Introduction

The dominance of English as a language both within and outside academia has significantly impacted the demand for reading in English as a second or foreign language. The importance of English reading skills has also led to considerable research on reading practices (Celce-Murcia, 2001; Duke et al., 2021; Koda & Zehler, 2008; Smith et al., 2021; Urquhart & Weir, 1998). The main concern of this line of research has been to help students acquire meaning from text. Research has shown that a lack of adequate reading proficiency and failure to comprehend various text types can hinder students’ progress in academic contexts (Koda & Zehler, 2008; Yapp et al., 2021).

Reading is considered an interactive process between the reader’s many cognitive processes and between the reader and the text (Cao & Kim, 2021; Grabe, 2009). Reading is a communicative act, in which the reader is not a passive receiver of information, and according to Fenner (2001, p. 24), acts as a “co-producer of language and meaning.” One way to facilitate communication between the reader and the writer is through rhetorical features called metadiscourse. Metadiscourse, defined as “embodying the interaction necessary for successful communication” (Hyland & Tse, 2004, p. 161), has attracted the attention of researchers for the last two decades (e.g., Ädel, 2023; Al-Subhi, 2022; Tang, 2021). For Hyland (2017), metadiscourse refers to the way writers select language out of consideration for readers, to help them process and comprehend what they are reading, through elaboration, clarification, guidance, and interaction. Understanding metadiscourse and how to use it when reading helps readers comprehend the author's textual strategy. Readers can discern the structure of a text—introduction, body, conclusion, or shift of topics, and realize the author’s attitudes toward the content and the reader (Crismore, 1982). Readers can also distinguish accepted claims from those disputed by the author (Vande Kopple, 1997). Thus, metadiscourse involves how writers and readers share their worlds and how such a world connects them discursively.

Metadiscourse has been the subject of many studies in different fields and genres. However, most of these studies explored the effect of metadiscourse on students’ writing skills rather than reading skills (e.g., Deng et al., 2021; Nesi, 2021; Yoon, 2021). Considering the importance of reading in educational settings, especially English as a second language (ESL)/English as a foreign language (EFL) settings, and the crucial role that reading plays in the academic achievement of learners (Koda & Zehler, 2008), the role that knowledge of metadiscourse elements can have on students’ reading comprehension is worthy of study.

In the EFL context of Iran, metadiscourse has been indicated to have facilitative effects on the students’ EFL reading comprehension (Jalilifar & Alipour, 2007; Parvaresh & Nemati, 2008; Soltani & Shokrpour, 2021; Tavakoli et al., 2010; Vasheghani, 2020; Zakeri & Samimi, 2021); however, few experimental studies have been conducted to examine the effect of teaching metadiscourse on students’ reading comprehension of English texts. Previous studies showed intriguing results regarding the impact of metadiscourse awareness on EFL learners of varying levels of language proficiency (Hashemi et al., 2011; Jalilifar, 2011; Parvaresh & Nemati 2008).

To clarify the role metadiscourse plays in EFL readers’ reading comprehension, this study aimed to investigate the effect of teaching metadiscourse on the reading comprehension of EFL students with varying levels of language proficiency as defined by the Oxford Placement Test (OPT), a standardized test of the English language for non-native speakers of English. Based on these objectives, the present study attempted to address the following research question:Does instruction about metadiscourse enhance the reading comprehension of EFL learners with different levels of language proficiency?

Literature Review

Strategy-based reading instruction

Over the last few decades, research in second language reading comprehension has focused on identifying strategies directly associated with comprehension performance. It is widely accepted that a critical component of comprehension is the ability to use appropriate reading strategies (Fathi & Afzali, 2020; Grabe, 2002; Koda, 2005; McBreen & Savage, 2021; Muhid et al., 2020; Wissinger et al., 2021). Another category of related studies has compared the patterns of strategies used by readers across different proficiency levels. Findings have indicated that, compared to poor readers, good readers use a more extensive repertoire of strategies, use strategies more effectively, and are actively engaged in reading comprehension (Grabe, 2009). Recent trends in this domain have shifted the emphasis from merely identifying and classifying strategies to strategy instruction and its benefits for learners’ reading comprehension performance. It has been found that explicit teaching of effective text-processing strategy helps students become strategic readers (Dignath & Veenman, 2021; Grabe, 2009; Janzen, 2002; Kavani & Amjadiparvar, 2018; Zhang, 2008). In strategy-based reading instruction, it is recommended that the teacher incorporate instruction of reading strategies through direct explanation, consistent modeling, scaffolding, extensive practice, and feedback (Grabe, 2009). It is this incorporation that helps learners effectively make sense of input in general and texts in particular.

