Reiteration Relations in Arab ESL Student Writing: Effect of First Language Transfer

Cohesive devices have long been a challenging aspect for second language learners of English, especially for learners whose native language is linguistically distant from English such as Arabic. This study investigates the frequency of using lexical cohesion by Arab ESL undergraduate students studying at the intensive English course at Ohio University. T-test was used to measure the frequency of cohesive types students employed in their writings. Due to the existence of cohesive ties in Arabic that are similar to repetition in English, the study hypothesized that students with lower proficiency levels are more likely to overuse repetition cohesive ties compared to higher proficiency students who have a better awareness of the second language’s linguistic features. However, results indicated otherwise. Repetition lexical ties were used more by the higher proficiency level students. This study findings contradict with other similar studies conducted on ESL students that stated lower proficiency students tend to rely on repetition to compensate for their lack of English vocabulary. The different educational systems and linguistic backgrounds of the participants can be the result of such different findings.


Introduction
Writing teachers, especially, EFL/ESL instructors, were mostly concerned about learners' production at the sentence level (Khalil, 1989).They tended to focus mostly on the technical errors produced by students that caused them to forget about the lexical errors that are far more important factors in creating and interpreting discourse (Morris, 2004).The fact that exploring lexical cohesion in students' writing and teaching is a challenge, led many teachers to focus on grammatical cohesion instead.As Sinclair (1998) points out, "the tools for lexical analysis remain unrefined, while grammar has gone through many stages of sophistication" (p.3).This, however, has changed in the recent years.That is, many researchers recognize the importance of lexis in the creation of the continuity of the text (Hasan, 1984;Hoey, 1991;Widowson, 1992), which, as a result, has influenced the teaching of cohesion in writing courses.Teachers became more concerned to have students produce coherent texts than before (Kafes, 2012).Because of the large quantity of discourse-based studies on the process of coherence and cohesion behavior of EFL/ESL learners, methodologists and language teachers have realized the importance for language learners to acquire more than vocabulary and grammar rules to produce meaningful and coherent texts in English (Palmer, 1999;Kafes, 2012).
However, such a notion remains absent to Arab EFL teachers.Many Moroccan EFL teachers, for instance, believe that teaching cohesion means teaching conjunctions (Hellalet, 2013).As a consequence, Moroccan EFL students' writings are characterized by an overuse of connectors.In addition to the lack of instruction, Arab students may confuse the use of lexical cohesion devises with the ones exist in their native language.In Arabic, the repetition of the same word or its synonyms is a strategy used as a means of persuasion and emphasis (Khalil, 1989).In this regard, it is important to note the impact of first language transfer on second language acquisition that causes Arabs to misuse English cohesion and coherence devices.According to Gvarishvili (2012), language transfer could be either negative or positive: negative by which it causes the learners to commit mistakes, or positive from which learners can benefit to learn the second language.As explained by Warsono (2016), the features of the English language that significantly differ from Arabic could cause the learners to commit errors because of negative language transfer.In contrast, other second language learners whose native languages have similar grammatical features to English could transfer their knowledge from their first language to facilitate learning the second language through the so-called positive transfer.Fauziati (2009) stated, "native language transfer has become and will always be the concern of second language acquisition studies."In this study, Arab students' compositions will be analyzed in terms of the frequency and correctness of employing cohesive lexical ties.Native language transfer will be taken into account in a way that its effect will be detected depending on students' proficiency level.

Background of the Study
Cohesion together with coherence "is often considered as a fuzzy concept", leaving teachers being aware of the benefits of teaching cohesion and coherence without knowing how to teach them (Lee, 2002).The binary between the two concepts can hardly be distinctly drawn, which, as a consequence, causes students and teachers to be uncertain how and when to employ each (Khafes, 2012).For this reason, this part of the study aims to review the two concepts briefly in order to provide a basic knowledge about the difference between cohesion and coherence.

