Article Title [Persian]
نوشتن یک فعالیت اجتماعی و عامل مهمی در روند سوادآموزی جامعه است. به نظر می رسد زبانآموزان - با فرهنگها و زبان مادری مختلف - به دلیل تداخل زبان اول عملکرد متفاوتی در نگارش متون به زبان خارجی نشان میدهند. در دهههای اخیر یکی از معروف ترین مطالعات در زمینه مقایسه استفاده از شاخص های فراگفتمانی در نگارش متون توسط زبانآموزانی بوده است که زبان و فرهنگ مادری آنها متفاوت از زبان دوم است. این پژوهش ، انشاهای اغنایی نوشته شده توسط دانشجویان زبان انگلیسی با ملیت ایرانی و چینی را با هم مقایسه نموده است. هدف پژوهشگران بررسی تاثیر زبان اول در استفاده از شاخص های بلاغی در دو محیط فرهنگی متفاوت بود. برای انجام این مطالعه از مدل هایلند که دارای شاخص های مرحلهای، تغییر موضوع، ارجاعی درون متنی، ارجاعی برون متنی، و نشانگرهای اطلاعاتی میباشد ، سود جسته شد. بدین منظور، 40 انشاء نوشته شده توسط دانشجویان فارسی زبان و 40 انشاء نوشته شده توسط دانشجویان چینی زبان به صورت غیر تصادفی انتخاب و مورد بررسی و تجزیه و تحلیل قرار گرفتند. پس از کاربست آزمون مان ویتنی یو - برای مقایسه میانگین استفاده از انواع شاخص های فراگفتمانی - یافته ها نشان داد که به جز یک مولفه، در استفاده از سایر شاخص های فراگفتمانی در نگارش متون توسط دانشجویان ایرانی و چینی رشته زبان انگلیسی تفاوت معنی داری وجود دارد. یافته های پژوهش می تواند موجب تعمیق بینش برنامه ریزان درسی زبان های خارجی در رابطه با ادراک تفاوت های بینافرهنگی در نگارش و استفاده از روش های موثرتر برای تعدیل این تفاوت ها گردد.
As the persuasive dimension of language, rhetoric involves organizing the text in such a way that convinces the audience or elicits their support by establishing the credibility of the reported events and engaging the readers or listeners. Put another way, rhetoric is "the art which seeks to capture in opportune moments that which is appropriate and attempts to suggest that which is possible" (Lucaites, Condit, & Caudill, 1999, p. 26). As Kaplan (1966) argued, the reasoning of a text written by a native speaker is different from those written by nonnative speakers. All the same, the features of each community as motivated by their unique linguistic and cultural traditions cannot be considered as superior over others (Canagarajah, 2002).
The textual-linguistic descriptions offered by contrastive rhetoric (CR) may improve second language writing instruction in two ways. First, teachers could use the empirical findings that CR provided to anticipate some challenges (Canagarajah, 2002). Second, CR findings can facilitate students’ access to language norms by drawing their attention to certain text features and structure (Colombo, 2012). In this way, CR helps to “create and maintain an atmosphere of tolerance for differences in L2 writing” (Leki, 1997, p. 244). Since Kaplan’s (1966) postulation of CR, a large number of studies have been published (e.g., Kaplan, 1966; Ying, 2000). Moreover, many studies have been conducted in this field to compare the characteristics of texts written by nonnative speakers from different cultures to describe the effect of these characteristics on EFL and ESL (e.g., Khodabandeh, Jafarigohar, Soleimani & Hemmati, 2013; Liu, 2007; Rashidi & Dastkhezr, 2009). Furthermore, there are numerous studies that contrasted the writings of Iranian or Chinese students with those of the English native speakers (Biria & Yakhabi, 2013; Faghih & Rahimpour, 2009; Liu, 2007; Noorin & Biria, 2010; Sabzevari & Sadeghi, 2013).
To date, the CR findings have justified the differences in the rhetorical patterns, written by students based on the following reasons: cultural background (Matalene, 1985; Zamel, 1997), educational background (e.g., Yang, 2003), philosophical background (Chen, 2007; Gu, 2008), linguistic background (Connor, 2012; Leki, 1997; Ying, 2000), and developmental factors (Mohan & Lo, 1985). However, in spite of comparative studies between Iran and China in other fields such educational systems (Shekari & Rahimi, 2009), positive phycology (Wang, Derakhshan, & Rahimpour, 2022), and medical education (Amini, Ghahremani, Moosaeifard, & Taghiloo, 2016), there is yawning gap in the CR research, which concerns the role of rhetorical diversities ubiquitous in almost all aspects of L2 learning and teaching (Kuo & Lai 2006).