Iranian university EFL learners have problems with reading comprehension partly due to the language teaching system in Iran, since the primary English teaching method in school and university is still the Grammar Translation method (Hadad Narafshan & Yamini, 2011). In schools, teachers spend most of the class time translating the texts into Persian, explaining grammatical structures, and providing Persian equivalents for new words during each lesson. Lack of explicit instruction in effective reading strategies is one of the main factors contributing to reading comprehension problems (Milaghardan, 2016). The goal of English instruction and learning is mainly for students to acquire linguistic knowledge of English, as it is all they need to know to pass their exams and be well-prepared for the university entrance exam. Beside emphasizing linguistic forms, they are also taught test-taking strategies that help them in the national university entrance exam. Farhady et al. (2010) considered this situation narrow instruction, which naturally leads to many problems for students who enter university and need to read texts in English.

Because English is a foreign language in Iran with no or very little use in the everyday lives of EFL learners, English texts for university students majoring in English are the primary source of language input. Thus, reading comprehension ability is of great importance for them to progress in their academic performance. However, most students have not acquired this ability in school, resulting in poor reading ability at university. The over-reliance on lower-level text processing strategies, such as paying attention to single words, translation, and looking up new words, which are regarded as insufficient reading strategies, takes EFL reader’s attention away from higher-order strategies (Fotovatian & Shokrpour, 2007). This is because lower-level text processing skills, though essential components of reading, do not guarantee reading comprehension. Reading is not a single-factor process; instead, it is a complex process requiring both low-level and high-level text processes to work together interactively as part of the reading process (Nassaji, 2003). The complex nature of reading processes at the tertiary level requires EFL/ESL readers to go beyond the word and sentence levels and expand their repertoire of reading strategies.

Recognizing and understanding the authors’ implicit message is one of the difficulties Iranian EFL learners at the tertiary level encounter while reading their academic English texts (Ahmadi et al., 2013). They regard reading as extracting facts from a written text rather than a communicative act in which they should participate in the co-construction of meaning (Wilson et al., 2004). One way to help students at the tertiary level cope with academic reading materials is to teach them metadiscourse strategies. Attention to these reading strategies is primarily neglected in reading classes at Iranian schools and at tertiary levels. An awareness of these features as facilitative means of communication can help EFL readers to engage in a dialogue with the writer, understand the authors’ text plan, and grasp the writer's stance when reading academic material.

Modified version of Hyland’s Taxonomy of Metadiscourse

The present study adapted Hyland’s (2005) taxonomy of interpersonal metadiscourse to teach metadiscoursal features of English tests to students in the experimental group. In Hyland’s model, metadiscourse comprises two dimensions of interaction: the interactive dimension and the interactional dimension. These two dimensions are believed to be defining characteristics of any spoken or written communication and are expressed through various rhetorical features. Interactive resources help organize propositional information in ways that readers find coherent and persuasive. These include transition markers that connect ideas, frame markers that structure and signal stages of the text, endophoricmarkers that refer to other parts of the text, evidentials that attribute information to sources, and code glosses that clarify or elaborate meanings for readers. The interactional dimension reflects the writer’s stance toward both the content and the reader. It includes hedges, which limit commitment and allow alternative views; boosters, which express certainty; attitude markers, which convey the writer’s evaluation or affect; self-mentions, which signal authorial presence; and engagement markers, which directly involve readers and guide them toward particular interpretations.

To suit the purpose of the study, a few changes were applied to the model: the category of evidentials and endophoric markers were omitted from the model, and the punctuation category was incorporated into it instead. The reason to omit the evidential and endophoric markers from the model is that these categories are most applicable to long passages such as academic articles, and the reading passages used to teach metadiscourse features in this study are of one- or two-page length and general genre, not discipline-based. The reason to include punctuation marks is that it has a vital role in reading comprehension. As Crismore et al. (1993) stated, “these marks can signal text glosses and clarification as well as uncertainty, certainty, and attitude” (p. 48). The classification of the modified model of metadiscourse is presented in Table 1 below.