Coherence
Coherence has always been considered as an essential element of good writing.One of the characteristics of coherence is that it allows a "text to be understood in real setting" (Connor, 2009).It is perceived by Bamberg (1984) and Richard (1990) to be an important quality of effective writing.However, it is important to note that the conception of coherence changed towards the end of the 1960s when the focus of linguists shifted from analyzing texts at sentence level to discourse level.(Lee, 2002, pp. 13-137) Coherence is defined in the contemporary time as the "connectivity of the surface text evidenced by the presence of cohesive devices."(Halliday & Hasan, 1976) It is also perceived to be "an information structure which guides the reader in understanding the text and contributes to the topical development of the text."(Connor &Farmer, 1990;Firbas, 1986;Lautamatti, 1987) And by Kintsnh & van Dijk (1978), it is understood to be the "connectivity of the underlying concept evidenced by relations between propositions and how these relations contribute to the overall discourse theme and organization".To sum up the previous definitions, coherence functions to connect the surface level and organization of a text.Unlike cohesion, it does not concern the semantic relations and ideas' connectivity.(Halliady& Hasan, 1976)

Cohesion
Cohesion is considered to be one of the important features to form an understandable English text.It helps the sentences as well as the ideas presented in a text to be linked together to form a unity for the text (Halliday and Hasan, 1976).It can be defined as "the way in which a sentence is connected to its predecessors in a passage by means of some lexical items and grammatical features; it refers to the elements on the surface level of text that connect its parts and help it form a unified whole" (Hellalet, 2013).Without cohesion, a text can be misleading and ideas can be ill-organized and hard to be followed.Halliday and Hasan (1976) explained that cohesion can only be achieved "when the interpretation of some element in discourse is dependent on that of another.The one presupposes the other" (p.4).
To better understand it, Halliday and Hasan (1976) identify and classify cohesion into five categories: reference, substitution, ellipsis, conjunction, and lexical cohesion.The first four types are called "grammatical cohesion", while the last one, which would be the focus of this study, is referred to as "lexical cohesion".According to Halliday and Hasan (1976), a marginal area between grammatical and lexical cohesion is the cohesive function of the class of "general noun"., "… a General Noun is itself a borderline case between a lexical item (member of open set) and grammatical item (member of closed system)".Thus, a distinctive feature between lexical cohesion and grammatical cohesion, as explained by Halliday and Hasan (1976), is that the former has a more generalized reference and comes in various shapes compared to the later.Example of general nouns are explained as follows: -People, person, man, woman, child, boy, girl [human] -Thing, object [intimate concrete count] -Business, affair, matter [inanimate abstract] As illustrated in the three examples above, the general nouns [human, intimate concrete count, and inanimate abstract] function as the umbrella terms that include a list of words.For instance, (people, person, man, woman, child, boy, girl) can all referred to as humans, thus the general noun used to identify them is [human]…etc.

Lexical Cohesion
Since lexical cohesion is going to be the major concern of this study, this section is dedicated to explaining the different types and functions of lexical cohesion.As can be inferred from its name, lexical cohesion is defined as the cohesion created in a text by the selection of vocabulary (Teich & Fankhauser, 2005;Kafes, 2012).Halliday and Hasan (1976) explain that the use of general nouns as cohesive elements is perceived to be a special case when seen from lexical point of view.It is considered to be a much general phenomenon that can be called reiteration.They defined reiteration as "a form of lexical cohesion which involves the repetition of a lexical item, at one end of the scale; the use of a general word to refer back to a lexical item, at the other end of the scale; and a number thing in between the use of synonym, near-synonym, or superordinate."(Halliday & Hasan, 1976).It has to do with the use of general nouns to create cohesive contexts by replacing one element with another in the ongoing text or discourse.As explained in the previous quote, Halliday and Hasan (1976) categorize the different formats that reiteration comes in into five categories: repetition, synonym, near synonym, superordinate, and general words.
To facilitate understanding the difference between categories, the following section illustrates examples for each.It is noteworthy, though, that synonym and near synonym as categorized by Halliday and Hasan (1976) were treated synonymously in this study.Additionally, in order to easily detect the difference between categories, the same example will be used in each category.The examples will be drawn from the following scenario: A boy went to a restaurant and got a contaminated food.He immediately started eating it because he was very hungry.

