total GCDs and writing scores, the number of each GCDs and writing

scores which are demonstrated in the table below.

Score Word Number Reference Conjunction Substitution Ellipsis

count of
GCDs

Scor 1 .419** .280* .275* .091 -.017 -.053

e

Table 10. Relationship between word count, total number GCDs, each types

of GCDs and quality writing

In terms of composition scores and the GCDs, Table reveals that the composition

scores insignificantly co-vary with the total number of GCDs (r=0.280). This finding

is similar to that of Liu & Brane (2005) and Trisnaningrum et al (2019), in which the

coefficient value were r=0.315 and r=0.422 respectively meaning that the overall

GCDs use and the quality of writing is barely correlated. The findings show a positive

relationship between the reference devices and the quality of composition (r=0.275),

the conjunctions devices and the writing score (r=0.091) which are genuinely

inconsiderable. On the other hand, substitution (r=-0.17) and ellipsis (r=-0.53) even

suggest that there are a negative correlation between them and the writing scores.

However, Trisnaningrum et al (2019)’s findings suggest that substitution and ellipsis

low correlate with the quality of the composition.

Problem with GCDs

The students in this study are able to apply a range of GCDs in their writings

to make them cohesive and coherent, as seen in the previous tables. However, the
 26

argumentative writing produced by these undergraduate non-English majors had some

cohesiveness issues. This is consistent with Othman’s (2019) findings that Saudi

students often struggled with cohesiveness. These issues mostly concerned with the

use of reference and conjunction.

Errors of reference

Despite the fact that reference devices are the most commonly utilized

category of cohesiveness in the students' writing, they appeared to be uncomfortable

with their use. The difficulties with reference devices were mostly of two types:

shifting pronoun and omission, misuse or overuse of the definite article.

The problem of shifting pronouns within or between clauses found in this

study is similar to the finding of Zhang (2000). Students are likely to switch from

singular pronouns to plural pronouns or from first person to second person pronouns.

Consequently, the incompatibility of pronouns and reference items creates a

confusion for readers and leads to comprehension issues. The following are some

samples of student writing with inconsistent referents.

Example 15. For example, a group of undergraduates who had already tests and
received their anticipative grades or some random business just earns a promotion in his
work.

Example 16. So we can see the importance of a plan when you want to obtain your
goal.

Example 17. Moreover, there are many games that not require player to win such as
farm games, city games, building games that player can build the house they like or breed
pigs, chickens and cows, which is the main point that help player relax and have fun.

As the most used GCDs in the corpus, it is undeniable that definite article ‘the’ causes
some problems for the students when utilizing it. Students are sometimes confused about the
use of definite and indefinite articles. Subsequently, students tend to exploit ‘the’ in their
essays whenever they refer to a noun. This confusion also results in the addition of definite
articles to indefinite nouns and vice versa. Another problem with ‘the’ is that students
occasionally eliminate the definite articles. Some examples of previous problems are
demonstrated below.
 27

Example 18. However, due to the time it takes to pick up the vaccine, the vaccine is
still far from perfect and much remains in the experimental phase.

Example 19. However, a certain number of people only enjoy playing games when
they gain the victory.

Example 20. On (the) one hand, maybe someone will think that praise in private is
enough, we do not need to make a big deal, but to a person or a collective, a team, there is no
meaning in the reward if no one knows about, it is clearly that praising is usually more
meaningful and effective in public.

Error of conjunction

In terms of conjunctions, they are the second prominent in the student’s

essays. However, students in this study also have problems using conjunction devices
which are the inappropriate use of conjunction items, overuse of conjunction, and
starting sentences with coordinating conjunctions.

Regarding the misuse of conjunction, the students sometimes use adversative

conjunctions without any sense of contrast or result between clauses in their
sentences. Moreover, the inappropriate uses of conjunction also vary which cannot be
fully described in this study due to the lack of a proper framework for conjunction
errors. Therefore, they are plainly categorized as misuse.