Table 1: Hyland’s (2005) Adapted Model of Metadiscourse for Use in Reading Instruction

Methodology

The present study is quantitative research employing a survey and experimental design. In many educational research studies, when it is impossible to select subjects randomly, it is recommended that the researcher employs a quasi-experimental design and uses intact classrooms as the experimental and control groups (Ary et al., 2009). Since administrative constraints imposed by the university where this experimental research was conducted made it impossible to select subjects randomly, a quasi-experimental design was employed. However, Seliger and Shohamy (1989) recommended that the researcher match subjects in groups in a quasi-experimental design to make groups more comparable. To increase the subjects’ comparability in this study, the researcher administered the Oxford Placement Test (OPT).

The study was conducted with two groups of subjects: the experimental group (EG) and the control group (CG). The EG received instruction on metadiscoursal features of expository texts following Hyland’s (2005) modified model of interpersonal metadiscourse. Students’ scores on a reading comprehension test (developed to assess students’ metadiscourse knowledge) were the dependent variable, and instruction of metadiscoursal features were the independent variable. A widely used quasi-experimental design in educational research, namely a nonrandomized control group, pretest-posttest design was used in this study (Ary et al., 2009), as presented in Table 2 below. To meet ethical standards, we asked the participants to sign a consent form, which involved expressing their agreement to participate in the study and that the data would only be used for research purposes.

Table 2: Non-randomized control group, pretest-posttest quasi-experimental design

Participants

The participants of this study were 113 undergraduate students of Teaching English as a Foreign Language (TEFL) at the Islamic Azad University in Tehran, Iran. The students were informed that information that could be used to identify them (students’ details, as well as their responses) would not be shared with third-party agencies or researchers not affiliated with the current research; further, they were assured that all the responses would be kept anonymous in the research report. They were chosen through purposive sampling to participate in the research. All related data was collected in their Advanced Reading classes, a compulsory course. According to the academic administration timetable, these 113 students were split in four different classes. For the purpose of this study, the four classes were randomly divided into two groups 57 students in the experimental group and 56 students in the control group.

Instruments

In order to collect the data, two instruments were employed: OPT and a Reading Comprehension Test (RCT). The Oxford Placement Test was used as a test of English language proficiency. The students’ scores on it were standardized by mapping them onto the Common European Framework of Reference (CEFR), and four levels of language proficiency were identified (Table 3).

Table 3: The Oxford Placement Test scores mapped onto the Common European Framework of Reference.

In order to assess students’ knowledge of metadiscourse, an RCT was developed. This instrument was used as a pre- and post-test. It consists of four expository texts on general topics. A large sample of expository texts from different reading textbooks currently in use in universities across the country of Iran along with some from the internet were reviewed to select the appropriate texts. These were meticulously reviewed to ensure that they include all interactional and interactive metadiscourse resources defined by the modified model. The texts were checked by two experts who hold a PhD in Applied Linguistics to ensure the accuracy and appropriacy of the texts for the present study. Other criteria in selecting texts included: text types, number of paragraphs, readability index, and text length. Based on the four selected passages, reading comprehension questions were developed. Each text was followed by five multiple-choice questions. The RCT and the outline of Hyland’s (2005) modified model of metadiscourse were given to three instructors to examine the content and face validity of RCT. The lecturers who were chosen for their expert judgments met the following criteria:

• EFL university lecturers
• Had taught reading courses for more than six years
• Knowledgeable in metadiscourse

The RCT was deemed appropriate for the intended purpose following some modifications suggested by the expert judges. A reliability procedure called internal consistency was used to determine the reliability of RCT. Internal consistency indicated that all the items in the test were measuring the same thing. An adequate level of reliability (.71) for RCT was shown by the computed internal consistency reliability coefficient.

Two textbooks, Reader’s Choice and Active Reading, were selected by the university to be used in the reading classes. Reading instructors were required to utilize these two textbooks as the main course material. Besides these two textbooks, which were the standard materials for both groups, experimental and control, a further 18 passages were selected to be used in classes for the experimental group. These passages on general topics were chosen by the researcher from other available reading textbooks or the internet. The passages were selected meticulously to include all the metadiscourse resources defined by the modified model of metadiscourse.

Data collection procedures

Data collection procedures began at the start of the semester for students in Advanced Reading classes. In the first session of the semester, the OPT was administered to both groups to determine the students’ level of language proficiency. In the second session, the RCT was administered to both groups as a pretest.

After administering the OPT and Pre-RCT in the first two sessions of the semester, the teaching procedures began in the second week of the semester. The same teaching materials were used for the control and experimental groups during the same instructional period (16 instructional sessions), except that metadiscourse instruction was incorporated into the reading instruction for the experimental group.

