Examining University Students' Scholarly Publication in English Journals: A Case for Postgraduate Students' Written Literacy Practices

Introduction

In English as a Foreign Language (EFL) educational settings, a uni-skill teaching of 'reading' is usually current. In so doing, seeing 'writing' as a neglected skill or act of communication is a common practice. As Tribble (1996) argued, in EFL situations, 'Writing to learn' prevails over 'Learning to write'. In the latter, learners are required to extend their textual knowledge (lexico-grammatical) along with rhetorical (whole-text) abilities, while in the former, which is characteristic of EFL contexts, language learners use L2 (here, English) writing system to practice new language knowledge or use writing to demonstrate their knowledge of language in the context of assessments and they find few opportunities to further extend their literacy in the target language (p. 68-69).

At tertiary levels, the same problem exists in Iranian universities. Disseminating and/or publicizing ideas through a non-native language might be challenging for Iranian university students and/or scholars, who have always been accustomed to 'Writing to learn', EFL trends in their formal, educational settings with an emphasis only on grammatical rules.  This is so while according to recent trends in Academic Literacy (AL) movements, one major goal behind learning English at tertiary levels is 'knowledge dissemination' (Turner, 2012). By definition, knowledge dissemination refers to the exchange of knowledge and information among scholars from different nations. This goal has, at times, been considered as one pertinent mission at academia which acts as a torch for carrying forward academic work (Akin, 2005).

In line with knowledge dissemination trends, a crucial text type at tertiary levels, which is mostly used by academics at various disciplinary fields of study for communicating their scientific achievements is Research Article (RA).

The number of research studies on L2 AL practices targeting RAs within Iranian contexts is few and the most recent research studies, in which 'essay writing' practices in Iran have specifically been the focus signifies a suboptimum condition among both English major and non-major students (Brijandi & Nosrat Nia, 2005; Ghahremani Ghajar & Mirhosseini, 2011; Hasrati, 2015; Jebreil, Azizifar & Gowhari, 2015; Mehrdad & Gazni 2010; Rezaee & Jafari, 2014, etc).

Micro vs. Macro Levels of Essay Writing

Regarding Iranian postgraduate students, within English non-major sciences, there are various ongoing arguments pertained to writing and publishing RAs in English. One line of research over the mostly experienced difficulties at higher levels has been recently focused on two distinct Micro vs. Macro level abilities. By definition, Micro level aspects of writing a research paper are usually defined by the scholars in the field at/below sentence level, which comprise clear sentences (coordination and subordination, parallelism, confused syntax), grammatical sentences (fragments, run-ons, comma splices, pronoun reference, verb forms), punctuation and/or mechanics of writing and richness of vocabulary (Nation, 2009, p. 141). Regarding Macro levels, whole-text, rhetorical aspects of various text types have long been discussed in the literature. In a writing genre such as RA, for instance, Macro aspects or features include 1) Identifying, analyzing and defining a problem to solve, 2) Determining information in a discipline to solve that problem, 3) Collecting data, 4) Offering viable solutions, and finally 5) Evaluating them via four successive sections including 'Introduction', 'Review of Literature', 'Method' and 'Discussion' (IRMD) (Carter, 2007). These four successive courses of actions are believed to be different among diverse university disciplines. As evidence, in Natural and Medical Sciences, a sequential order over major parts of an RA includes IRMD, which must closely be observed by the researchers in order to 1) propose a research hypothesis, 2) test it, 3) report the results, and 4) evaluate the findings. In Humanities, however, IRMD might not be needed to be followed as such. Instead, critically evaluating data sources like documents and authenticating a target context might depend on giving defendable arguments. In other words, the Macro-level aspects of writing an RA consist of the writing process regarding generating of ideas, inspired by ways of knowing as Carter asserted above. Other scholars considered other aspects about Macro levels of an RA, though. As an instance, Swales & Lindemann (2002) included the 'aims and purposes' of writing, 'library searches', 'taking notes' within Macro level abilities of writing RAs and within Micro level abilities, factors such as ‘tense’, ‘citation’, ‘reporting verbs’, and ‘adjuncts of reporting’ were included. In general, they considered Macro functions as broadly representing reading activities, while Micro features denoted mechanics of writing.

In general, research studies over giving prominence to lower-order (Micro) vs. higher-order (Macro) levels of research writing in another language (here, English) are mixed and aligned with compounding results in the literature. In effect, decision over the priority of Micro vs. Macro levels of writing a research paper is unclear (Alnufaie & Grenfell, 2013; Bern Stein, 2006; Hinkel, 2004; Zeng, 2005). Some scholars believe that at undergraduate levels, language levels such as lexis and grammar are in need of refining while at graduate studies such needs tend to be associated with discourse and rhetorical features of English language (Hinkel, 2004). On pedagogical grounds, some scholars referred such complexity to the inter-related nature of thinking processes with language along with motional abilities in first language (L1) contexts (Bern Stein, 2006; Zeng, 2005) and mixing Micro vs. Macro level abilities of writing in L2 contexts (Alnufaie & Grenfell, 2013).

Requirements over the superiority of Micro vs. Macro levels of writing also differ along goals of instructions. In ESL settings, for instance, where students have more access and consequently more control over lower level lexico-grammatical aspects, concerns for higher-order (Macro) levels were felt as more pertinent in the reviewed literature. As an example, Johns (1991) illustrated some ESL aspects of illiteracy at macro levels in an American academic setting including lack of background knowledge, problems in interpreting text macrostructures, lack of planning, conceptual imagination, and unwillingness to be objective about their value systems were just some underlying reasons for this situation. In EFL settings, however, where university students also complain over lower order aspects of language, this issue could be interesting to figure out how the problems at Micro vs. Macro levels might entangle them. The present research has the same concern in an EFL setting.