Example 21. It can help people avoid the virus but with one hundred percent

Example 22. Cause at last, it just a game, no need to be so serious about

winning.

Example 23. However, as we have the way to neutralize the vaccine to save
people, but we have no cure for the COVID-19, choosing to be injected would be the
wise choice.

The overuse of conjunction is presented in the use of additive items which are ‘and’
and ‘or’. Overusing them results in the production of a stringy sentence with unclear
ideas and meaning.

Example 24. In today’s society, the COVID19 pandemic has been a major
problem to the world, and to solve that problem, many countries have tried to make
Corona virus vaccines and many large countries like Russia, America or China have
successfully created it for the safety of humanity.

Example 25. For examples, in Germany, faking vaccine certification or refuse

to vaccinate can put you in the jail because this is an order from the government so
once the police stopping you just to check wherever you get vaccinate or not, you will
 28

either show them the certification or let them drive you to the vaccinate center to get a
shot

Example 26. Firstly, vaccines have been tested on many people and scientists
and doctors can guarantee the safety of vaccines, every people can feel safe to take
the vaccination without any fear.

Another problem that the students in this study have is the use of ‘and’, ‘but’ and ‘so’.
Instead of using other conjunctions such as: therefore, as a result or consequently,
they tend to use ‘and’, ‘but’ and ‘so’ at the beginning of a sentence in order to connect
to the previous sentences. However, ‘and’, ‘but’ and ‘so’ function as coordinate
conjunctions that link words, phrases or clauses together not sentences (Stern,
2002:103). Below are some examples form the students’ essays.

And some individuals have more serious reactions than others

Example 27. But some days after getting these vaccines, we need to take a
rest.

Example 28. So we can see the importance of a plan when you want to obtain
your goal.

Example 29. So that we have to balance and analyze all the factors, ways that
we can do as well as all the outcome that may happen

Example 30. So the risk of getting other illnesses is quite small while taking
the vaccination will help them become immune to COVID-19.

Discussion

Regarding the first research question, the findings show that reference

devices (65.4%) accounts for the highest percent of total number of GCDs used in the

argumentative essays, followed by conjunction devices (33.1%) and substitution

devices (0.89%), whereas ellipsis devices (0.51%) are the least used GCD. The

highest percent of reference devices in this study can be attributed to the calculation

of the definite article ‘the’ which is similar to many studies (Liu & Brane, 2005;

Bahaziq 2017). This also explains why among the subcategories of GCD, definite

article (27.96%) is the most commonly used. Personal reference (23.6%) placed

second and followed by additive conjunction (18.24%). It can be seen that all

subcategories of GCD can be found in the corpus and the majority of them are
 29

reference devices and conjunction devices. However, there is only one instance of

verbal ellipsis and it is used wrongly.

A further analysis of statistic reveals that the writing scores are highly correlated with

the number of word counts. However, the correlation coefficient value between

writing quality and the number of GCDs is 0.28. The value answers the third research

question that there is insignificant correlation between the number of GCDs and the

composition scores. This finding indicates that the high use of GCDs perhaps do not

result in higher scores. The study by Zhang (2010) also suggests the same result that

there is a weak relationship between the total number of GCDs and writing quality.

In order to answer the third research question, a qualitative analysis is employed to

investigate the GCD errors usage in students’ writings. These common errors are

presented by deviation in the use of reference and conjunction. In terms of reference,

they are manifested as shifting pronoun and omission, misuse or overuse of the

definite article. The errors related to conjunction are classified into three types:

inappropriate use of conjunction items, overuse of conjunction, and starting sentences

with coordinating conjunctions. As this section mainly focuses on the common errors

of students when employing GCDs. The errors of substitution and ellipsis which are

rarely found in the corpus due to the low instance of substitution and ellipsis, are

not discussed.