In the first session of the second week, the instructor introduced the concept of metadiscourse and its function as a comprehension strategy that can help students improve their reading skills. The purpose of this general information on metadiscourse was to help students acquire an overall picture of the concept. The instructor explained the general function of interactive and interactional metadiscourse. Students were also made aware of the major function of interactive metadiscourse as devices that make a text easy to follow by conforming to conventional text patterns and predictable directions, enabling the reader to process the text by recognizing relationships and ordering material. The primary function of interactional metadiscourse as devices that focus more on the interaction between the writer and the reader were also explained.

Metadiscourse instruction

Teaching metadiscourse markers, or the actual process of teaching reading, began in the second part of the second week of the semester. Every week, students were introduced to one kind of metadiscourse based on the modified model of metadiscourse. For each kind of metadiscourse, two texts were chosen and named accordingly for the instructor to keep in a file.

Through discourse analysis of different passages during this study, the students were taught different metadiscourse forms and their functions. Metadiscourse was first introduced and explained, then practiced giving students metacognitive awareness of metadiscourse functions as discourse organizing and an indication of the writer’s attitudes towards the reader and the text. The instruction of metadiscourse strategies in every session was based on the following guidelines in reading strategy instruction proposed in earlier research (Grabe, 2009; Janzen, 2002):

• Explicit teaching of the metadiscourse element in its context
• Explanation and definition of its function through mental modeling and asking questions,
• Practice through discourse analysis: asking students to identify the element by doing different exercises
• Constant recycling of the strategies over new texts

The introduction and teaching all metadiscourse categories were completed by the end of the tenth session. During these ten weeks, all metadiscourse markers were introduced, explained, practiced, and reviewed. From the eleventh session on, the class worked with just the course books while working on instances of metadiscourse markers they encountered. In this way, in the five remaining sessions, all metadiscourse markers were reviewed again. When teaching from the textbooks, the instructor called students’ attention to metadiscourse markers they encountered and pointed out their functions in different contexts. The instructor also asked students questions and responded to their answers with feedback.

Data Analysis and Results

In order to compare students’ performance in the experimental and control groups on pre-RCT, a two-way between-group analysis of variance (ANOVA) was conducted. The analysis revealed that both groups were different in their performance on pre-RTC. The descriptive statistics, Table 4, indicated that the students’ pre-RCT mean scores in the control group for all proficiency levels (lower advanced (M=15.91, SD=0.83), upper-intermediate: 14.70, lower-intermediate: 12.26, elementary: 9.92) were higher than those of the experimental group (lower-advance: 14.17, upper-intermediate: 11.70, lower-intermediate: 8.35, elementary: 8.61). The result showed that the groups’ performances on the pre-test were not similar. When a pretest indicates that the groups are not equivalent on the dependent variable, the literature suggests using ANCOVA to statistically adjust the post-test scores for the pre-test differences (Ary et al., 2009).

Table 4: Descriptive statistics for pre-reading comprehension test

Consequently, ANCOVA was conducted to compare groups’ mean scores on post-RCT analysis of covariance. A 2-by-2 between-group analysis of covariance was conducted to assess the effectiveness of the intervention (metadiscourse strategy instruction) on students’ reading comprehension performance. The independent variables were the intervention and students’ levels of language proficiency. The dependent variable was students’ post-RCT scores, administered after the intervention's completion. By using ANCOVA and setting the pre-RCT scores as the covariate and the post-RCT scores as the response variable of interest, the study statistically controlled for the initial difference between the groups.

The between-subjects test effects indicated a significant main effect for independent variables (groups and levels within them) and whether the interaction between these two variables was significant. The results in Table 5 show that the interaction effect of the two independent variables is not significant, GROUP*OPT LEVEL: F (3, 10)=2.49, p=.06>.05. This indicates that there is no significant difference in the effect of levels of language proficiency on groups’ differences on post-RCT. This finding means that the treatment effect was the same across different levels. According to Pallant (2013), when the interaction effect is not significant, the researcher can safely interpret the main effect. As Table 5 shows, there is a significant main effect for the independent variable Group: F(1, 10)=38.85, p=.000<.05. That is, both groups significantly performed differently in the post-RCT.