Before explicating the present research concerns in this regard, it is worth noting micro- vs. macro level arguments related to publishing RAs within wider EFL situations of use which has mostly been discussed by the scholars in the field via two strands: linguistic vs. non-linguistic challenges. As to academic writing practices, literacy scholars have recently focused on both Linguistic reasons regarding L1 vs. L2 incompetency (Brijandi & Nosrat Nia, 2005; Edalat, 2005; Ferris, 2009; Gee, 2012; Grabe & Zhang, 2013; Hasrati, 2015; Nation, 2009; Strickland, 1991; Strickland & Morrow, 1988 etc.) as well as non-linguistic reasons at macro levels (Davies, Swinburne, & Williams, 2006; Ghahremani Ghajar & Mirhosseini, 2011; Hyland, 2015; Jebreil, Azizifar & Gowhari, 2015; Rezaee & Jafari, 2014).

Linguistic Challenges in Writing English RAs

In the reviewed literature, the number of research studies on academic writing practices targeting writing English RAs within Iranian contexts was few and the most recent research studies, in which 'essay writing' practices in Iran had specifically been the focus signified a suboptimum condition. In an extensive research study, covering a twenty-year span from 1990-2010, Mehrad and Gazni (2010) alleged that Iran's publication share in major databases like Institute for Scientific Information (ISI) had been only 42%. The rest (58%) of the interested researchers had found their way in Islamic Science Citation (ISC) databases. This situation in some university disciplines like Chemistry, Physics, Materials Science, Mathematics, Plant and Animal Sciences, Geosciences and Engineering sciences had been more optimum compared with Social Sciences and Humanities (SSH) fields. In other words, publication share in the latter forefront-students and scholars in SSH- had been to the minimum compared with the former. They claimed that, in general, 'language barrier' at Micro levels had been one major problem among the above target group and contended:

"Thus, considering language barrier in ISI Web of Science, {Instead} ISC has provided appropriate opportunity and proper channel for indexing and analyzing scientific publications, and for measuring scientific impact of the Islamic countries" (p.44).

Regarding AL skills and abilities in Iranian EFL academic settings, Hasrati (2015) did a comprehensive research on the status of literacy practices among Iranian academics at Higher Education (HE) with a specific focus on their L1 writing in the country and found that the situation is far from satisfactory. He contended that Iranian university professors in different disciplines had followed different strands in their writing assignments, which could hardly be said to involve writing for academic purposes at all. Engineering students, for examples, claimed that they had done their assignments through progress reports to their professors, which did not need any writing in the real sense of scripting and/or productive practice. As evidence, they had just reported their projects to their professors by tracking some computer screen signs and printing what had been done, in the end. In another case, Geography students claimed that their assignments regarding literacy practices at college had mainly included delivering 'map reading' and designing but not putting any words on paper (p.117). Hasrati argued that description and discussion of ideas had nearly no place in the L1 literacy courses of action at Iranian universities. This study by Hasrati signified how 'writing' is neglected in the Iranian AL curriculum even in their L1 contexts.

In an earlier study by Birjandi and Nosrat Nia (2005), this suboptimum situation found by Hasrati had also been confirmed in an Iranian context among English-major (TEFL) university students to be related to their AL abilities in L1. Birjandi and Nosrat Nia had confirmed that students with more advanced writing acts in their L2 (English), had the same quality in their L1 writing abilities as to paragraph writing, writing summaries and/or paraphrases.

In another research study by Edalat (2005), the root sources of the major essay writing problems among Iranian students were proved to be associated with the effects of the learners' L1 as the main cause of inability to attain maximum fluency in L2. The author had analyzed a group of university students' essay writing at sentence and paragraph level. Having classified the errors (of Interlingual errors kinds) made in the process of sample writings, the author had analyzed the errors in four distinct areas involving wrong word choice, ambiguous English structures caused by Persian expressions, incomplete use of English sentence formation including modifiers and coordinators, which were thought to have been all originated from their L1.

Regarding linguistic reasons comparing L1 with L2 settings outside Iranian contexts, there was a large number of studies, which had focused on how L2 contexts may create a different set of criteria for the learners to be more successful because the nature of shortfalls in the two contexts is gravely different. Raimes (1994, cited in Hinkel, 2004) reported that in an L1 situation, deficits in 'essay writing' could be attributed more to the fluency and conventions of expository discourse, while in L2 writing, learners require a developed L2 proficiency, along with discourse conventions and 'organizing the information flow' (p.10). Later, this idea was called into question by Cummins (1991) in that L2 proficiency could not lead to a better writing in the same language. Bruce and Lewkowicz (1991) also had a similar result for reverting a focused view from lower level features and contended:

“it may be true that L2 learners have problems at the lexico-grammatical level, it is not at all clear that remediation at that level is likely to improve their ability to write coherent responses to academic essay assignments” (p. 363).

In a study by Johns (1991, cited in Hinkel, 2004) it was asserted that "if you can write [or read] an essay, you can write [or read] anything" (p. 122).  The author then concluded that doing the same kind of writing for a long time might affect the performance of the writers in that they can excel in some certain tasks of writing compared with other tasks. He maintained:

When NN students are exposed to largely one type of writing task, they come to believe that "this is the only way to write. Such limited experience with writing actually does students a disservice and causes problems in their academic and professional careers (p. 25).

Non-Linguistic Challenges in Writing English RAs

Among non-linguistic strands, diverse reasons were prevailing in the literature including psychological (Horwitz, Horwitz, & Cope, 1986; Jebreil, Azizifar & Gowhari, 2015; Rezaee & Jafari, 2014) and pedagogical (Ghahremani Ghajar & Mirhosseini, 2011) strands.