1-Repetition
-The boy is going to get a food poisoning if he doesn't stop eating.
"boy" in this example refers to the same word "boy".No substitution of the word occurred.It is the same word with the same meaning; therefore, it is a repetition reiteration.

2-Synonym
-The lad is going to get a food poisoning if he doesn't stop eating.
In this example, "boy" was replaced by "lad".Since both words convey the same meaning and function virtually similarly, "lad" in this sentence is an example of a synonym reiteration.

3-Superordinate
-The person is going to get a food poisoning if he doesn't stop eating.
The "boy" in this example functions as a subordinate to "person".In other words, "person" can be a boy, a girl, a man, or an old man, and they are all in a subordinate level to the broader term "person".

4-General word
-The poor thing is going to get a food poisoning if he doesn't stop eating.
The phrase "poor thing" refers to the "boy" but with different level of generality.They both serve a similar semantic function as cohesive references.
As can be seen in this section, differentiating between these four types of reiteration can rather be critical, especially to nonnative speakers.At the same time, ESL as well as EFL teachers could find it challenging to point out such subtle difference of each lexical reiteration to their students, which makes this topic in need for further research.

Previous Studies
Several studies were conducted to analyze the use of cohesive devices by nonnative speakers (Atari, 1983;Ferris, 1994;Palmer, 1999;Hinkel, 2001;Kafes, 2012, Wenxing & Sun, 2012;Scott, et al., 2016).They either investigated the density of using a certain cohesive device or comparative studies that compared the number of lexical cohesion devices used by native to nonnative speakers.Since the primary purpose of this study is to analyze the use of types of reiterations by ESL Arabs who are nonnative speakers of English, reviewing the studies that investigated the use of lexical cohesion by nonnative speakers would be crucial.It is worth noting that only two studies were conducted on Arab participants, the vast majority of the studies were dealing with nonnative speakers as a group without considering the students' native language that can have a tremendous effect on the use of cohesive devices due to possible L1 transfer.
The most important and most cited resource in studies concern cohesive devices is Halliday's and Hasan's (1976) book: Cohesion in English."It was this book that made cohesion an important concept in many fields and has evoked wide discussion and application ever since" (Xi, 2010, pp.139).In their book, they studied the grammatical and lexical devices very comprehensively and interpreted cohesion as "the set of semantic resources for linking a sentence with what has gone before".Halliday and Hasan (1976) suggested the categorization of lexical cohesion that is dominantly used by most of the studies that analyzed students' use of lexical cohesion.They identify two major categories of lexical cohesion: lexical reiteration and collocation.The subcategories for lexical reiteration are: repetitions, synonyms, near synonyms, superordinate, and general words.They also present pedagogical approaches for teaching cohesion as well as numerous examples to facilitate understanding the meaning and the function of each cohesive device.Despite the critique of the book's inexplicit organization and lack of illustration to cohesive devices by Hoey (1991), Schiffrin (1994), Verschueren (1999), andXi (2010), it is still the most, if not the only, valuable resource that explained cohesion in English thoroughly.Not only does it play an important role for helping researchers who intend to learn more about cohesion, but also for teachers who teach cohesion to their students.
After the publication of Halliday's and Hasan's (1976) book, the concept of cohesion has been applied by numerous studies in different fields such as translation (e.g. Baker, 1992;Neubert & Shreve, 1992;Hatim & Munday, 2004;), psycholinguistics (e.g.Garrod & Sanford, 1994;McCabbe, 1998;Carrol, 2000), and second language writing (e.g.McCarthy, 1991;Liu, 1999;Zhang&Liu, 2003;Hyland, 2005;Rost, 2005).One of the second language writing studies that not only benefited from Halliday and Hasan's (1976) definition and categorization of cohesion, but also implemented their taxonomy to analyze students' papers is Connor's (1983) study.Connor (1983) conducted a comparative study analyzing cohesion features in ESL students' composition and explained how they differ from the ones native speakers used in their writings in terms of the number and the types of cohesive ties used.Her findings revealed that there is a relatively high frequency of lexical cohesion in both groups' writings in general, yet the only distinctive feature that distinguishes ESL students from native speaker is the implementation of the subcategories of lexical cohesion.That is, ESL students showed higher percentages of lexical reiteration use, with relatively small numbers of synonyms and collocation compared to native speakers who demonstrated more use of collocation and less of lexical reiteration.Connor's (1983) findings suggest that native speakers have better vocabulary awareness that allowed them to elaborate the concepts they introduce.In contrast, nonnative speakers seem to lack such word richness.They instead tended to use lexical reiteration devices to compensate for their lack of L2 vocabulary.