Table 5: Tests of between-subjects’ effects on Post-Reading Comprehension Test

Table 6 shows the descriptive statistics of post-RCT scores for the experimental and the control group. It shows that students’ mean scores in all four levels of language proficiency in the experimental group (C1, M=16.67, SD=1.03), (B2, M=15.20, SD=2.48), (B1, M=13.57, SD=1.37), (A2, M=13.06, SD=2.20) are higher than those of the control group (C1, M=15.73, SD=1.00), (B2, M=15.30, SD=2.49), (B1, M=12.00, SD =2.89) (A2, M=10.08, SD=1.73). Although the upper intermediate group in the experimental group showed a lower mean (15.20) score than the control group (15.30), incorporating the initial difference that existed between the two groups by the ANCOVA revealed that the upper intermediate students in the treatment group performed significantly better than the students in the control group at the same level of language proficiency. The results are in line with studies that investigated the influence of metadiscourse instruction on language learners’ reading comprehension performance with intermediate (Tavakoli et al. 2010; Vasheghani, 2020), and pre-intermediate (Jalilifar & Alipour, 2007) levels of language proficiency.

Table 6: Descriptive statistics of Post-Reading Comprehension Test scores for both groups

Discussion

The findings of this study indicate that the metadiscourse instruction had a significant effect on tertiary EFL learners’ reading performance. It was found that despite the control group’s initially higher proficiency, the experimental group outperformed the control group in the post-Reading Comprehension Test. This finding indicates that metadiscourse instruction could help students in the experimental group improve their reading comprehension. The results of the present study lend further support to previous research demonstrating the facilitative role of metadiscourse in EFL learners’ reading comprehension where learners at all proficiency levels performed better with metadiscourse-enriched texts (Zarrati, et al., 2014). These results are also consistent with earlier studies by Jalilifar and Alipour (2007), Tavakoli et al. (2010), and Vasheghani (2020) which similarly reported the positive impact of metadiscourse on reading comprehension. As Nesi (2021) argues that because metadiscourse provides a multidimensional representation of text construction/comprehension, it helps readers develop their understanding through various strategies, including elaboration and clarification. It seems that the participants of this study have also developed such competencies, enabling them to comprehend the texts from multidimensional aspects.

Furthermore, by comparing students’ mean scores across the four levels, the study found no statistically significant difference in the effect of language proficiency level on groups’ differences in post-RCT. The positive effect of metadiscourse instruction was similar across the four levels. This finding suggests that metadiscourse instruction was a practical approach to helping students progress in their reading skills regardless of their levels of language proficiency.

Previous studies provided different results on the effect of metadiscourse markers in reading English texts, and instruction of these markers on reading comprehension for other proficiency groups. Some studies showed that students with a high level of language proficiency benefited more from the instruction of metadiscourse (Jalilifar, 2011). Other studies showed that low-proficiency language learners benefited more from metadiscourse markers (Parvaresh & Nemati, 2008). Still, other studies showed that while these markers helped the advanced students perform significantly better on a reading comprehension test, they had no significant effect on the reading performance of the intermediate group unless they were consciously aware of them (Hashemi et al., 2011). These inconsistent findings reveal that proficiency level may be a key variable in how metadiscourse affects students’ reading comprehension.

Implications

While metadiscourse may be an essential part of instruction in English reading classes, Iranian EFL learners are overly reliant on lower-level text processing strategies at the word and sentence level, so they have difficulties deciphering the global meaning and intention of English texts. Thus, to help EFL learners improve their reading comprehension, reading instructors are encouraged to integrate teaching metadiscourse in their classes. One of the salient implications of the present study is its classroom application. As the results of the present study indicate that university students of varying levels of English proficiency benefited from the intervention. Drawing on the findings, the study provides a few general guidelines for effectively implementing metadiscourse strategy instruction in EFL reading classes.

The first issue in teaching metadiscourse in reading classes is selecting a model that best suits its purpose. This study adapted Hyland’s (2005) model of metadiscourse to teach metadiscourse elements. The researcher made a few changes to the model to make the model more applicable to reading classes. The practical application of the model showed some positive outcomes in this study. Thus, the adjusted model is recommended to be used by reading teachers when teaching metadiscourse strategies in their classes.

The second issue concerning metadiscourse instruction is the period and procedures of teaching metadiscourse strategies. Metadiscourse must be integrated into the overall class time. As a ubiquitous aspect of everyday communication (Hyland, 1998), metadiscourse is not separated from general language use; hence it should not be taught separately. Having regular class time for teaching reading and allocating a portion of class time to teaching metadiscourse may suggest that metadiscourse is not as crucial as discourse. Students’ attention should be directed to these elements during class time when reading passages, doing exercises, explaining expression and vocabulary, and whenever else the teacher thinks it is necessary.