Given the preceding background, it seems that no single study has ever tried to unearth the rhetorical differences between the native Iranian and Chinese EFL students that have learned English in these two Asian EFL contexts, which have some similarities (e.g., priority of commitment over expertness, curricular change, etc.), and differences (e.g., higher education controlled by ministry of education, admission criteria, etc.) as both of them have experienced the ideologically-driven Cultural Revolution (Sobhe, 1982). Hence, this inspires more studies to conduct cross-cultural studies to investigate whether any similarities and differences exist in the argumentative rhetoric produced by their academia. Therefore, to address these gaps, the current study was an attempt to compare and contrast the types of metadiscourse markers in the English argumentative essays, written by the EFL students from Iran and China.
Metadiscourse, also called metatext or metalanguage in many studies (Bunton, 1998; Farrokhi & Ashrafi, 2009; Mauranen, 1993; Rahman, 2004), is "self-reﬂective linguistic expressions referring to the evolving text, to the writer, and to the imagined readers of that text” (Hyland, 2004, p.133). The term ‘metadiscourse’ was first coined by Harris (1959, as cited in Sultan, 2011), who tried to describe text elements which comment on the main information of a text, but which themselves contain only unessential information. It is based on a view of writing as a social engagement and, in academic contexts, reveals the ways writers project themselves into their discourse to signal their attitudes and commitments (Hyland, 2004).
Because metadiscourse analysis involves taking a functional approach to texts, writers in this area have tended to look to the systemic functional theory of language for insights and theoretical support (Hyland, 2005). According to Halliday and Matthiessen (2014), there are three main language functions, considered as the underlying issue of metadiscourse. Hence, when people produce a message, their speech involved three different kinds of meaning; which is ideational, interpersonal, and textual. As a result, language communication is the product or the result of the process of interplay between these functions of language. Through this interplay, the meaning potential of language is realized. Then, learning a language entails “learning to mean” (Kumaravadivelu, 2006). For Halliday and Matthiessen (2014), textual, interpersonal, and propositional (ideational) elements of the texts are not discrete and separable. As a result, writers should simultaneously create propositional content, interpersonal engagement and the flow of text as they write a discourse. However, the creation of a text is a means of creating both interpersonal and ideational meanings, and textual features cannot be considered as ends. It should be recognized that it is “interaction in a text if metadiscourse is the way writers engage their readers and create convincing and coherent text; it expresses the interpersonal dimension and how both interactive and textual resources are used to produce and continue relations with readers” (Hyland, 2005, p. 27). Furthermore, According to Khedri and Kritsis (2018, p. 51), engagement markers help writers "bring readers into discourse, inviting them into argumentations." Commonly, engagement markers are in the forms of "personal pronouns, directives, asides, and interrogative structures". In the same vein, various frameworks and classifications of metadiscourse have been presented as follows:
Attitudinal metadiscourse, according to Crismore (1983), refers to this point that an author can also explicitly or implicitly signal his attitude toward the content or structure of the preliminary discourse and toward the reader in order to give directives to readers about the importance or salience of certain points, about the degree of certainty he has, about how he feels, and about the distance he wishes to put between himself and the reader. Crismore (1983) used four subtypes of attitudinal metadiscourse:
The basic assumption that convinces researchers to do CR studies and compare the discourse of students of different social groups is that there are some differences among the groups of students in terms of rhetoric. Hence, it is important to identify the main causes of differences between EFL learners’ writing. Yang (2003) identified three main causes of differences in organizational patterns in ESL texts: linguistic, cultural, and educational. Some researchers claimed that the Sapir-Whorf hypothesis of linguistic relatively is the basis of principles of contrastive rhetoric because it suggests a great relation between languages, thoughts and cultures (e.g., Connor, 2012; Leki, 1997; Yang, 2003; Ying, 2000). Sapir-Whorf hypothesis refers to a strong relationship between languages and the society or, in other words, cultural boundaries, which is exactly the main idea of rhetoric.