On psychological accounts, in a recent attempt by Jebreil et al, it was noted that EFL learner-writers' anxiety levels, in general, might have influenced their writing performance. This study was in line with Horwitz et al in which they had also claimed that although language-anxious students study more compared with their classmates with lower indices of anxiety, their level of achievement does not reflect that effort at all. Factors such as 'communication anxiety', 'test anxiety', and 'fear of negative evaluation' were thought to impede those involved L2 writers to produce sound and well-designed sentences. This was in line with the study by Rezaee and Jafari (2014) who attributed the causes of writing in EFL contexts to writing anxiety such as 'preoccupation with performance', 'high expectations', 'fear of teacher’s negative feedback', 'low self-confidence', though they also added 'poor linguistic knowledge' to the list in the end.

In some recent studies, the degree of writing anxiety had been associated with learner-writers' writing self-efficacy levels, which was thought to be initiated from their writing and reading performance or behavior at first stance (Tanyer, 2015).These and other similar studies denote how linguistic and non-linguistic features can be related though, and have interactive features to be pondered upon.

Associated with pedagogical reasons behind inadequacies in producing an RA, some other scholars had discerned suboptimum condition regarding literacy practices in academic settings related to ways of knowing. The reasons had been sought in educational practices employed by literacy teachers. In a recent study, a pedagogical undertaking of the writing teacher was related to 'essay writing' practices themselves as being not right. Mistaken methodologies, which were not conductive of proper 'essay writing' tasks, were believed as having had de-skilled university students in producing a proper piece of writing at academia. Ghahremani Ghajar and Mirhosseini's (2011) study was a case in point. In this article, the fake roots of research writing practices based on decolonization aims of some over-indulgent imperialists in the world of academia had been sought in entirety within ELT domains. Research methodology courses based on neutral positivistic views, not relevant to real pains of Iranian educational contexts for language learning and teaching had been criticized not to have had inspired Iranian students in their writing courses. This last research paper gave prominence to the role that higher level thinking abilities at Macro levels could play in the writing abilities of university students to flourish their competencies. This was in line with Lee and Street's study (1998) and Carter (2007). In response to a project supported by Economic and Social Research Council (ESRC), Lee and Street had explored the reasons for students' failing at literacy abilities including 'writing' at HE in the United Kingdom. They pursued the issue in the students' and the academic staff members' perception in an ethnographic design to find what it meant to be academically literate. In the end, they contended that many of the writing difficulties university students had experienced were due to Macro level abilities pertained to ways of knowing including 'the conflicting and contrasting requirements for writing on different courses and from the fact that these requirements were frequently left implicit {emphasis mine}(p. 162). On the other hand, their tutors complained that although they knew what they expected from their students, this could not be expressed clearly in their approaches toward how of writing. For instance, the degree to which teachers gave prominence to surface structure linguistic elements was in some cases concealed if a student had been able to write critically and argue well.  Accordingly, their teachers reacted by giving positive feedback to a student even if s/he had had many linguistic problems at surface levels of the sentence.

They brought the impressions of a university lecturer about student writings in this regard as telling evidences:

I need my students to have an introduction which sets the scene and a main body which covers a number of issues highlighted in the introduction and introduces economic theory, application and analysis. Students need to be critical, to evaluate, to try and reach some sort of synthesis and then to simply summarize and conclude. You need a good solid introduction leading into your main body and each part of your main body will be crafted and it will link with the next. It will have a professional feel about it and will not describe but will critically analyze and then it will lead into a summary and conclusion (p.163).

They concluded that 'elements of successful student writing are in essence related to particular ways of constructing the world and not to a set of generic writing skills', (p. 163). This could signify, in itself, that lower-order abilities at Micro level (lexico-grammatical) elements of language could eventually be inundated by Macro level abilities. Along the same line, Hyland (2015) believed that within those researchers from native users, one could even find problems. He thought regarding scholarly publications at international levels, Native/Nonnative competencies might, at times, be obscured by the dichotomous native vs. Non-Native users of the language. He talked of a special competence, which to him was needed if any scholar wanted to participate in knowledge dissemination practices.

Writing for publication is a specialized competence which both Native and non-Native English speakers must acquire, a fact which is obscured by two key assumptions of the linguistic disadvantage orthodoxy (p. 61).

This issue created a second concern in this research project to see through how micro-macro levels of writing had blocked Iranian scholars at postgraduate levels in L1 and L2 comparatively.

Another research strand regarding essay writing at academic levels had focused on the extent of written literacy practices. This last point added another loop to be considered for exploring the issue in more depth. First, a brief description is also given for a clear picture of what has been involved in the literature regarding this third concern.

Reading and Writing as a Nexus Network

Scholars in the reading-writing nexus lines of enquiry believe that the reading and writing abilities grow together and/or in tandem (Nation, 2009). In recent years, language scholars have linked AL lines of research to New Literacy Studies for schools (Gee, 2012) where innumerable courses of actions for accelerating written acts of learning at academia including 'Reading' and 'Writing' for students studying at HE are taken into consideration in tandem with some higher-order levels of thinking. Regarding 'essay writing' skills, some other scholars also had the same view. They believed that L2 writers at academic settings could alleviate their problems by reading a good deal of texts written by native speakers (Grabe & Zhang, 2013; Nation, 2009; Strickland, 1991; Strickland & Morrow, 1988). Bringing a diverse set of reading-writing activities and tasks in his studies, Nation proved how 'Reading like a writer' was in effect where the learner reads an article or text like the one s/he wants to write. He asserted that while reading, the learners write for instance the questions that a typical writer had seemed to be answering in an essay.