The third issue is following a systematic lesson plan in metadiscourse instruction. It is recommended that teachers organize their lesson plans to allocate enough time to introduce each category of metadiscourse, explain its function, practice it in different contexts, and provide students with various examples of each category. The lesson plan organized to incorporate metadiscourse instruction in this study could also be used in similar reading courses.

The fourth issue concerning metadiscourse instruction is the student's level of language proficiency. To teach metadiscourse strategies to EFL learners, it is reasonable to think that they need to have a minimum level of language proficiency. At the tertiary level, however, as this study indicated, students across the four levels of language proficiency (lower advanced, upper intermediate, lower intermediate, elementary) benefited equally from metadiscourse instruction.

Besides classroom applications, the findings of this study also have implications for materials developers who oversee the design and selection of materials to be used in reading classes. They can include sections for introducing, defining, and practicing metadiscoursal features. However, metadiscourse elements must be presented in ways that highlight their primary function in influencing the interaction between writer and reader (Intaraprawat & Steffensen, 1995).

Conclusion

The results of this study add to this key variable by showing that proficiency level may not feature significantly in metadiscourse. Yet, this study was limited in terms of the number of students. We believe that if proficiency level were defined distinctly (i.e., elementary versus advanced), how learners of different levels differed could be explored more effectively . This issue could be explored in future research.

It is hoped that the present study's findings cast new light on the relationship between metadiscourse knowledge and EFL/ESL learners’ reading comprehension, and have filled an existing gap in the metadiscourse literature. As Hyland (2005) states, metadiscourse is a new and growing field of investigation with considerable potential for description and explanation, promising to reveal the underlying interaction of all communication and help us see how discourse is produced and received.

References

Ädel, A. (2023). Adopting a ‘move’ rather than a ‘marker’ approach to metadiscourse: A taxonomy for spoken student presentations. English for Specific Purposes, 69, 4-18. https://doi.org/10.1016/j.esp.2022.09.001  

Al-Subhi, A. S. (2022). Metadiscourse in online advertising: Exploring linguistic and visual metadiscourse in social media advertisements. Journal of Pragmatics, 187, 24-40. https://doi.org/10.1016/j.pragma.2021.10.027  

Ary, D., Jacobs, L. C., Razavieh, A., & Sorensen, C. K. (2009). Introduction to research in education. (8^th ed.), Cengage.

Cao, Y., & Kim, Y.-S. G. (2021). Is retell a valid measure of reading comprehension? Educational Research Review, 32.https://doi.org/10.1016/j.edurev.2020.100375

Celce-Murcia, M. (Ed.). (2001). Teaching English as a second or foreign language. (3^rd ed.). Heinle & Heinle.

Crismore, A. (1982). The metadiscourse component: Understanding writing about reading directives files (ED217374). ERIC.https://files.eric.ed.gov/fulltext/ED217374.pdf

Crismore, A., Markkanen, R., & Steffensen, M. S. (1993). Metadiscourse in persuasive writing: A study of texts written by American and Finnish university students. Written Communication 10(1), 39-71. https://doi.org/10.1177/0741088393010001002

Deng, L., Fatemeh, B., & Gao, X. (2021). Exploring the interactive and interactional metadiscourse in doctoral dissertation writing: A diachronic study. Scientometrics, 126, 7223-7250. https://doi.org/10.1007/s11192-021-04064-0  

Dignath, C., & Veenman, M. V. J. (2021). The role of direct strategy instruction and indirect activation of self-regulated learning: Evidence from classroom observation studies. Educational Psychology Review, 33(2), 489-533. https://doi.org/10.1007/s10648-020-09534-0

Duke, N. K., Ward, A. E., & Pearson, P. D. (2021). The science of reading comprehension instruction. The Reading Teacher,74(6), 663-672. https://doi.org/10.1002/trtr.1993  

Farhady, H., Sajadi Hezaveh, F., & Hedayati, H. (2010). Reflections on foreign language education in Iran. TESL-EJ, 13(4). https://tesl-ej.org/pdf/ej52/a1.pdf

Fathi, J., & Afzali, M. (2020). The effect of second language reading strategy instruction on young Iranian EFL learners’ reading comprehension. International Journal of Instruction, 13(1), 475-488. https://doi.org/10.29333/iji.2020.13131a

Fenner, A.-B. (2001). Dialogic interaction with literary texts in the lower secondary classroom. In A.-B. Fenner (Ed.), Cultural awareness and language awareness based on dialogic interaction with texts in foreign language learning (pp. 13-46). Council of Europe.