Similar to Connor (1983), Ferris (1994) analyzed the composition of 160 ESL students whose native language was Chinese, Arabic, and Spanish.She aimed to determine the lexical and syntactic features used by students with considering their proficiency level.Ferris (1994) found that the advanced students at higher proficiency levels demonstrated a better use of the targeted textual features in their composition than those at lower levels.More importantly, less advanced students use more lexical repetition ties in their essays to promote textual cohesion than any of the other reiteration subcategories.
Unlike the previous two studies that looked at ESL students' composition from various L1 backgrounds, Khalil's (1990) study was one of two that targeted only Arabic speakers.In his study, he analyzed EFL west bank Arab students' composition for the purpose of determining the frequency and the type of lexical cohesion ties used in students' writings.He implemented Halliday and Hasan's (1976) model to analyze students' use of cohesion and coherence in their writings.Findings revealed that Arab students used lexical cohesive ties twice as much the number of grammatical ones."The analysis of cohesion showed that the students overused lexical reiteration as a cohesive device and underused the other devices available in English."(Khalil's, 1990, p.366) Repetition, in particular, was the most frequent lexical reiteration tie used among the other subcategories.He attributes this finding to the religious and literacy written Arabic where there is a tendency to repeat words and phrases (p.363).He suggested that teachers "should pay more attention to the teaching of other lexical cohesive ties such as synonym and collocation" (p.366) to help them use a variety of ties.Hellalet's (2013) study is the second one that targeted only Arab students.In her study, Hellalet (2013) looked into the writings of Moroccan EFL students to investigate their use of reiteration relations.Her aim was to determine the extent to which Moroccan college students rely on lexical cohesion to produce native-sounding texts.After analyzing twenty essays written by college students from two different proficiency levels, results of her study, similar to the previous studies, revealed that repetition is the most dominant cohesive relation used by students from both levels.Hellalet (2013) also found that students from both levels overused conjunctions to connect their sentences as a result of their limited vocabulary and their lack of awareness of the role of lexis in the creation of continuity and connectedness in their writings.

Justification of the Study
This study concerns the use of reiteration relations used by Arab ESL students.The aim is to investigate the lexical ties as categorized by Halliday and Hasan (1976) and analyze how frequently they are used by Arab ESL students depending on their proficiency level and the class they are taking.Since the only two studies that targeted Arab participants (Khalil, 1990;Hellalet, 2013) were conducted on EFL Arab students, the major justification of this study is that it analyzes ESL Arab writings.Considering the difference between EFL and ESL students in terms of the amount of exposure to the second language and the different teaching style, it is expected that the findings of this study would differ from those of Khalil's (1990) and Hellalet's (2013).Also, since Hellalet (2013) attributed the overuse of repetition ties by Arab EFL students to the poor teaching styles in EFL classrooms, this paper targets Arab ESL students taught by different teachers in a different educational system.Additionally, it is worth noting that although there were a number of Arab ESL students in Ferris's (1994) study, her study was more focused on the general lexical and syntactic features of the texts than it is particularly limited to the reiteration ties.Thus, in this study, I solely focus on the reiteration ties (repetition, synonym, superordinate, and general word) to accurately detect the type and number of ties that are often used by Arabs.Furthermore, Ferris (1994) did not consider the class environment that each student attends since her study is a corpus analysis study that only considers the level of proficiency.This study, however, aims to analyze students' writings based on the type of the class they are attending, and whether it has an influence on the frequency of ties when they take a class with native speakers or only nonnative speakers, like themselves.Compositions from three different classes were analyzed with taking into account the different features and requirements for each course.