The next important cause of differences is cultural and logical characteristics of EFL learners. The cultural background of the writer influences his/her writing strongly. Among the Western rhetorical values, the importance of “originality and individuality” and “self-expression and logical argument” in writing is emphasized (Matalene, 1985, p. 790). However, as Chen (2007) argued, under the influence of Confucian traditions, Chinese teachers have always superiority. They have deep knowledge and they are superior to students, as an authority and expert. Teachers transfer their knowledge to students, and students follow them. The Chinese students may feel that they do not know enough to express their ideas or do something original; consequently, they usually quote past masters and authorities. Hence, when Chinese students come to write in English, argumentative discourse might be problematic for them to construct their own ideas in order to persuade readers (Chen, 2007).
Gu (2008) explained that rhetoric is intertwined with and attached to philosophy, religion, ethics, psychology, politics, and social relations. He claimed that the heritage of Western rhetoric owes a great deal to the doctrines of Aristotle and Cicero. However, Buddhism, Confucianism, and Taoism strongly affected the heritage of Chinese rhetoric. Gu (2008) argued that Chinese rhetoric, due to its unique culture, is like a puzzle for western readers. He continued that "Chinese students and scholars feel equally alien to the Western rhetorical tradition when they are challenged to speak in class in a Western institution of higher learning. Some of them even find such a practice frustrating, especially when they are first exposed to the Western culture and the clash between Western and Chinese rhetorical traditions are most apparent" (p.46), a claim even made by Chinese students themselves in their diaries and expressions (e.g., Shen, 1989, as cited in Yang, 2003).
Another cause of differences between EFL learners' writing, according to Yang (2003), is education. Yang (2003) argued that the focus of teachers and educational settings in China is on grammatical structure and at the sentence level; however, in western countries the focus is on organization at the discourse level. Mohan and Lo (1985) suggested that developmental factors may be relevant to organizational problems in academic writing by second language learners. In order to support this claim, they compared the composition practices in Hong Kong and British Columbia. They found out that Chinese school experience with English composition was oriented more toward accuracy at the sentence level than toward the development of appropriate discourse organization. They realized that students also see their writing problems as sentence-level problems.
In the Chinese context, Wu and Yang (2022) compared the use of interactive metadiscourse marker by native English for academic purposes (EAP) teachers in the UK and their non-native counterparts in the Chinese context. The corpus included two sub-corpora, composed of instructor’s contributions to classroom discourse (i.e., eight sessions of EAP lessons from both contexts). They used an interpersonal model of metadiscourse to examine the similarities and differences in their use of interactive metadiscourse in the two sub-corpora. The findings revealed that transition markers and frame markers were heavily used in both contexts to organize the teachers’ lessons. They concluded that this may have been resulted from such factors as logical preferences, development order of acquisition, discourse community, and speech community.
Afzaal, Chishti, Liu, and Zhang (2021) looked at the differences in the use of metadiscourse markers in the introductions of the English and Chinese university students’ theses. To this end, twenty introduction chapters of MA theses were selected. Then, the researchers used Hyland’s model of metadiscourse markers to analyze the corpus of the study. The results of statistical analysis revealed a statistically significant difference in the use of metadiscourse markers used in the MA theses written by the Chinese and English university students.
Furthermore, Mu, Zhang, Ehrich, & Hong (2015) in their research article identified that English RAs differed in the employment of metadiscouse features from Chinese RAs. Hedges were Preferred in English RAs to qualify the claims when making the inferences. Chinese RAs tended to use more evidential, Chinese RAs paid much attention to citing resources in academic writing. Also, Chinese RAs were found to prefer using Boosters and self-mentions.
Li and Wharton (2012) found out that in similar discipline, context is a more powerful factor that influence the use of metadiscourse by students. They argued that UK students employ metadiscourse more frequently than Chinese writers. Students in the UK use smaller account of transition markers than Chinese students. They explained that self-mentions are almost absent in Chinese writings corpus, but are frequent in the essays of UK students. The results of this study also showed that Chinese writers use strong assertions in their rhetoric, and use expressions such as "we must" and "you should" to engage with readers. UK students use more hedges, indicating a preference to diminish their commitment to propositions. UK writers show slightly less use of unquoted evidentials than do Chinese. Li (2011) collected a corpus of article abstracts and indicated that abstracts display differences in the writers' disciplinary and linguistic background.
In the EFL context of Iran, Goltaji and Hooshmand (2022) explored the interactive metadiscourse markers used in the textbooks on language-related issues (i.e., language testing and language materials development), written by native and non-native writers. For the purpose of this study, they selected four technical textbooks, written by the English and Iranian authors and analyzed the corpus using Wordsmith software. Finally, they compared the interactive markers used in the selected corpus based on Hyland's (2005) model. The results of comparative analyses indicated that the transitions were the most frequently used metadiscourse markers. Moreover, the minor variations in the application of the interactive markers were attributed to the differences in the authors’ rhetorical strategies, cultural conventions, and idiosyncratic writing styles.
Soltani and Shokrpour (2021) also investigated the use of metadiscourse markers in four sections (i.e., introduction, methods, results, and discussion) of 120 ISI- indexed English medical research articles, written by the Iranian and non-Iranian authors. Using Hyland’s (2005) taxonomy, they identified the target metadiscourse markers. The results of statistical analysis indicated that there was no significant difference between the metadiscourse markers used by Iranian and non-Iranian authors of the medical papers
Furthermore, Noorian and Biria (2010) realized that the use of hedges, boosters, and attitude markers are similar in both groups. However, there were significant differences between the two groups regarding the occurrences of such interpersonal markers as commentaries and personal markers. Estaji and Vafaeimehr (2015) compared the use of metadiscourse in mechanical and electrical engineering research papers. The results of their finding suggested that there is no significantly difference between these two fields of study in the use of metadiscourse.
Simin and Tavangar (2009) also made an attempt to look at the foreign language learners written products from a pragmatic perspective, focusing on the use of metadiscourse markers. To that end, they asked ninety Iranian EFL students to participate in the study. Based on their Oxford Placement Test (OPT) scores, they were divided into three proficiency groups (i.e., lower-intermediate, intermediate, and upper-intermediate). For a period of one semester, their sample essays, which they wrote on argumentative topics assigned to them, were collected and analyzed. They used Vande Kopple’s (1985) criteria for classification of metadiscourse markers. Then, the number of correct uses of metadiscourse markers was counted and calculated across the given tasks. Running a Chi-square test, the differences in metadiscourse use were shown to be significant for different levels of proficiency. From the above observations, it can be inferred that the more proficient learners are in a second language, the more they use metadiscourse markers. Moreover, it would appear that metadiscourse instruction has a positive effect on the correct use of metadiscourse markers. They also found out that textual metadiscourse is used more than interpersonal metadiscourse by all groups.
Moreover, Zarei and Mansoori (2007) did a quantitative analysis of metadiscourse differences between English and Persian. They concluded in their article that "metadiscourse provides a link between texts and community culture, defining the rhetorical context which is created to conform to the expectations of the audience for whom the text is written." The results of their study suggested that Persian writers employed more metadiscourse elements.
Given the preceding literature review, many researchers and linguists in Iran and China have been interested in doing CR studies. The findings of the previous studies have shown that the use of metadiscourse is different among different sociocultural groups (e.g. Heng & Tan 2010; Mirshamsi & Allami, 2013). Be it so, there are only a few comparative studies between two nonnative groups of EFL students. All the same, most of the researchers have compared a group of nonnative students with native speakers of English (e.g., Marandi, 2003; Jalilifar, 2011). However, it is hard to find a comparative study of compositions written by Persian native speakers and the natives of other languages except English. Hence, the present study tried to address this gap and compare the use of metadiscourse markers by two groups of Iranian and Chinese students. To that end, the following research question was proposed:
Major RQ: Are there any statistically significant differences between the means of interactive metadiscourse markers in argumentative essays written by the native Iranian and Chinese EFL students?
Minor RQ1: Is there any statistically significant difference between the transitions in argumentative essays written by the native Iranian and Chinese EFL students?
Minor RQ2: Is there any statistically significant difference between the frame markers in argumentative essays written by the native Iranian and Chinese EFL students?
Minor RQ3: Is there any statistically significant difference between the endophoric markers in argumentative essays written by the native Iranian and Chinese EFL students?
Minor RQ4: Is there any statistically significant difference between the evidential markers in argumentative essays written by the native Iranian and Chinese EFL students?
Major RQ5: Is there any statistically significant difference between the code glosses in argumentative essays written by the native Iranian and Chinese EFL students?
This non-randomized comparative study is “a combination of qualitative and quantitative research that includes both a quantitative data tabulated to illustrate the use of metatextual features and a description of the types of metatextual features” (Kim & Lim, 2013, p.4). The independent variable of the study was the use of metadiscourse markers by the participants of the study. It is noteworthy that the assumption of the comparability of the genre under study (i.e., argumentative genre) was met to present reliable and valid findings from the corpus, written by two different groups of the participants, without interpreting the probable results as mirroring genre-related differences (Adel, 2006). The decision as to delimit the study in terms of field of study (i.e., English language) was also made on the basis the researchers’ expertise in the field and the availability of both physical and virtual materials.
The sampling technique in this study was convenience sampling because a random sampling was impossible for the researchers. A total of 80 male and female students studying at two universities located in central Iran and East China participated in this study. The reason for choosing the two countries is that they have enough comparable cultural beliefs, outlooks, values, and orientations (Wang, Derakhshan, & Rahimpour, 2022). The Iranian group was comprised of forty students in Qom Islamic Azad University and the Chinese group contained forty students in School of Foreign Studies in An Hui University. The participants were adult university students, whose field of study was English. Both groups of students in Iran and China majored in English language. The age of the students ranged from 19 to 35. The only differences between these two groups were their L1 and C1. The L1 of the students in Iran was Persian, and the L1 of Chinese students was mandarin Chinese. The Chinese participants from An Hui University were accessed through personal contacts by one of the researchers with a faculty member of that university, where the same researcher had already done a BA course in Chinese language. Moreover, the Iranian participants were in the intact class that one of the researchers had in the target university in Iran. The students voluntarily participated in the study. However, the point that must be taken into account is the disproportionate size of the Chinese and Iranian populations and the geographical area of both countries, which may be taken into account as a limitation of the study.
To compare and analyze the differences between metadiscoursal characteristics of the essays written by Chinese and Iranian students in this study, it was essential to have an appropriate model. The instrument that was used during this study was Hyland's (2004) model. Hyland's (2004) taxonomy involves two parts as follows:
The following procedural steps were taken in order to fulfill the requirements of this study:
Initially, the participants of the study were selected through convenience sampling technique from an Iranian university and a Chinese university. Then, every participant was asked to write a five-paragraph argumentative essay in English based on a topic from the second task of an academic IELTS writing module. The participants were given the writing rubrics with its translation into Persian and Chinese. The students wrote the argumentative essay as an in-class assignment in a writing course. The topic was presented to students and the teachers provided the prompts orally.
The data from Chinese students were collected by a Chinese university professor, who taught English language in An Hui University. She voluntarily accepted to collect the data for this study out of her acquaintance with one of the researchers and send the essays to the same researcher via E-mail. The same procedure was followed by one of the principal researchers at the target university in Iran. At the end, eighty argumentative essays were collected from the two groups of this study. Then, two of the principal researchers of this study identified and rated the use of interactive metadiscourse markers in the essays in accordance with Hyland’ model, and scored the papers accordingly. The inter-rater reliability of the ratings was calculated and found to be acceptable (0.88).
Finally, the collected quantitative data were subjected to the statistical analysis with Statistical Package for the Social Sciences version 22. The data were analyzed using Mann-Whitney U Test, which is an appropriate nonparametric statistic to examine the differences between the mean ranks of two independent groups (i.e., Iranian and Chinese groups) on a continuous measure (i.e., the number of each metadiscourse marker used by each group). When the data is not normally distributed, this test is used to convert “the scores on the continuous variable to ranks across the two groups. It then evaluates whether the ranks for the two groups differ significantly. As the scores are converted to ranks, the actual distribution of the scores does not matter (Pallant, 2010, p.227).
In order to choose the appropriate statistical test for comparing the means of the performance of the two independent groups of Iranian and Chinese participants, a test of normality was run (Table 1).
Test of Normality Distribution of Scores of Iranian and Chinese Students in Writing Argumentative Essays
As shown in table 1, the significant values suggest the violation of the assumption of normality. The results of the Shapiro-Wilk test, which is used for small sample sizes, show that the p value is less than 0.05 for all of the interactive resources. Therefore, the nonparametric Mann-Whitney U Test was used to compare the mean ranks of the two groups’ number of metadiscourse markers. Table 2 presents the result of Mann-Whitney U test for the mean of six interactive metadiscourse markers in the written performance of the Iranian and Chinese students in their English argumentative essays.
Mann-Whitney U Test for Interactive Resources
Table 2 shows the p value is less than 0.05 for the mean ranks of the two groups’ use of interactive resources (p=0.008). Therefore, the null hypothesis for the first research question (i.e., Are there any statistically significant differences between the means of interactive metadiscourse markers in argumentative essays written by native Iranian and Chinese EFL students?) is rejected. Moreover, the mean ranks of the performance of the Iranian and Chinese participants’ argumentative essays were significantly different for the following metadiscourse markers: transitions (p = .000), frame markers (p = .000), endophoric (p = .010), and evidential markers (p = .000). Therefore, the null hypotheses of the first four minor research questions are rejected. However, there was no significant difference in the mean ranks of the use of code glosses by the Iranian and Chinese students (p = .24). Therefore, the null hypothesis of the fifth research question (i.e., Is there any statistically significant difference between the code glosses in argumentative essays written by the native Iranian and Chinese EFL students?) is not rejected.
The purpose of this study was to explore whether there were any statistically significant differences between the use of the interactive metadiscourse marker types by Iranian and Chinese EFL students in their argumentative essay writings in English. By adopting a quantitative research design, this study investigated the effect of students' L1 and C1 on the use of metadiscourse in the context of English argumentative writing.
The findings revealed that there were significant differences between the Iranian and Chinese students in writing argumentative essays in terms of interactive resources through transitions, frame markers, endophoric markers and evidential; however, no significant difference was found in the use of code glosses. To be more specific, it was revealed that the distribution of interactive metadiscourse markers across the compositions of students with different mother tongues and different cultural backgrounds vary significantly. As some researchers have expressed concern that the cultural background and first language of writers have an influence on the use of metadiscourse markers (Li & Wharton, 2012; Zarei & Mansoori, 2007), an initial conclusion based on the quantitative analysis in the current research is that both Iranian and Chinese groups used all subtypes of metadiscourse markers in their writings, although the use of metadiscourse markers has different functions depending on the cultural context. This finding may demonstrate the almost universal application of metadiscourse markers.
However, the different application of metadiscourse markers by the Iranian and Chinese students proves the claim about the noticeable influence of local culture on writers’ use of metadiscourse markers. This finding is in line with Zarei and Mansoori's (2007) conclusion which supported the interlingual rhetorical differences in the use of metadiscourse resources in English and Persian articles, Li and Wharton (2012), who found out that in similar discipline, context is a powerful factor that influences the use of metadiscourse markers by students, and Li (2011), who firmly indicated that abstracts display differences in the writers' disciplinary and linguistic background. On the other hand, the results of current research suggested that the presence of a certain type of metadiscourse markers (i.e., code glosses) in the argumentative writings did not vary in the argumentative writings of the Iranian and Chinese students. Similarly, Dehqan and Chalak (2017) found no significant difference between native English and Iranian writers in their use of code glosses. Hence, this can be either the result of their mastery over the use of this metadiscourse marker in English or because of the compatibility of writing conventions in all these cultures having facilitated the mastery of just this variable.
In addition, the results of present research lend support to the applicability of Hyland's (2004) analytic tool for metadiscourse analysis. As CR treats the features of each community as motivated by their linguistic and cultural traditions that one cannot be generalized as superior over others (Canagarajah, 2002), educators can get insight from students’ linguistic and cultural backgrounds and raise their awareness of L2 students’ struggles with language and writing. In this way, CR may help teachers foster tolerance for differences in L2 writing. It is hoped that the results of this study may help researchers to gain a better understanding about the nature of the cross-cultural and cross-linguistic variation in argumentative written discourse.
Additionally, the findings of this study may assist students to find out that their rhetorical choices are not just individual mistakes or errors, but can be related to culturally-based preferences, they can validate their own rhetoric. This prevents students from feeling that they are lacking something when producing texts in their second language. The findings can also facilitate students’ access to language norms by drawing their attention to certain text features and structure. Future researchers are recommended to conduct some mixed-methods studies in order to unearth the cultural and linguistic reasons for the discrepancies and resemblances that exist among various groups with differing levels of L2 proficiency, experience in L1/L2 writing, and interference of L1.