Limitations and suggestions

Limitations

The findings of this study have to be seen in light of some limitations. First, the

subjects of the study are 60 essays composed by undergraduates from International

 30

University – VNU HCMC which may not be sufficient for the research topic's

generalization. The second limitation concerns the English level of the students, due

to the training program of International University – VNU HCMC, English is the

main language used for instruction. Therefore, most of the participants' English ability

was at an advanced level, better than that of undergraduate students from other

universities in Vietnam. Consequently, it is not possible to conclude that the outcomes

of undergraduate students' GCDs use in this study may be relevant for all

undergraduates in Vietnam settings as their English competencies may vary. The third

downside of this study is that it cannot cover all the aspects of cohesive devices in

particular and writing skills in general. Due to the time limitation and the complex of

other cohesive factors, this paper mainly focuses on the small element of cohesion

which are GCDs that cannot significantly contribute to the improvement of students’

writing ability.

Suggestion

As demonstrated by the findings and discussions presented above, this study offers

some implications for the teaching of English writing to undergraduates in Vietnam in

general.

First of all, in order to produce quality writings, students should understand the

marking criteria consisting of GCDs as a part of cohesion, along with other elements

namely organization, grammar and language use, content are the key factors that

contribute to the quality. Teachers should be responsible to initially introduce and

explain these marking criteria to the students. Consequently, their awareness of these

factors will enhance and when writing in English, students will know what to

emphasize.
 31

Secondly, although the majority of students are competence enough to employ

reference and conjunction in their essays, some of them still have problems using

these devices correctly and effectively. The misuse and overuse of certain devices

mentioned in previous part limit their abilities to produce a quality text. Furthermore,

the use of substitution and ellipsis as a method to avoid repetition are hardly found in

the corpus. Writing teachers should deliver explicit instruction with examples in class.

In other words, specific activities relate to the variety of GCD items and their usage

should be introduced to students. Teachers could assign students to write short

paragraph using various GCDs. Peer review might then be utilized to assess the GCDs

employed in each other's writing and provide feedback on the impact of using such

cohesive devices. Following the completion of each work, the writing teacher should

select a sample composition for critique, emphasizing the necessity of using GCDs

effectively and responsibly, as well as reminding students to avoid the inappropriate

use of GCDs.

Conclusion

Overall, the findings revealed that non-English major undergraduates in

International University – VNU HCM are high competence and proficient in using

English. In other words, they are at an advanced English level and capable of

producing quality English argumentative essays. This can be explained by the

program provided the International University – VNU HCM which uses English as

the main language for teaching and learning. With regard to teaching and learning

writing skills, cohesion and coherence are the prominent factors that required to

develop a high quality text. It is important that students know how to utilize these

devices, in particular GCDs to produce a cohere composition. This study aimed to

study the use of GCDs in Vietnamese undergraduates’ writing essays. In addition, the
 32

study also wanted to examine whether there is a correlation between the frequency of

GCDs used in the writing and the writing quality. Furthermore, the study focused on

investigating the common errors committed by Vietnamese undergraduates when

using GCDs. The researcher applied quantitative research to examine the frequency of

GCDs, while employing Pearson correlation coefficient to identify the relation

between GCDs used and the quality of the text. Qualitative research was utilized to

analyze the common GCD errors. The result suggested that students prefer to use

reference and conjunction devices, while the uses of substitution and ellipsis were

limit. In addition, the number of GCDs insignificantly co-varies with the writing

quality. Inappropriate use of references and conjunction were the main problems that

students encountered. From this study, it is suggested that more work needs to be

done in the writing classroom to raise students' awareness of the value of and usage of

GCDs in their writing.

 References

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

Blanchard, K., & Root, C. (2010). Ready to write 3: From paragraph to essay. Pearson
Education.

Palmer, J. C. (1999). Coherence and cohesion in the English Language Classroom: The
use of lexical reiteration and pronominalisation. RELC Journal, 30(2), 61–85.
https://doi.org/10.1177/003368829903000204

Chiang, S. (2003). The importance of cohesive conditions to perceptions of writing

quality at the early stages of foreign language learning. System, 31(4), 471–484.
https://doi.org/10.1016/j.system.2003.02.002

Dastjerdi, H. V., & Talebinezhad, M. R. (2006). Chain-preserving deletion procedure in

cloze: A discoursal perspective. Language Testing, 23(1), 58–72.
https://doi.org/10.1191/0265532206lt318oa

Nunan, D. (1993). Introducing discourse analysis. Penguin.