Study Hypothesis
The hypothesis of this study is that it is not only the inadequate EFL teaching styles that causes Arabs' overuse of repetition like what Hellalet (2013) argues; rather, there are several other impactful factors such as language proficiency and classroom environment.Also, even though some EFL students might be highly proficient in English, they may not be necessarily aware of the different lexical ties other than what they have been introduced to by their teachers or textbooks.Therefore, in addition to taking into account students' proficiency levels (intermediate and advanced), class environment will also be considered.In a way of explanation, the advanced level learners are divided into two groups: students who are taking ENG-1610 and students who are taking ENG-1510.Those who are taking ENG-1610 are taking the class with only ESL students, whereas those who are taking ENG-1510 are taking the class with native English speakers.Given the amount of the target language and the different teaching style that ENG-1510 students are exposed to, less repetition ties are expected to be used by them compared to those who are taking ENG-1610.
Additionally, worth considering, it is the effect of first language transfer on the use of second language.According to Attari (1984) and Khalil (1990), Arab students tend to employ some oral style strategies of their native language in their English writing.They explain that such style can briefly be defined by the way in which Arabs use synonymous words to emphasize a certain point.This repetition, however, is considered an act of redundancy in English.Due to the existence of such linguistic feature in Arabic, an overuse of the reiteration tie repetition is anticipated by the intermediate group because of the less linguistic awareness they have compared to the advanced level group.

Methods
This descriptive study analyzes students' papers from two different levels: intermediate and advanced.Both groups are ESL Arab students studying at the Intensive English Language Program (ELIP) at Ohio University.Although their majors differed, the classes from which these compositions were adapted are required before joining any program, thus they all took the class under the same conditions regardless of their future field of study.After sending a recruitment email targeting only Arabic native speakers, sixteen Arab students volunteered to participate, eight intermediate and eight advanced.They were a mix of Saudi and Omani students whose native language is Arabic.They were asked to share a sample of their writing through a mediator who omitted their names and sent them to the researcher to protect participants' privacy.Only the course number and the assignment information that remained on the papers when they were received.

Proficiency Level
The way students' level of proficiency was determined is by the class they were writing the paper for.In other words, students who were taking classes ENG-1510 and ENG-1610 were identified as advanced, whereas those who were taking ENG-D160 were considered as intermediate level students.The number of participants from the advanced level were a total of eight students, four from ENG-1610 and four from ENG-1510.And from the intermediate level (or ENG-D160), the total number of students was eight.It is worth noting that ENG-1510 and ENG-1610 are classes designed for higher proficiency level students, thus students would have to pass ENG-D160 in order to be able to take either one.For further information about the three courses, the course overview of each class as described in the university webpage (Note 1) is explained as follows:

Writing and Rhetoric I (ENG-1510)
This class is a requirement for all university students despite their major.Its primary focus is to help students to practice composing and revising in a logical and coherent way.Although this class is designed for native speakers, nonnative speakers have the option to either take it or take ENG-1610.This class is under the supervision of the Rhetoric and Composition program in the English department.

Freshmen Composition and Business Writing (ENG-1610)
Freshmen Composition and Business Writing is a class designed for ESL undergraduate students to help them improve their professional English writing skills.This course fulfills the requirements for a freshmen composition course ENG-1510.A perquisite to take this course is ENG-D160, both of which (ENG-1510 and ENG-D160) are supervised by the ELIP program in the linguistics department.