Ferris (2009) noted lack of integrated reading-writing tasks in L1 situations as probable reasons behind the suboptimum condition of L2 English for Academic Purposes (EAP). Inspirations from this and similar research articles may, in and of itself, signify the fact that one main source of such difficulties might be the learners' L1 literacy.  In L1 contexts, at arrival to college, it seems crucial that the nature of writing practices alters due to new requirements that academia sets. Accordingly, writing assignments at college level might turn into more complex types like producing Summary writing, Precise writing, Report writing, Reviewing and Editing other people's writing as well as 'essay writing', etc. By intuition, in L1 contexts, the situation should not create many difficulties for the students; nevertheless, research findings refer to a suboptimum status even in L1 contexts in recent years.

Grabe and Zhang (2013), on the other hand, focused on challenges in integrating reading and writing tasks in academic settings. They felt that limited vocabulary knowledge and grammatical accuracy, lack of L2 intuitive knowledge, less extensive reading practices, weak and in many cases, varying proficiencies in reading and writing among L2 students and Less cultural knowledge in their L2 were just some factors having obstructed writing teachers in integrating reading with writing tasks at tertiary levels (p. 11).

Overall, it seems that reading-writing integration might have helped L2 writers in alleviating the suboptimum condition of 'essay writing' abilities.  In the surveyed literature, research studies on locating the roots of essay writing problems within EFL non-major students esp. at postgraduate levels contexts was scarce in this regard.

In line with these recent enquiries of research in AL, the main focus in this study was directed toward determining in what literacy practices (reading vs. writing) Iranian, English non-major students at postgraduate levels had observed the most difficulties in writing for publication as well as general aims in both English and Persian languages. Ultimately, two research questions were suggested in this study:

Research Questions

1- In which major Micro (lexico-grammatical) vs. Macro (rhetorical or whole-text) levels of 'essay writing' had Iranian EFL Non-major university students at postgraduate levels (MA and PhD) already received the most revising comments by journal reviewers in the mainstream, English journals?

2- In what written literacy abilities including writing as a production and reading as a recognition act of communication, had Iranian EFL Non-major university students at postgraduate levels (MA and PhD) often perceived the most difficulties in using them for General and Academic purposes in Persian and English languages?

Method

This study had an etiological taint to it and via a survey-led study, the authors focused on those 'essay writing' arenas wherein there are the most reported difficulties by the targeted population. Responses to each question above could be insightful in knowing what linguistic strands (L1 or L2) at (Micro vs. Macro) levels, Iranian EFL, non-major postgraduate students had experienced the most difficulties in writing for the two aims or language uses (General vs. Academic). Two measures were used to assess written literacy competencies (reading and writing) across language uses: A) participants' assumed, self-reported written literacy competencies in using English and Persian languages for General and Academic purposes were denoted as 'perceived' measures and B) further supported by actual measures: mostly received comments from reviewers in the mainstream, English journals by the target group above. Response to the first research question provided evidences as 'actual' and response to the second question provided evidences as 'perceived' measures.

Participants

At the outset, an enquiry was made over some one hundred Iranian English Non-major scholars, out of whom, fifty-five were randomly selected as Iranian students at MA (24%), PhD (74%) and Post-doctorate (2%) levels from among both gender with Male (74%) and female (26%) scholars from diverse universities (public and private) and across diverse university disciplines (Humanities, Engineering, Medicine and Basic Sciences) around the country. Random stratified sampling from the volunteering respondents to the group-administered questionnaire in this study was applied conveniently only over university discipline strata in order to have diverse discipline groups from among Iranian English non-major university students at post-graduate levels. In line with the purposes of this study, university discipline was more important variable compared with university type and participants' gender. Nevertheless, in the original, wider, one-hundred students to whom inquiry was sent, other variables including their university type (public, private), their educational level (MA, PhD and Post-doctorate) and gender (Male vs. female) was considered to include selected participants as diversely as possible.  In the sampling stage, stratified selection was randomly made over those who had already published their articles (written in English) at international, peer-reviewed journals or had frequently attempted in this regard but their article had been rejected during the blind-reviewing processes. Since the participants were also to be screened over their experiences in publishing Persian articles inside the country, some filled questionnaires, in which they had announced as having no Persian articles, had to be discarded. This decreased the number of participants. Nevertheless, this could be noteworthy which had to be checked otherwise as to the underlying reasons why this target group had managed to publish their scientific achievements in their disciplines only in English.

At initial screening, about 60% of the respondents announced that they had managed to publish between 1-5 articles in English journals, 25.5% had more than five English articles, and 14.5% had not been finally able in publishing their works at international journals. Some 36.4% had managed to publish less than five articles in Persian, 54.5% had more than five published articles in Persian and 9.1% had not published any articles in Persian journals.

After initial screening, fifty-five volunteering, randomly selected participants cooperated with us in sending their attitudes. Their age range was 30-52. Among the respondents, 63.6% were studying inside and 36.4 % outside Iranian universities. Table 1 below is a chart of participants in terms of their first language backgrounds and major fields of studies.

Table 1. Participants' first language varieties and their major fields of study

Instrumentation

In this research, a group-administered questionnaire helped the present researchers in responding to the two proposed questions. The utilized questionnaire (Appendix A) was an adapted version pertained to an intercultural study of academic discourse among an assimilated target group known as ENEIDA project in Spanish. The original questionnaire had been adapted from Spanish to English by Moreno (2011). The questionnaire was tailored for the aims of this study, translated into Persian by the first author of the present research and crosschecked by the first co-author as well as two PhD candidates in TEFL to ensure that the same propositions had been included in the translated version to Persian. The original questionnaire had been devised for online administration and dealt with post-doctoral Spanish learners of English, whose aims were using English for Research Publication Purposes (ERPP). In view of that, on validation accounts, the same reported procedures for designing the items in ENEIDA project could be relied on for our aims as well. Nevertheless, to clarify our stance, a very brief account on changes done through item construction processes and phases and specific procedures for construct and content validation of this questionnaire are first given for a quick reviewing.