Witte, S. P., & Faigley, L. (1981). Coherence, cohesion, and writing quality. College
Composition and Communication, 32(2), 189. https://doi.org/10.2307/356693

Medve, V. B., & Takač, V. P. (2013). The influence of cohesion and coherence on text
quality: A cross-linguistic study of Foreign Language Learners’ written production.
Second Language Learning and Teaching, 111–131. https://doi.org/10.1007/978-3-642-
35305-5_7

Liu, M., & Braine, G. (2005). Cohesive features in argumentative writing produced by
Chinese undergraduates. System, 33(4), 623–636.

Toadithep, N. (n.d.). Relationship between cohesive devices and writing quality in Thai
EFL Students’ Written Works, Srinakharinwirot University

Bloor, T. (2013). The functional analysis of English. Taylor & Francis. 93-105.

Jamalzadeh, Mehri.,2017. Cohesive Conjunctions in Medical Articles. The paper is

retrieved from the ResearchGate website. Volume 1,1-25.

Trisnaningrum, Y., Alek, A., & Hidayat, D. N. (2019). Discourse analysis of

grammatical cohesion devices in college students’ academic writing essay. IJEE
(Indonesian Journal of English Education), 1(1), 79–90.
https://doi.org/10.15408/ijee.v1i1.12502

Alarcon, J.B., & Morales, K.N. (2011). Grammatical cohesion in students'

argumentative essay. International Journal of English and Literature, 2, 114-127.
Abdurrahman, N. H. (2013). Grammatical cohesion analysis of students’ thesis writing.
Jurnal pendidikan dan pembelajaran, 2(11).

Bahaziq, A. (2016). Cohesive devices in written discourse: A discourse analysis of a

student’s essay writing. English Language Teaching, 9(7), 112.
https://doi.org/10.5539/elt.v9n7p112

Nasser, A. N. A. (2107). A study of errors in the use of grammatical cohesive

devices in argumentative texts written by Yemeni EFL learners. International
Journal of Applied Research, 3(10), 172-176.

Nurhidayah, N., & Jismulatif, J. (2020). A study on reference as cohesive device in

essays written by the fourth semester students of the English Study Program Universitas
Riau. Journal of Educational Sciences, 4(1), 212.
https://doi.org/10.31258/jes.4.1.p.212-219

Stern, G. (2002). An outline of English grammar with exercises and answer key (p.103).
Learners.

Zhang, A. (2010). Use of cohesive Ties in Relation to the Quality of Compositions by

Chinese College Students. Journal of Cambridge Studies, 5(2-3)

Rao, P. (2017) Developing writing skills among the EFL/ESL learners. Retrieved June
4, 2022, from
https://www.researchgate.net/publication/334762842_DEVELOPING_WRITING_SKI
LLS_AMONG_THE_EFLESL_LEARNERS

Lane, D. (2003). Values of the Pearson Correlation. In Introduction to statistics \ (pp.

170–175). essay, Open Textbook Library.
 Appendices

Appendix A. Consent form

INFORMED CONSENT
Informed Consent to Participate in a Research Study

International University – Department of English

Quarter 6, Linh Trung Ward, Thu Duc District, HCMC

Title of Research Project: The use of grammatical cohesive devices in college students’
essays

Name of Principal Investigator: Le Minh Toan

Email of Principal Investigator: [email protected]

Phone number of Principal Investigator: +84 949 9517 02

A. PURPOSE AND BACKGROUND

B. Le Minh Toan from the Department of English, International University – VNU HCM is
conducting research on the use of grammatical cohesive devices in college students’
essays. The purpose of your participation in this research is to help the researcher
investigate the use of grammatical cohesive devices. The study is conducted to identify the
grammatical cohesive devices in the essays and examine their role in students’ written
discourse quality. You were selected as a possible participant in this study because you are
currently attending the course of Writing AE1 and Writing AE2 and your essays are
appropriate for the study. PROCEDURES
If you agree to participate in this research study, the following steps will occur:
1. The researcher will collect your essays in the form of Microsoft Word
documents.
2. The researcher will analyze the instances of grammatical cohesive devices
and categorize them into subtypes.
3. Collected data will be presented into tables and percentage of grammatical
cohesive devices’ usage will be calculated.
4. The researcher will assess the correctness and incorrectness of grammatical
cohesive devices in your essays.
This process is expected to take a few months.

C. RISK
There are no foreseeable risks involved in participating in this study.

D. CONFIDENTIALITY
The records from this study will be kept as confidential as possible. No individual
identities will be used in any reports or publications resulting from the study. All data
collected under Microsoft Word document form will be given codes and stored
separately from any names or other direct identification of participants. Research
information will be kept in locked files at all times. Only research personnel will have
 access to the files and Microsoft Word documents and only those with an essential need
to see names or other identifying information will have access to that particular file.
After the study is completed by 15 June, 2022, all collected data will be destroyed.
th

E. BENEFITS OF PARTICIPATION

There will be no direct benefit to you from participating in this research study. The
anticipated benefit of your participation in this study is to improve college students’
knowledge, comprehension and ability of grammatical cohesive devices use in order to
overall enhance their writing quality.

F. VOLUNTARY PARTICIPATION
G.
Your decision whether or not to participate in this study is voluntary and will not affect
your relationship with the department of English. If you choose to participate in this study,
you can withdraw your consent and discontinue participation at any time without prejudice.

H. QUESTIONS

If you have any questions about the study, please contact Le Minh Toan by calling +84 949
517 02. You can also contact Department of English Linguistic at International with any
questions about the rights of research participants or research related concerns.

CONSENT

YOU ARE MAKING A DECISION WHETHER OR NOT TO PARTICIPATE IN A

RESEARCH STUDY. YOUR SIGNATURE BELOW INDICATES THAT YOU HAVE
DECIDED TO PARTICIPATE IN THE STUDY AFTER READING ALL OF THE
INFORMATION ABOVE AND YOU UNDERSTAND THE INFORMATION IN THIS
FORM, HAVE HAD ANY QUESTIONS ANSWERED AND HAVE RECEIVED A
COPY OF THIS FORM FOR YOU TO KEEP.

Signature Date__________

Research Participant

Signature Date__________

Interviewer
 Appendix B. Final Exam Argumentative Essay Rubrics

ORGANISATION
- The essay follows the outline with 3 clear parts:
Introduction (ends with the thesis statement) 10 9

Body
10 7
At least 1 paragraph discusses the counterargument.

Each paragraph discusses a particular point and

begins with a 10 8
clear topic sentence.
1 Each paragraph has specific supporting details
(fact, examples, etc.). There are no sentences that are 10 6 The essay is quite well organized.
off-topic.
Each paragraph has cohesion and coherence.
There are
transition signals to show the relationship among 10 8
ideas and to
link paragraphs.
Conclusion (summarizes the main points by
paraphrasing the thesis statement, begins with a
10 9
conclusion signal, and leaves the final thoughts on the
topic)
CONTENT
- Acknowledgement of the opposing points and
Rebuttals Counterarguments are not really valid.
2
20 13 Refutation Points are not well
- Own arguments expressed nor elaborated.

GRAMMAR AND LANGUAGE USE (15 pts)

- A wide variety of sentence patterns (simple
sentences, compound/complex sentences, relative
clauses, passive/active voice) and vocabulary are
3 presented correctly.
20 14 Satisfactory language use
- Language control is good, and meaning is clear.
- Students demonstrate command of spelling,
capitalization, punctuation.

TOTAL 100 74