Fundamental English Usage Skills (ENG-D160)
ENG-D160 class is designed to assist nonnative English speaking undergraduate students in their development of becoming more skilled writers in their undergraduate coursework.International students join this class after they pass the preparatory language courses.This class is considered to be an intermediate writing course that prepares students to take ENG-1610 (or ENG-1510).

Method of Analysis
Similar to the previous studies (Hellalet, 2013;Khalils, 1990;Connor, 1983;and Ferris, 1994), this study implemented Halliday and Hasan's (1976) taxonomy to identify the subcategories of reiteration cohesion ties that students used in their writings.Each tie was identified, counted, and described in terms of the type of lexical reiteration it represented (repetition, synonym, superordinate, and general word).T-test was conducted to measure the cohesive ties frequency usage by the two groups.

Results
Table1 demonstrates the difference between the two proficiency level groups in terms of their use of the reiteration relations.The independent samples T-test show a significant statistical difference between the two proficiency groups in their use of the repetition words with a p value = 0.02.This indicated that the less proficient students used significantly less repetition word ties than did the advanced students.Regarding the other three reiteration relations: synonym, superordinate, and general word, results showed that there is no significant difference in students' use of each reiteration relation.As indicated in Table 1, the difference in using the superordinate word ties seems to be the most noticeable one among other ties with 0.177 p value.This shows that students whose level is intermediate used less superordinate ties compared to advanced students.The use of general word ties, on the other hand, seems to be virtually the same across the two proficiency groups.Figure 1 displays the results by using clustered columns.

Proficiency Level Factor
The results of the study are compatible with Khalil's (1989), Ferris's (1994), and Hellalet's (2013) findings who found that repetition ties are the most frequently used among the other cohesive ties.In this study, both groups regardless of their proficiency levels or the class type demonstrate a high frequency usage of repetition ties.That is, the mean of intermediate students' use of repetition ties was 26.25 and only 9 for the synonyms, 10.25 Class Type superordinate, and 1 for the general word.Similarly, the mean of repetition ties by advanced students was 34.5, while it is only 10.5 for the synonyms, 8 superordinate, and 7.25 for general words.This finding, as Hellalet (2013) explains, is a result of first language transfer.Because repetition is an aspect of Arabic rhetoric, Arab students tend to use repetition ties more frequently than the others (Atari,1983;Khalil, 1989;Hellalet, 2013).
In terms of whether the L1 influence will decrease as the students' proficiency level increases, which was the focus of this study, the results came rather surprising.The study hypothesized that repetition ties will be used more frequently by less proficient students due to the stronger L1 influence and lack of linguistic awareness, yet results show that they are more frequently used by the higher level instead.As demonstrated in table 1, the mean of using repetition ties by intermediate level students is 26.25, while it is 34.5 by advanced learners.This finding can be attributed to several possible factors.One of which is that the fact that L2 learners are always instructed to avoid writing ambiguous sentences.As a negative side effect of that, advanced learners may have overused repetition ties to make their sentences clearer.Additionally, as L2 students are encouraged to compose clear sentences, they are also encouraged to diversify their use of vocabulary.Intermediate student's concern about using the same word repeatedly may have caused them to look up synonymous words that they may not necessarily know the meaning of to use in their texts.
The results of using the synonym ties, on the other hand, was as expected.-"This self-assessment allows me to rethink and reconsider my behaviors…" -"Working with people from different countries will be very beneficial and helpful for both sides" -"I fully explained and described the skills I have for the job" The explanation of such misusage of synonym ties of English is that Arab students tend to employ their spoken Arabic synonym style to their English writing (Attari, 1984).It is considered to be an act of emphasis, in Arabic, to use a synonymous word that has virtually the same meaning and function of the previously mentioned one, yet it is can be perceived as redundancy in English.
Similar to repetition, it was unexpected that intermediate level students would use more superordinate ties than advanced level students due to the difficulty of using such ties professionally in English.The study hypothesized that advanced learners would perform more usage of lexical ties compared to intermediate.Results, however, showed that intermediate level students used superordinate ties more frequently and fairly properly.The possible interpretation of this is that less proficient students benefited from their reliance on their native language to use the superordinate ties.Although a few of the referenced items were ambiguous in their writing, the intermediate group demonstrated more usage compared to the advanced group.Another interpretation of the less usage of superordinate ties by advanced level students is similar to why they did not use repetition as much as intermediate level did.Being worried about the clarity of the text may have inhibited them from using repetition and superordinate ties in their writing.
Lastly, it is also worth noting how higher proficiency level used significantly more general word ties than intermediate level did.Based on the findings of Hellalet's (2013) study and the findings of this study, it is logical to conclude that both ESL and EFL students are not explicitly taught the different ways of how to link their ideas, rather it is left to them to figure out.This study proves this claim by demonstrating a higher percentage of general word ties use by the higher proficiency level students and those who had more exposure to the target language.Both advanced level learners (see Table 1) and students who are taking writing course with native speakers (see Table 2) demonstrated a better usage of general word.