Outlining ENEIDA Questionnaire

In this questionnaire, a large group of researchers had already been involved in constructing the items from multiple interrelated sources and from various multidisciplinary informants for investigating Spanish researchers’ writing difficulties as to publishing in English-medium international journals from intercultural and cross-cultural perspectives. Their team involved a group of scholars from Applied Linguists who had been involved for years in EAP courses and researchers who had already analyzed learners’ interlanguage/ errors in their studies. These researchers had in their experiences 'Genre analysis' of academic discourses in English and Spanish, using research methods in inter-cultural and cross-cultural studies of academic discourse, ethnographically-oriented methods for the study of academic discourse and of the research activities of Spanish scientists, sociolinguistic studies, psycholinguistic research studies, experience as members of journals’ scientific committees and editorial boards in various fields.

In constructing the items of this questionnaire, a large-scale online survey had first been conducted aiming at creating a database of the main writing difficulties that Spanish scholars had felt in their publication tasks. Various variables, which were thought to influence writing and learning to write for ERPP, had been taken into consideration such as the researcher’s first language, age, gender, disciplinary field, research qualifications, type of institution where the researcher was employed, etc.

The tailored questionnaire for the present research was calibrated in three main sections in order to find some possible responses for our two research questions. Section A rated the participants' demographic data including their professional and personal information. Section B measured their self-reported academic vs. non-academic skills and abilities in using English and Persian languages for the purposes of our research. In so doing, all specific examples of skills and abilities had been specified for the respondents to self-rate their proficiency in using English and Persian for both General and Specific uses. In this research, language for General purposes (GP) uses signified using the two languages (Persian and English) for non-academic aims. For reading, language uses such as reading short stories, newspaper, and magazines and in writing, language uses such as diary writing, letter writing etc., were suggested within general uses to the respondents to self-rate themselves. By the same token, Language for Academic uses in the two written literacy skills, on the other hand, denoted using Persian and English for Academic purposes (AP) such as, reading articles, searching through English databases, corresponding with editors and reviewers of foreign journals etc. Finally, section C rated their experiences in publishing English articles in foreign journals under the study. In all, in each case, various examples regarding academic communications, writing to editorials, etc. were given to the respondents so that they have a clear view on rating their abilities. In line with the purposes of this study, only the results from part 3 and 4 of this section have been reported here.

The adapted ENEIDA questionnaire was devised in Persian so that respondents could understand the items with ease.

Construct validation had been managed through a two-step procedure based on a qualitative (respondent debriefing) and quantitative approach (a pre-test survey) which ensured a diverse group of experts and knowledge area by Moreno, Rey-Rocha, Burgess, Martín-Martín, Gea-Valor, López-Navarro, I.,Garzón, & Sachdev (2011). The original Spanish version of the questionnaire had been validated among a sample of 200 informants and then piloted among a population of staff with doctorates (n = 8794) for reliability aims. Accordingly, since it had been piloted among a large sample, its validity and reliability indices were taken for granted as suitable for our aims in this study. In addition, since the original questionnaire had been devised for an EFL context like Spain, we thought the propositions could also be utilized for our purposes in our Iranian setting as EFL^*.

Data Analysis Procedures

At initial stages, in order to gain access to the target group, the utilized questionnaire was distributed via a group-administered strategy among English non-major MA and PhD students in various universities and among different university disciplines (Humanities, Engineering, Medicine and Basic Sciences) and diverse university settings (Public and Private) across the country, who had publications in foreign journals across five years from 2010 to 2015. Apart from distributing the adapted ENEIDA questionnaire through both email groups, the authors used online research networks such as 'Research Gate' available at www.researchgate. net in order to find the focus groups. This website was helpful as a major source of data collection to gain access to a large number of Iranian scholars in this regard.

Self-reported measures of the participants over their competencies in written abilities were relied upon as 'perceived' vs. 'actual' measures drawing on Gardner (1985, cited in Moreno et al, 2012) who had approved' self-reported measures of proficiency correlated with objective measures very well' (p. 165).

Findings and Discussion

To compare and contrast self-reported measures within participants' written literacy abilities in the two English vs. Persian language across both GP and AP, a rigorous statistical analysis

^*For further details about this initial implementation and descriptive data for most variables, interested readers may refer to http://hdl.handle.net/10612/1824

catering for simultaneous three by three matrix over language uses (GP vs. AP), language variety (English vs. Persian) and written literacy abilities (reading vs. writing) could be done over the whole dataset, however for clarifying each variable in more depth via qualitative strands and since attitudes and not precise measures had been sought, just frequency and percentage rates are presented here to have an overall estimated view. Data gained at this stage were analyzed through SPSS statistical package ver. 21.

Interpreting Perceived Abilities as to Written Literacy Competencies in Persian

To reiterate, in line with the posed research questions in this study, we aimed at screening actual vs. perceived self-rated measures of a group of English, non-major university students at post-graduate levels. The respondents' written literacy abilities (reading and writing) for General and Academic purposes in the two languages (Persian and English) were mapped over some actual measures involving micro vs. macro levels of literacy skills via inquiring about their mostly received revising comments from scholarly peer-reviewed journals in which they had publications.

In keeping with the administered questionnaire, the results pertained to the second research question are first reported. To recap, this question dealt with respondents' written literacy abilities including writing as a production and reading as a recognition act of communication across General vs. Academic purposes in Persian and English languages. Here, individual uses of each written literacy skill (reading and writing) across General vs. Academic Purposes for the two Persian and English languages are reported separately below.

In the first section of the tailored questionnaire, respondents' self rating of their abilities regarding language for GP vs. AP uses across two written literacy competencies were checked in the two Persian vs. English languages. Respondents had to self rate their abilities in a Likert Scale of Very low (1), Low (2), Medium (3), High(4), to Very high(5). First, discerned difficulties in Persian vs. English across GP vs. AP uses were compared for reading. Table 2 displays frequency counts and percentage rates for perceived reading abilities in Persian language across GP vs. AP.

Table 2. Perceived reading abilities of postgraduate students in Persian language across GP vs. AP

According to Table 2, ratings were clue for an overall better perceived condition of reading abilities for General compared with Academic purposes in Persian. The majority of respondents with 81.8% (n=45) had rated their competency in reading for GP in 'Very high' scale while this had a lower rating index for AP with 76.4% (n=42). At the surface, this could signify, in itself, that in reading skills, postgraduate students felt to be less proficient in reading for academic aims like locating essential disciplinary knowledge in their majors, comprehending their own specialized materials on the internet, compared with general aims like reading short stories, newspaper, magazines etc. in Persian. The same comparison was made for writing abilities in the same language to see if the same condition was prevailing for the supremacy of General over Academic aims in both written literacy skills and abilities or not. Table 3 displays frequency counts and percentage rates in Persian language across GP vs. AP for perceived abilities as to writing.

Table 3. Perceived writing abilities of postgraduate students in Persian language across GP vs. AP

As Table 3 clearly indicates, the difference between GP and AP abilities had been estimated less compared with reading with Academic 63.6% (n=35) a little bit less estimated competency compared with General uses 70.9% (n=39) in the 'Very high' scale.

At this stage, it was possible to compare reading with writing at language uses to find out at which languages uses and what literacy skills, the most rated abilities had been documented by the respondents. A quick look at Table 2 and 3 revealed that reading for GP (81.8%), reading for AP (76.4%), writing for GP (70.95) and writing for AP (69.1%) had the most to the least reported rate counts respectively. At this point, one could possibly figure out that 1) writing abilities had a lower status as a whole compared with reading, 2) in specific uses such as using literacy skills for Academic purposes, respondents felt more difficulties in Persian.

Interpreting Perceived Abilities as to Written Literacy Competencies in English

The same comparisons across language uses and literacy abilities were made for English language as well to see if the same supremacy of reading over writing was discerned for English as well. Table 4 displays distribution of self-rated counts and frequency proportions for reading aims (GP vs. AP) in English.

Table 4. Perceived reading abilities of postgraduate students in English language across GP vs. AP

As Table 4 displays, in all, the lower scales below 'high' and 'very high' have been filled more compared with Persian language. This could clearly show how lower self-rating scales had been spotted for English compared with Persian language in reading skills. In order to figure out the distinctive superiority of each language use (GP vs. AP), the most rated frequency counts were again considered. As clear, for reading skills in English, 43.6% (n=24) had rated themselves in the 'High' scale regarding GP while this rate for AP was observed to have been loaded in 'Very high' for AP. This showed that although reading skills had been discerned to be toward higher scales, in AP, this supremacy had been felt to be more for Academic purposes rather than General uses within the respondents' competencies. So, it could be said that, overall, postgraduate students believed they were able to act better in their disciplinary fields compared with reading texts of general interests like reading English magazines, English newspapers, non-specialized books in English, etc. This could also show how focused view they might have had over their disciplinary knowledge compared with other areas of general interest in English language. In other words, they might have usually preferred themselves better in reading in English for their disciplines for Academic aims and not general reading comprehension skills and abilities. At this point, we also checked collected datasets for writing abilities in English to see if similar patterns could be discerned. Table 5 displays self-rated counts and frequency rates for writing in English across GP and AP.

Table 5. Perceived writing abilities of postgraduate students in English language across GP vs. AP

As Table 5 illustrates, the proportions of self ratings in writing abilities within English language revealed that for GP, the majority of respondents with 38.2% (n=21) had announced that their abilities in writing for General aims was average (medium), while for AP, this proportion had been mainly loaded more on 'High' scale (29.1%), which could denote that generally speaking writing for Academic aims had had a better status compared with GP uses. This was in line with the same discerned pattern for reading abilities in English in that for academic uses, the target group had felt to have been more proficient. The fact that was pertinent here was that, in all, in both reading and writing aims, their status had been estimated to be higher in Academic compared with General uses in English language. In all, in English language, reading and writing for General aims were not discerned to have had an optimum status. Possible routes of the problems in writing for Academic purposes could perhaps then be sought in reading and writing for General aims in which case micro features of language are solidified.

In order to find out how Persian and English languages had been perceived to be different or similar across the two GP and AP, some other comparisons this time regardless of language uses could be made.

Inferring Discerned Difficulties in Written Literacy Competencies in Persian and English

Overall, above interpretations from self-rated competencies could in some way signify to what extent and in what written literacy skills across AP and GP the target group had probably perceived their abilities to be more deficient however, in order to have a better overview over L1 contexts of use compared with English, another set of comparisons could be used this time for languages (Persian vs. English) themselves. Firstly, altogether, at this stage, regardless of language uses, it became clear that Persian speakers had regarded their overall linguistic knowledge in the two reading and writing competencies in their first language to be better than English. In an EFL context like Iran, this could obviously be expected to happen and need not be mentioning here but a crucial point to be considered at this stage was that participants had rated their abilities in line with what we had expected to happen. This could ensure us to initiate the examination of data still further. At this stage, to reiterate, the purpose of study, in line with research questions no.2, was to determine in what written literacy abilities including writing as a production and reading as a recognition act of communication, Iranian English Non-major university students at postgraduate levels (MA and PhD) perceived the most difficulties in the two Persian and English languages. Therefore, another set of comparisons were also made over the datasets to compare English with Persian over the targeted language uses.

Interpreting Perceived Abilities in Written Literacy Skills in English vs. Persian

At this stage, based on the above-cited frequency counts, in general, the following inferences could also be made on the discerned difficulties over the written literacy abilities, this time comparing and contrasting the variables consecutively in the two English vs. Persian languages. For brevity reasons, case comparisons where some major contrastive dissimilarities were seen in frequency counts are reported below. At this point, in order to have an overall view over all three dyad variables above simultaneously, a mean comparison was also run on the dataset. It was assumed that this overall view could also display part of the realities in the targeted sample participants. Accordingly, Table 6 below summarizes comparisons over mean and standard deviations as to reading and writing in each GP vs. AP and in English and Persian languages.

Table 6. Descriptive Statistics for the GP vs. AP abilities in Persian and English across written literacy abilities (Reading and Writing)

As Table 6 evidently displays, an overall view showed how reading estimates have been rated higher than writing in both Persian and English languages. This is also happening in writing abilities in the two languages. Here, we could also see how each literacy skill had been estimated lower across both language uses as well as in the two languages. This comparison seemed to be quite probable since in Iranian communities, Reading Approach is current in mainstream, formal L2 education and writing as another act of communication is neglected in the majority of cases. Table 5 above displayed the tallied frequencies for writing uses for GP, for instance, within the perceived abilities of the target community in English. As clear, some 27.3 % i.e., one-third of the screened respondents had asserted that their writing abilities in English for GP had been less than medium. This could be a case in point here note-worthy in itself for the claims above.

Some other implications over (within and between) comparisons in Persian and English were also probable that seemed worthy of mentioning. Below, some general implications are given in this regard to compare English with Persian.

Reading for AP and GP in Persian and English was not Perceived to be the same

An initial mean comparison over Table 6 indicates that perceived reading abilities for GP and AP in Persian had been rated as nearly the same (4.78 vs. 4.70), respectively. This could indicate in itself that regardless of English language, reading for GP and AP had been following a similar discerned condition. It was interesting to know if the same condition was prevailing in English as well. Initial mean comparison for GP (3.68) vs. AP (3.98) uses, in English displayed, however, a rather more suboptimum condition compared with Persian with AP being more highly rated than GP. This could indicate partly that the target community considered themselves better readers for their disciplinary fields rather than reading for general aims as it was mentioned earlier in this study. This could also be noteworthy, in itself and thoughtfully but not assuredly it could mean that in English, AP uses have been valued higher compared with GP. Probable reasons could be sought among the sampled participants who all had English publications. This could indicate that since they were involved in lots of reading in their fields of study before starting to write in English, reading for AP had been perceived to be more optimum compared with GP reading to their view. This should be notable regarding recent literacy education trends in which reading and writing are considered to be interacting with one another. Recent literacy trends appreciate Reading and Writing acts of communication as a nexus (Strickland, 1991; Strickland and Morrow, 1988; Nation, 2009). They signify writing and reading are related to one another to a great extent. People who read a lot have a much easier time getting better at writing. In order to write a particular kind of text, it helps if the writer has read that kind of text but this should not be thought as if reading for GP cannot be considered important. Here, it was interesting to know that if to the respondents' views, reading for AP was perceived to be easier than GP in the same language, it has also been the case that better discerned abilities in reading for AP, for example, could have influenced the respondents to be better writers for AP or not? So, another comparison in the dataset could be run again. In the next part, this contrast was conspicuous.

Writing for AP and GP in English and Persian was not Perceived to be the same

Again, a quick look over mean distributions in Table 6 displayed unlike perceived competencies for reading, writing abilities of the respondents for GP in Persian (4.62) was perceived to be higher than their AP writing abilities in the same language (4.56), but this was not the case with English. In other words, AP writing in English (3.32) had been rated higher than GP writing (3.04) in the same language. This could indicate that Iranian scholars were probably experiencing more difficulty concerning writing for academia within their own native contexts. However, quite interestingly, in English they had asserted that their AP writing skills and abilities were higher than those of GP. This had also nevertheless been likely to happen since in Iranian academic settings, various researchers have to communicate in English for knowledge dissemination if they want to get published at international levels but writing for GP in English, due to not being ESL, is not compulsory in Iran and scarcely taught among English Non-major participants of the present research.

Concerning Academic vs. General writing abilities in Persian, it was noted that for Iranian scholars, writing for GP in Persian language with (µ= 4.62) had been perceived as easier than AP (µ= 4.56) in the same language, but in English this pattern was not the same with writing for AP (µ=3.32) presiding over GP writing (µ=3.04). In ENEIDA project by Moreno et al (2011) Spanish researchers had perceived their writing level of proficiency in 'Academic' 25.6% (1.28 points) lower than their writing level of proficiency in Spanish for General aims. In the present study, Iranian scholars had rated their abilities in Academic better than general purposes both in writing and reading in English language. In writing for 'General' purposes, this could probably indicate the Iranian respondents' root of problems in English language in the deficiencies they felt in their overall writing abilities due to excluded writing practices in their content courses at both academic and non-academic settings deeply-rooted in their previous educational settings before entering the college in line with Hasrati's study (2015). This was note-worthy since General abilities as basic and in many cases a pre-requisite for initiating practice in Academic skills and abilities should be noted by the scholars in the fields of written literacy abilities for Iranian scholars.

This line of research could still be ripe for further exploration for possible other interpretations but for the sake of brevity, just main interpretations over overall perceived abilities were discussed here. In the next section and in line with the research question no.1, actual written literacy difficulties of the participants regarding 'essay writing' practices were interpreted in the light of gained evidences in the next part of ENEIDA questionnaire. To reiterate, at this stage, the present researchers intended to see in which major Micro (lexico-grammatical) vs. Macro (rhetorical or whole-text) levels of 'essay writing' practices, Iranian English Non-major university students at postgraduate levels (MA and PhD) had received the most revising comments by mainstream journal reviewers.

Interpreting Actual Areas of Difficulty as to 'Essay Writing'

In response to the first research question, section C of the ENEIDA questionnaire sought the respondents' experiences in publishing their studies in various foreign venues like English journals.

Data mining as such could give us more valid and reliable responses since they had been approved by a diverse group of involved participants and across a wide time span in their professional life history. Table 7 below displays the most reported problematic parts within listed subcategories of RA Micro vs. macro levels. Item numbers 4 and 5 in the distributed questionnaire denoted Micro, and the rest of suggested items were counted as data denoting Macro levels.

Table 7. The tallied frequencies for Iranian postgraduate university students' actual linguistic difficulties at Micro and Macro levels in essay writing practices

From among proposed areas of difficulty concerning writing activities for scholarly publication, it became clear that grammatical flaws (69.4%) at Micro, and lack of fluency within Macro levels (55.1) had caused the most difficulties respectively for the scholars in this study. The next order of complications had been announced with lesser extents as to 'writing to editors in English' (36%) and 'formatting' (24.5%).

Item no. 4 in section c of the questionnaire required the respondents to postulate their views over the major areas of a research article, which had caused more difficulty for them during their research publication experience. Data gained from this section provided further data for Micro vs. Macro linguistic levels as to essay writing practices within the target group of this study. Table 8 displays tallied frequencies in this regard.

Table 8. The most problematic parts to write in RAs

As clear in Table 8, the most problematic part to write in an RA had been the 'Discussion' section (62%). This was in line with Nesi and Gardner's (2006) and Bacha's (2010) studies in that many students have difficulty in critically discussing of their arguments. Possible reasons for rhetorical moves regarding argumentation (Wingate, 2012) as a crucial genre in 'Discussion' section of an RA should be considered in further research studies.

Conclusion

This study was undertaken as part of a multiphase larger-scale inquiry pertained to the Micro/Macro incompetency in writing English RAs on the part of Iranian postgraduate university students in their Persian vs. English competencies for essay writing practices.

Various inferences over each dyads of the three intended variables including language (Persian vs. English), language uses (GP vs. AP), and written literacy skills (Reading vs. writing) revealed various implications that were described in full in the previous section. In short, regarding academic practices of the target group in this study, however, it seems that in all reading-writing integration and esp. within EFL learners' experiences in their L1 long before they started their practices in their L2 should be taken into close consideration. It is also highly probable that English non-major scholars in this research might have also made use of other recourses for elevating their writing proficiency and had faced with other blocks in this process that had been due to other reasons leading to this suboptimum conditions in their written literacy practices for scholar publication aims. Implications related to the discerned major difficulties, which were discussed in this small-scale research, could be aimed for EFL contexts where indication of literacy abilities concerning writing abilities as writing-reading nexus might show a sub-optimum condition but with the interaction of some social factors, this sub-optimum condition might be alleviated to a certain extent. Still, a myriad of other factors at social as well as cultural factors apart from micro-macro levels might have been involved, which other researches might focus on such aspects to shed more light on this topic.

Limitations of the Study

In a nutshell, in this precursory research project, the participants sought possible reasons for the essay writing deficiencies mainly within linguistic strands among Iranian scholars In fact, proper research methods toward exploring difficulties as to written practices regarding RA text types could be several. One such route might be followed via analyzing writing performance of a large body of population having attended in nation-wide tests such as Tests of English as a Foreign Language (TOEFL) where examining specific Micro vs. Macro level aspects of writing are possible along an extended time span. Such studies, which are otherwise known as 'big data' research can help spotting the suboptimum conditions in this regard in the long run since there might exist a large database of the spotted Test of Written English (TWE) to be taken into account (Wassan, 2015). Nevertheless, exploring the complex nature of writing even through recent 'big data' studies within educational arenas can be challenging and may hardly give out accurate data over probable sources of problems regarding Micro vs. Macro levels of writing an RA due to different reasons like time limit strategy taken in such nation-wide test settings that might hinder some test takers. In other words, due to time limit policy, test takers might not be able to actualize their full potentiality or competencies on paper. Accordingly, data taken from such big data testing situations might be impeding rather than facilitating in gaining actual measures from the test candidates. Another alternative route toward investigating this issue could be directly scrutinizing the test takers' views themselves as major informants or data sources. Insights received as such might be much more revealing since they can best reflect their views over discerned difficulties during a longer processing rather than a single session as such. In addition, more focused data over research writing processes gained from big data studies may not be possibly spotted on aspects of writing esp., at Macro levels, which need to be observed in more depth. Exploring participants' attitudes as to their L2 writing processes has been current during the recent decades as well O' Keefe (2002). Thus, not relying on 'big data' strategies for collecting the required data, this research study followed latter strand that included focus group as informantsstrategy via two significant channels to reach more insightful information over writing RAs in English by the target group. Maybe further research can possibly maintain a more focused view so that direct assessment of measures could be gauged to reveal still other complex hidden reasons in this regard.

In addition, due to the limited number of scholars with both English and Persian publications in our academic settings and across only five years, maybe it could lead to more generalizable results if we made inquires over a more extended time and among more respondents. This could not necessarily distort the picture as to the targets of the present research, but this should be taken as having more credence over the authors' responsibility in communicating all the features of the context for doing this research.