Fotovatian, S., & Shokrpour, N. (2007). Comparison of the efficiency of reading comprehension strategies on Iranian university students’ comprehension. Journal of College Reading and Learning, 37(2), 47-63. https://doi.org/10.1080/10790195.2007.10850197

Grabe, W. (2002). Dilemmas for the development of second language reading activity. In J. C. Richards & W. A. Renandya (Eds.), Methodology in language teaching: An anthology of current practice (pp. 276-286). Cambridge University Press.

Grabe, W. (2009). Reading in a second language: Moving from theory to practice. Cambridge University Press.

Hashemi, M. R., Khodabakhshzadeh, H., & Shirvan, M. E. (2011). The effect of metadiscourse on EFL learners’ reading comprehension. Modern Journal of Language Teaching Methods, 1(2), 112-126. https://profdoc.um.ac.ir/articles/a/1025702.pdf

Hyland, K. (2005). Metadiscourse. Continuum.

Hyland, K. (2017). Metadiscourse: What is it and where is it going? Journal of Pragmatics, 113, 16-29. https://doi.org/10.1016/j.pragma.2017.03.007

Hyland, K., & Tse, P. (2004). Metadiscourse in academic writing: A reappraisal. Applied Linguistics, 25(2), 156-177. https://doi.org/10.1093/applin/25.2.156  

Intaraprawat, P., & Steffensen, M. S. (1995). The use of metadiscourse in good and poor ESL essays. Journal of Second Language Writing, 4(3), 253-272. https://doi.org/10.1016/1060-3743(95)90012-8

Jalilifar, A. R., & Shooshtari, Z. G. (2011). Metadiscourse Awareness and ESAP comprehension. Journal of College Reading and Learning, 41(2), 53-74. https://doi.org/10.1080/10790195.2011.10850342

Jalilifar, A., & Alipour, M. (2007). How explicit instruction makes a difference: Metadiscourse markers and EFL learners’ reading comprehension skill. Journal of College Reading and Learning, 38(1), 35-52. https://doi.org/10.1080/10790195.2007.10850203

Janzen, J. (2002). Teaching strategic reading. In J. C. Richards & W. A: Renandya (Eds.), Methodology in language teaching: An anthology of current practice (pp. 287-294). Cambridge University Press.

Kavani, R., & Amjadiparvar, A. (2018). The effect of strategy-based instruction on motivation, self-regulated learning, and reading comprehension ability of Iranian EFL learning, Cogent Education, 5(1). https://doi.org/10.1080/2331186X.2018.1556196  

Koda, K. (2005). Insights into second language reading: A cross-linguistic approach. Cambridge University Press.

Koda, K., & Zehler, A. M. (Eds.). (2008). Learning to read across languages: Cross linguistic relationship in first- and second- language literacy development. Routledge.

McBreen, M., & Savage, R. (2021). The impact of motivational reading instruction on the reading achievement and motivation of students: A systematic review and meta-analysis. Educational Psychology Review, 33, 1125-1163. https://doi.org/10.1007/s10648-020-09584-4

Hosseinpour Milaghardan, R. (2016). A survey of the causes of reading difficulty of the Iranian EFL learners. Indian Journal of Applied Research, 6(10) 241-244.

Jafarinejad, R., & Tavakoli, M. (2011). Investigating the relationship between discourse markers, language proficiency and reading comprehension: A case of some Iranian university students. Procedia-Social and Behavioral Sciences, 15, 1526-1530. https://doi.org/10.1016/j.sbspro.2011.03.324

Jalilifar, A. R., & Shooshtari, Z. G. (2011). Metadiscourse awareness and ESAP comprehension. Journal of College Reading and Learning, 41(2), 53-74. https://doi.org/10.1080/10790195.2011.10850342

Muhid, A., Amalia, E. R., Hilaliyah, H., Budiana, N., & Wajdi, M. B. N. (2020). The effect of metacognitive strategies implementation on students’ reading comprehension achievement. International Journal of Instruction, 13(2), 847-862.https://doi.org/10.29333/iji.2020.13257a

Narafshan, M. H., & Yamini, M. (2011). Policy and English language teaching (ELT) in Iran. Iranian EFL Journal, 7(5), 179-189.

Nassaji, H. (2003). Higher-level and lower–level text processing skills in advanced ESL reading comprehension. The Modern Language Journal, 87(2), 261-276. https://doi.org/10.1111/1540-4781.00189

Nesi, H. (2021). Sources for courses: Metadiscourse and the role of citation in student writing. Lingua, 253. https://doi.org/10.1016/j.lingua.2021.103040  

Pallant, J. (2013). SPSS survival manual: A step by step guide to data analysis using IBM SPSS. (5^th ed.). Routledge.

Parvaresh, V., & Nemati, M. (2008). Metadiscourse and reading comprehension: The effects of language and proficiency.Electronic Journal of Foreign Language Teaching, 5(2), 220-239. https://e-flt.nus.edu.sg/wp-content/uploads/docroot/v5n22008/parvaresh.pdf

Reza Ahmadi, M., Ismail, H. N., & Kabilan Abdullah, M. K. (2013). The importance of metacognitive reading strategy awareness in reading comprehension. English Language Teaching, 6(10), 235-244. https://doi.org/10.5539/elt.v6n10p235 

Seliger, H. W., & Shohamy, E. (1989). Second language research methods. Oxford University Press.

Smith, R., Snow, P., Serry, T., & Hammond, L. (2021). The role of background knowledge in reading comprehension: A critical review. Reading Psychology, 42(3), 214-240. https://doi.org/10.1080/02702711.2021.1888348  

Soltani, R., & Shokrpour, N. (2021). Metadiscourse markers in Iranian medical articles: A comparative study. Journal of Mazandaran University of Medical Sciences, 31(196), 92-100.

Tavakoli, M., Dabaghi, A., & Khorvash, Z. (2010). The effect of metadiscourse on L2 reading comprehension: A case of Iranian EFL learners. English Language Teaching, 3(1), 92-102. https://doi.org/10.5539/elt.v3n1p92  

Tang, K.-S. (2021). The interconnections among metadiscourse, metalanguage, and metacognition: Manifestation and application in classroom discourse. Linguistics and Education, 65. https://doi.org/10.1016/j.linged.2021.100977  

Urquhart, S., & Weir, C. (1998). Reading in a second language: Process, product and practice. Routledge.

Vande Kopple, W. J. (1997). Refining and applying views about metadiscourse. Paper presented at the 48'h Annual Meeting of the Conference of College Composition and Communication, Phoenix, AZ. https://files.eric.ed.gov/fulltext/ED411539.pdf  

Vasheghani Farahani, M. (2020). Explicit instruction of metadiscourse features and intermediate EFL learners’ reading comprehension ability. Linguistics Journal, 14(2), 87-107.

Wilson, K., Devereux, L., Macken-Horarik, M., & Trimingham-Jack, C. (2004). Reading readings: How students learn to (dis)engage with critical reading. In F. Sheehy & B. Stauble (Eds.), Transforming Knowledge into Wisdom: Holistic Approaches to Teaching and Learning, Proceedings of the Annual International HERDSA Conference, Miri, Sarawak, 4-7 July, 2004 (pp. 341-348). http://www.herdsa.org.au/system/files/wlson.pdf

Wissinger, D. R., De La Paz, S., & Jackson, C. (2021). The effects of historical reading and writing strategy instruction with fourth- through sixth-grade students. Journal of Educational Psychology, 113(1), 49–67. https://doi.org/10.1037/edu0000463

Yapp, D., de Graaff, R., & van den Bergh, H. (2023). Effects of reading strategy instruction in English as a second language on students’ academic reading comprehension. Language Teaching Research. 27(6), 1456-1479. https://doi.org/10.1177/1362168820985236  

Yoon, H.-J. (2021). Interactions in EFL argumentative writing: Effects of topic, L1 background, and L2 proficiency on interactional metadiscourse. Reading and Writing, 34(3), 705-725. https://doi.org/10.1007/s11145-020-10085-7  

Zakeri, D., & Samimi, F. (2021). A cross-linguistic study on the reflection of metadiscourse markers in the acknowledgements of the Iranian and native English dissertations. Language Teaching Research Quarterly, 26, 52-64. https://doi.org/10.32038/ltrq.2021.26.04

Zarrati, Z., Nambiar, R. M. K., & Nor Rizan Mohamad Maasum, T. (2014). Effect of metadiscourse on reading comprehension of Iranian EFL learners. 3L: Language, Linguistics and Literature: The Southeast Asian Journal of English Language Studies, 20(3), 27-38. https://journalarticle.ukm.my/7736/1/5324-18843-1-PB.pdf

Zhang, L. J. (2008). Constructivist pedagogy in strategic reading instruction: Exploring pathways to learner development in the English as a second language (ESL) classroom. Instructional Science, 36(2), 89-116. https://doi.org/10.1007/s11251-007-9025-6