Class Type Factor
In an attempt to test if taking a class with native speakers would differ from taking a class with only nonnative speakers, this study presented the difference in students' answers from two different classes settings, ENG-1510 being with native English speakers and ENG-D-160 and ENG-1610 being with only ESL students.Results showed no significant difference in the use of any lexical tie.Instead, both groups used the lexical ties relatively similarly.This finding affirms that the type of class students takes (native or only-nonnative speakers) does not affect their choice of the lexical ties as much as the proficiency level does.

Conclusion
The use of lexical ties by Arab ESL students was tested in this study.Considering the level of proficiency and the class types as independent variables, results show that students' L1 language and their L2 proficiency level plays a pivotal role in their usage of lexical ties.Class type, however, showed no statistical significant difference.This study finding adds to the scholarship in this topic the following: 1-ESL students taking composition class with native speakers makes no difference in improving their L2 grammatical knowledge.Instead, taking composition classes that are designed for second language learner learners can be better for them since teachers in such classes tend to discuss L2 grammatical features more often.
2- Khalil's (1990) findings regarding Arab's overuse of repetition ties are affirmed in this study.Despite students' level of proficiency, both groups demonstrated a higher percentage of repetition ties use.This indicates that the overuse of repetition ties has to do with the students' native language not the language learning (ESL/ EFL) setting.
3-This study's findings contradict with Ferris's (1994) who stated that that less advanced students "tended to rely on repetition to promote textual cohesion" (p.417).This study finding reveal opposite results.Advanced level students demonstrated more usage of repetition ties than intermediate students.

Limitation of the Study
This study looked into the reiteration lexical relations but not the grammatical cohesion as Halliday and Hasan (1976) label.In their book, they assert that the boundary between lexical cohesion, which is referred to as reiteration in this study, and the grammar cohesion of the reference type is never a clearcut.Also, it is important to note that only 16 papers composed by Arab ESL students were analyzed in this study.Other studies might consider analyzing more samples to have more generalizable results.Comparing ESL students' papers with English native speakers' papers is another possible future research in this area of study.

Figure 2 .
Figure 2. Difference between the use of reiteration relations based on class type

Table 1 .
Difference between the use of reiteration relations based on proficiency level

Table 2
demonstrates the difference in terms of students' use to the reiteration relations based on the class they took.Students 9,10,11, and 12 took ENG-1510 course, whereas 13, 14, 15, and 16 took ENG-1610.The data shows no statistical significant difference between the two groups, yet it demonstrates a higher frequency of using repetition ties by ENG-1610 students compared to ENG-1510.

Table 2 .
Difference between the use of reiteration relations based on class type It was anticipated to have more synonym ties used by advanced students than by intermediate due to the fact that advanced level students have more vocabulary richness than those whose level is intermediate.According to Carrell (1983), cohesion development is related to vocabulary development, and, in this case, it is salient in advanced and ENG-1510 students' use of synonymous words.Although results show that intermediate level students used quite similar number of synonym ties to those whose level is advanced, they used it in such a redundant way as they do in Arabic.The following examples demonstrate how intermediate students misused the synonym ties: