Teachers' beliefs and practices in teaching reading at Islamic secondary schools in Indonesia

The importance of English reading skills in developing overall English literacy has made reading instruction one of the primary focuses in Indonesia's English teaching. Regardless of the high value of reading skills, the teaching of reading has not been successful in developing the reading comprehension ability of Indonesian students. This exploratory case study research examines teachers' beliefs and practices in teaching reading in Islamic secondary schools. It focuses mainly on how teachers view good reading instruction and the problems they encounter in teaching reading. Additionally, this paper explores how teachers implement reading instruction in their contexts. Data were obtained from 31 senior Islamic secondary school English teachers from six different provinces in Indonesia. Data collection methods comprised document analysis, participant classroom observation, personal in-depth interviews, written guided reflections, and focus group discussions. The data were analyzed using the constant comparative method. The research findings show that despite admitting the importance of teaching reading in schools, the English teachers of Islamic senior high schools found teaching reading challenging. The teachers identified several problems in reading instruction in their classrooms, such as lack of strategies to teach vocabulary, limited repertoire of strategies to teach reading, the use of teaching materials, and improving students' reading motivation and autonomy. As a result, they have not been able to implement classroom practices that effectively develop students' reading comprehension. This issue calls for the innovation and development of teaching reading strategies to improve reading instruction quality in Islamic secondary schools in Indonesia.


INTRODUCTION
Reading comprehension instruction has become a prominent interest of research in English Language Teaching (ELT) in the context of teaching English as a Foreign Language (EFL) these past few decades. This trend has emerged due to the position of reading as one of the important skills to acquire to develop the learner's overall English literacy skills. Reading skills are considered essential to enable students to prevail in their academic and nonacademic life (Cahyono & Widiati, 2006;Pang, 2008). In addition, the possession of sufficient skills to comprehend written texts also facilitates students to improve the language acquisition process, obtain good models for writing, and stimulate discussions to support their learning (Richards & Renandya, 2002). For these reasons, English instruction in the school system in Indonesia aims to provide well-developed reading skills among Indonesians to read texts written in English (Renandya, 2004). The prime role of reading in language acquisition has elevated the position of reading instruction and research on reading instruction to an essential level in ELT in Indonesia.
However, regardless of the ever-increasing attention towards EFL reading instruction in Indonesia, Indonesian students have not shown satisfactory achievement in reading skills. The results of the Program for International Students Assessment (PISA) test in 2015 indicated that Indonesian students ranked 69th out of 76 countries in the reading domain. Three years later, in 2018, the result of a very similar test did not show significant improvement as Indonesia ranked 72th out of 77 countries in the literacy domain (OECD, 2019). On a national level, the 2016 national examination results for English at senior high school and Islamic secondary school levels were also alarming. The average score of the 2016 national examination for English at these two schools was 50.73/100 and 48.96/100, respectively. Students' reading achievement was manifested in their examination score, as 70% of the test consisted of reading comprehension questions, while the other 30% consisted of listening comprehension questions. These problems of low reading achievement call for a rigorous investigation if we are to improve the success of reading instruction at the senior high school level.
Research on EFL reading instruction in senior high schools in Indonesia has focused on various aspects that potentially affect the instruction's effectiveness, including the students, teachers, teaching materials, and reading pedagogy. Yusuf et al. (2017) examine the use of self-assessment to diagnose students' reading problems in a senior high school in Banda Aceh, Indonesia. Their study reveals several problems encountered by good readers and poor readers in EFL reading classes that the teacher needs to anticipate to provide better teaching. Among the problems are students' inability to identity unfamiliar words, to determine the language features of English texts, and to predict main ideas in several paragraphs. Hamra and Syatriana (2010) propose the development of an interactive model for teaching reading comprehension in five senior high schools in South Sulawesi, Indonesia. They found that the proposed model of Interactive Model of Teaching Reading Comprehension (IMTRC) was effective to improve students' reading comprehension through emphasizing five key aspects: previous knowledge, reading skills, reading knowledge, reading participation, as well as reading environments.
In addition, Rusmawaty et al. (2018) conducted a case study on teachers' belief in using authentic materials for teaching EFL reading in a senior high school in Samarinda, Indonesia. The teachers are reported to show a positive attitude towards the use of authentic materials in reading classes. Implications for teacher training are then drawn based on the findings. Tabiati (2016), who conducted a study on factors affecting learners' autonomy in reading at an English department in an Indonesian university, suggested that there were at least 28 factors, both internal and external, which affected the development of autonomous readers. Internal factors could be attributed to, for instance, students' knowledge on the importance of EFL reading and motivation to read English texts. Additionally, she reported several external factors that affected students' autonomy in reading, such as the reading environment and the support of their significant others, i.e., parents, teachers, and family members.
The plethora of research on EFL reading classrooms mentioned above can help ELT stakeholders navigate current reading classroom issues. However, among this existing research, there is a lack of explanation about how teachers perceive the problems of EFL reading instruction in the classroom and how their views are manifested in classroom practices. Teachers' beliefs and perceptions are important as they can show what knowledge is worthy to be presented in action (Kumaravadivelu, 2012). Therefore, they are influential in shaping teachers' language instruction in the classroom. The compatibility between teachers' beliefs and the way they practice their beliefs in the classroom provides insight into what the teachers have understood and how they put their knowledge into action in the teaching and learning process (Farrell, 2013). Therefore, this study attempts to examine the beliefs and practices of English teachers in teaching EFL reading in Indonesia, which is currently underexplored in existing literature.
Furthermore, this study took place in Islamic secondary schools in 16 cities in Indonesia. Research on ELT in Islamic secondary schools remains scarce, although this setting's particularity is of interest to consider. Since the English language in such schools is often challenged by the abundant load of religious teaching in the curriculum, it is intriguing to see how teachers practice the teaching of EFL reading in this context.
To fill the above mentioned gap in existing literature, this research aims to identify the perceptions of English teachers about EFL reading instruction and see how these views are enacted in the classroom practices in an Islamic secondary school setting. This exploratory research is the initial phase of a bigger significant research project with a design-based approach to develop a sociocognitive model for teaching reading in Islamic secondary schools in Indonesia. Specifically, this research attempts to answer the following research questions: (1) How do teachers view the problems and needs of teaching EFL reading in English classrooms in Islamic secondary schools in Indonesia? (2) How do teachers enact their views on the teaching of EFL reading in their classroom practices?

Definition and aspects of reading comprehension
In the most traditional view, reading is seen as a linear process in which the reader decodes a text from its smallest units (words). Informed by behaviorism theory in the 1950s, reading began to be viewed as an activity where the reader responds to stimuli in printed texts with no sufficient explanation of how this process happens (Samuels & Kamil, 1988). This view has been criticized due to its strong reliance on learning the formal structure of sentences in a text. In the late 1960s and 1970s, researchers started to focus more on the reader's role and the reader's fluency, which led to the development of cognitive theories of reading that acknowledged the active cognitive process of activating background knowledge in the reader's mind (Goodman, 1967). Readers are active agents who sample the text, make propositions, and then confirm or reject the propositions (Goodman, 1967). In recent years, Goodman's psycholinguistic reading theory has been expanded to meta-cognition, which involves thinking about what one is strategically doing while reading (Grabe & Stoller, 2011). Carnine et al. (1990) also state that metacognition concerns learners' knowledge of and use of their cognitive resources, predicting, summarizing, clarifying meaning, and retelling. Thus, metacognitive theory deals with activities in prereading, while reading, and post-reading stages that learners need to undertake to comprehend a passage successfully.
Various definitions of reading from different schools of thought present different understandings of what reading comprehension is. Grabe and Stoller (2011) point out that reading comprehension is a set of complex abilities that cover rapid, efficient, interactive, strategic, flexible, evaluating, purposeful, comprehending, learning, and linguistic processes. These abilities are interrelated to support readers to achieve the ultimate goal of reading, that is, a full understanding of a particular text. In addition, reading comprehension is an interactive process between the reader and the text (Javed et al., 2015;Nuttall, 2005). There are several aspects essential to support better reading comprehension: comprehending the main idea of a text, recognizing the type of text, arranging the sequence of the text, predicting what a text is about, guessing the meanings of particular words from the text, identifying the specific information of the text, and using different reading strategies (Grabe & Stroller, 2011). Mastering these aspects leads students to comprehend the text better. Furthermore, making predictions using their prior knowledge of the structural organization and text features can further enhance students' understanding of the text being read (Palincsar, 2003). Besides, the amount of vocabulary also affects reading comprehension, where vocabulary development leads to improved reading comprehension (Renandya, 2004).
In sum, the structural, cognitive, and metacognitive theories of reading provide us with a better understanding of how reading is understood. These theories will be used as a guide to analyze the teaching of EFL reading in the Indonesian context. Based on the above discussion, teaching EFL reading should develop students' skills to comprehend written texts by developing various sub-skills, as mentioned previously. The teacher's role, then, should be as a mediator who facilitates students to go through activities that help them understand the text. In the next section, we will discuss the development of models for teaching reading implemented in EFL contexts.

Models of pedagogy to teach EFL reading.
In teaching EFL reading, three reading models are widely known: the bottom-up model, the top-down model, and the interactive model (Aebersold & Field, 2011;Clark & Silberstein, 1987;Zhang, 2016). The bottom-up reading model fosters reading instruction practices, which develops learners' decoding abilities from the bottom up, starting with the smallest units, single letters, letter blends, and building up to words and phrases. As this model started to become outdated, the top-down model emerged.
The top-down model views reading as a process of reconstructing meaning and stresses comprehension of units of meaning larger than words and phrases. The proponents of this model include Kenneth Goodman and Frank Smith (Dubin & Bycina, 1991). In the EFL context, this theory's application is evident in the adoption of instructional strategies that emphasize the active role of learners as readers. Examples of these activities include guessing the meaning of words from the context, previewing an article before reading it to have an overall view of the article, and actively engaging in predicting what the author might say next. In such a process, the reader uses his/her knowledge of vocabulary, sentences, discourse, and knowledge of the world.
The third model, the alternative model, combines the two earlier bottom-up and top-down models of reading. According to this model, reading involves the interplay of all meaning-gathering activities. This interactive theory acknowledges the role of previous knowledge and prediction, but, in the same time, reaffirms the importance of rapid and accurate processing of the text's actual words (Harmer, 2001). As he put it, "… sometimes, it is the individual details that help us understand the whole; sometimes it is our overview that allows us to process the details" (Harmer, 2001, p. 201). This model strives to develop students' metacognitive skills in reading. Harmer (2001) argues that reading instruction can be facilitated in various ways, such as providing an advance organizer (i.e., to activate the students' background knowledge) before reading, giving examples and more practice when reading, and providing feedback to students for the activities conducted in the pre-reading and while reading.
In the Indonesian context, the preferred model of pedagogy to teach EFL reading is the alternative model. It is suggested that teachers use the integration of the models of the reading process in EFL reading instruction (Widodo, 2016). Teachers are expected to strive for an optimum balance between the background knowledge presupposed by the texts the students read and the background knowledge the students possess. This is in line with the aim of the enactment of the scientific approach in the 2016 curriculum that is designed in a humanistic and constructivist paradigm in order to enable students to construct knowledge through reading ( Research on EFL reading instruction in Indonesia Yusuf et al. (2017) identify the problems that students in a senior high school in Banda Aceh encounter in the narrative reading classroom through the students' self-assessment. Dividing the students into two categories, good readers and poor readers, based on their reading test scores, the data are collected through questionnaires. This study reveals that both groups of students share similarities and differences in identifying the aspects of reading they find difficult. The significant difficulties faced by poor readers are identifying the main idea, understanding unfamiliar words, and understanding the characters' characterization. On the other hand, good readers find problems in predicting the possibilities in the story, determining the language features, and understanding unfamiliar words.
In addition, Hamra and Syatriana (2010) propose the development of an interactive model for teaching reading comprehension in five senior high schools in South Sulawesi, Indonesia. They propose an Interactive Model of Teaching Reading Comprehension (IMTRC) consisting of several learning aspects, including previous knowledge, reading skills, reading knowledge, reading participation, and learning environments, instructional reading materials, teacher participation and strategies, class interaction and conditions, group learning, individual learning, reinforcement, thinking in English, and evaluation. The study reports students' improvement in reading achievement after the implementation of this model. However, it is important to note that contextual factors will affect the success of this model if it is to be implemented in different contexts. Rusmawaty et al. (2018) conducts a case study on teachers' belief in using authentic materials for teaching EFL reading in a senior high school in Samarinda, Indonesia. The teachers are reported to show a positive attitude towards using authentic materials in reading classes. They believe that authentic materials, for example song lyrics and travel magazines, can boost students' motivation in reading. Implications for teacher training are then drawn based on the findings.
Tabiati (2016), who conducts a study on factors affecting learners' autonomy in reading at an English department in an Indonesian university, suggests that there are at least 28 factors, both internal and external, affecting the development of autonomous readers. Among the internal factors, motivation is found to be the biggest contributing factor to students' reading autonomy. Among the external factors, parental encouragement and feedback from the teacher are found to be the most significant.
There are a number of researches conducted in Islamic senior high school in Indonesia. One of them is Aratusa (2017), who investigates the use of content-based summarizing technique in improving students' reading skills. The result of her study reveals that the use of content-based summarizing technique has significantly improved students' reading skills to comprehend a reading text. Another research is conducted by Afriyeni (2018), who explores extensive reading to improve students' reading ability. Based on the study result, she proposes that teachers should help students to improve their reading ability by engaging them in a systematic extensive reading program. Rodli (2018), who examines the use of the reciprocal teaching method in reading instruction, asserts that the method may be used to improve students' reading comprehension.
As shown by the review of the literature above, research on teachers' views and practices of reading instruction in an Islamic secondary school setting remains a gap to be addressed. This research, therefore, attempts to fill this gap. The results of the study are analyzed and compared with the existing literature reviews in order to navigate its significance for extending knowledge in the field of reading instruction in the Indonesian Islamic secondary school EFL classroom.

METHOD
An explanatory case study design was used in this research since it is well-suited for the aims of this study, that is, to examine insight, discovery, and interpretation rather than hypothesis testing (Merriam & Tisdell, 2016), which extends to the 'how' and 'why' of an issue (Yin, 2018). Moreover, a case study design recognizes the particularity of the cases explored (that is, the quality of English reading instruction at Islamic senior high school in Indonesia) and does not seek to generalize research findings (Creswell, 2018). Data were collected from 31 senior English teachers at 31 Islamic secondary schools in six different provinces, including East Java, Central Java, Special Region of Yogyakarta, Lampung, West Kalimantan, and West Nusa Tenggara. The data collection methods comprised document analysis (i.e. the analysis of the English teachers' lesson plans and teaching materials), participant classroom observation (i.e.10 classes from different schools), personal in-depth interviews with 13 English teachers, written guided reflections about the teachers' beliefs and practices in the teaching of reading (31 teachers), and focus group discussions about what the teachers believe, understand, and experience about English reading instruction (31 English teachers). The data collection took approximately 12 months: six months in 2018 and six months in 2019. The data were analyzed using the constant comparative method (Olson et al., 2016;Corbin & Strauss, 2015). As categories emerged from the different data collection methods, they were compared to one another and the research participants' data. The data were read, and initial impressions that seemed relevant to the research questions were noted in the margins. Common themes were then noted in the margins, and these became the codes. This process yielded a rich understanding to answer the research questions as presented in the following section.

FINDINGS AND DISCUSSION
The analysis of the data revealed three emerging themes: (1) teachers' perception of good reading instruction; (2) teachers' views about the problems in teaching reading; and (3) teachers' practice of teaching reading in the classroom.

Teachers' perception of good reading instruction
The teachers' perception of good reading instruction is affected by their belief that reading is the most important skill in English language teaching. Therefore, they support the idea that the employment of resources and time in the English classroom should be for teaching reading. Furthermore, they propose that reading instruction should take up a bigger portion in the curriculum structure than other language skills. As the participants state in their guided reflection entries:

I believe that reading should constitute 50% of the whole English instruction. It's a vital skill tested in the final examination. (ER) Overall, my school gives reading a priority in the allotment of time in the English curriculum structure. Up to 70% of English instruction in the classroom is in the form of teaching reading. I think it's an essential skill to master for students living in rural areas like our school. As long as you can read in English, you can learn. It's not like they will use English with native speakers or something in our context (laughter). So speaking and listening is not a priority. But reading is. (JH)
From a total of six hours of English instruction at my school, probably 75% of it is for teaching reading. We just implement the policy from the top level. It is believed that reading is the main skill in learning English; that is why it is tested in the final examination. (AS) As evident in the data, the teachers believe that reading instruction should be the focus of ELT in their schools. The reasons range from the importance of mastering reading skills, preparation for final examinations, and policy implementation which mandates the importance of reading instruction. The focus group discussions' results indicate that teachers believe in the importance of reading instruction as one of the primary goals of language teaching.
Acknowledging the importance of reading instruction, the teachers believe that it is their duty to facilitate their students' best reading instruction (Richards & Rodgers, 2014). Their perceptions of good reading instruction, however, vary from one to another. Some teachers believe that good reading instruction simply means teaching students how to answer the test questions correctly. ES, for example, states: I think successful reading instruction enables students to understand the text to answer the questions based on the pre-designed indicators. It is about practice, practice and practice to do the test well. (ES) As visible in the excerpt, ES's concept of good reading instruction is affected by English testing's washback. She does not see the genuine purpose of reading as an activity to fulfill various goals, such as reading for information, research, or pleasure. Instead, she views reading and reading instruction as a skill to teach simply because of the curriculum and testing demands. The above-mentioned view corresponds with Bakken and Lund's (2018) study of Norwegian EFL teachers' understanding of teaching reading. Emphasizing the "means-oriented actions", teachers in their study reported particular classroom procedures-reading, translation, rehearsing, and testing the vocabulary-which often downplays students' agency in the classroom reading activities (Bakken & Lund, 2018, p. 82). Moreover, these routines are legitimized by the testing culture of ELT in many countries, such as Norway and China (Bakken & Lund, 2018;Luo et al., 2019). In the context of the present study, the teachers' belief in prioritizing teaching reading to the test can be understood within the heavy examination orientation characterizing ELT in Indonesia (Sukyadi & Mardiani, 2011).
Another view of good reading instruction shared by the participants is that good reading instruction should help students comprehend the texts effectively.
As far as I know, good reading instruction equips students with the techniques to comprehend texts. As teachers, we have to motivate students to read and enable them to obtain the benefits of reading the text. (JH)

Good reading instruction facilitates students to understand general and specific information in the texts to conclude the texts. (LW)
The above extracts indicate that the teachers have moved beyond a structural understanding of reading. Their views also extend beyond the belief that teaching reading is merely teaching students to answer test questions in English tests. They believe that they should facilitate students to activate their cognition to understand the text. These views correspond with Goodman's (1967) cognitive theory of reading comprehension as a process of activating background knowledge in the reader's mind to comprehend texts.
However, none of the participants' views about good reading instruction reflect the need to develop students' metacognitive reading skills. This finding is in line with Chinese EFL teachers in a study conducted by Luo et al. (2019) in which teachers are reported to find difficulties in using metalanguage in their reading instruction. The teachers are aware that reading involves the students' activation of schemata in the cognitive process (Barnawi, 2013), although they have not been able to extend their understanding to help students make efficient use of the cognitive process. Although the 2016 curriculum proposes the development of students' metacognition in reading as one of the teaching goals, the teachers have not incorporated this advice into their conceptual understanding of what good reading instruction entails. If these teachers are to teach reading effectively, they should reorient the goals of reading instruction to develop students' higher reading skills to enable them to achieve a deeper and richer understanding of the texts (Xu & Durgunoglu, 2020).

Teachers' views about the problems in teaching reading
To improve the quality of reading instruction in English classrooms in Islamic secondary schools, it is imperative to gain a thorough insight into the teachers' perceptions of teaching reading issues in their own contexts. The challenges of teaching reading as elaborated by the teachers participating in this research are: (a) lack of strategies to teach vocabulary; (b) limited repertoire of strategies to teach reading; (c) the use of teaching materials; and (d) improving students' reading motivation and autonomy.

Lack of strategies to teach vocabulary.
All the teachers in this study repeatedly mention vocabulary as the most challenging aspect of teaching reading that needs special attention.
The primary factor behind students' failure to understand texts during reading classes is their limited vocabulary. As a teacher, it is my job to teach vocabulary effectively. (AS) Most of my students find it challenging to comprehend texts because of their limited vocabulary. This becomes a problem as they are reluctant to bring and use a dictionary when they are reading. They rely on me as the only source of knowledge. Therefore, I am challenged to find a teaching strategy during reading instruction to enrich students' vocabulary. (PW) Vocabulary is a big problem for my students when reading. How can they understand what they are reading when they don't know what the words mean? So I struggle to teach them vocabulary in reading. Furthermore, the newest textbooks in the 2016 curriculum do not provide a list of keywords alongside each reading passage. (IS) The above excerpts show that vocabulary is the most influential reading aspect that teachers believe to be the most difficult challenge that students encounter in reading. The students' poor mastery of vocabulary is also associated with their reluctance to consult a dictionary when they find difficult words in the reading. One of the teachers also added that the omission of vocabulary lists in the 2016 curriculum textbooks is also problematic. These findings are in line with Yusuf et al.'s (2017) research, namely, that vocabulary is the biggest challenge in reading instruction. Similarly, this problem is also persistent for EFL Chinese teachers in teaching the vocabulary aspect of reading (Luo et al., 2019). This particular issue can be attributed to "teachers' insufficient knowledge of basic constructs related to literacy acquisition" (Luo et al., 2019, p. 279). This issue calls for innovation in reading instruction to support teachers to develop students' vocabulary in order for them to read effectively.

A limited repertoire of strategies to teach reading.
In addition to the above problem, the teachers in this study also explain that their limited understanding of strategies to teach reading has been an issue in their classrooms. They report that in teaching reading, they tend to use the same techniques for most genres of texts. One of the participants, PW, argues that the monotony of strategies used to teach reading has presented a big challenge for successful instruction.

She states:
Lack of knowledge about the most up-to-date teaching strategies is a problem. It can be disastrous because in the long run it will bore the students. Of course it will affect the reading instruction achievement in the classroom. (PW) I think there should be support for teachers to develop strategies to teach grammar and vocabulary for students. I used to teach students vocabulary by asking them to memorize words. It worked at some stages, but students won't understand how to use the words in contexts if I keep using it. Therefore, I need to teach them differently. (JH) From the data, it is apparent that teachers need support in developing teaching strategies that will improve the effectiveness of reading instruction. Although in this post-method era, the appointment of one teaching method as the most effective is no longer relevant (Kumaravadivelu, 2003), there is still a need to equip teachers with a range of teaching methods to teach reading (Richards & Rodgers, 2014). These findings corroborate Hamra and Syatriana's (2010) study, which claims that the development of new models and strategies to teach reading is essential in Indonesia.
In the FGD, the teachers revealed several strategies used in their reading instruction, such as group presentation, vocabulary memorization, pair work, and reading games. However, they believe that they need more strategies to optimize reading instruction in their classrooms.

The use of teaching materials
Another problem in teaching reading that the teachers raise is teaching materials that fail to attract the students' attention and improve their motivation to read. All the teachers in this study report that they use government textbooks in teaching reading. They seldom use authentic teaching materials that attract students' attention. AS explains the issue in the following extract.
Students' low motivation and achievement in reading can be attributed to the use of low-quality materials. I had the experience of using a celebrity magazine in teaching reading, and it worked very well to boost students' motivation to read. So we need to use materials that are interesting and suitable for the students' local context. Especially for students in Islamic secondary schools, it is sometimes more appropriate to use texts with certain religious content and context. (AS) As evident in the excerpt, AS believes that teachers need to be supported to provide students with engaging reading materials. It is further elaborated that teachers need to consider the particularity of the Islamic secondary school context in selecting and designing reading materials. This finding is in line with the research of Rusmawaty et al. (2018) in which teachers have a positive attitude towards authentic materials with a slight modification to boost students' motivation in reading instruction. Echoing Fuchs et al. (2019), we argue that this finding implies that teachers need to go beyond the prescribed textbooks in order to be able to utilize a broader range of reading materials to promote students' reading fluency.

Improving students' reading motivation and autonomy
The teacher participants in this study believe that reading instruction's success is also influenced by students' low motivation and autonomy in reading. This issue is explained by several teachers in the extracts below. The data indicates that one of the elements influencing reading instruction's success is the students' motivation. Consistent with many research studies in the ELT field, reading motivation is a strong predictor of reading achievement (Lin et al., 2012;Xu & Durgunoğlu, 2020). Therefore, the teachers believe that they need to be equipped with strategies to improve students' motivation and autonomy to read English passages (Benson, 2013). This issue echoes Tabiati's (2016) research findings, namely that there needs to be an improvement in students' motivation to improve their reading autonomy and achievement.

Teachers' practice of teaching reading in the classroom
The result of classroom observation on reading instruction by all the teachers reveal that the practices are dominated by testing students' reading sub-skills with a limited explanation about how to improve these sub-skills, such as locating the main idea of a text, explaining the purpose of a text, and explaining the meaning of certain expressions in a text (Brown, 2007).
ER, for example, started her reading instruction with a grammar explanation and drilling. Then, she asked her students to listen to and imitate her reading the passage to elicit correct pronunciation. After that, she distributed passages and questions to the class. The main activity was the students reading and doing the reading exercise on their own. We saw minimum supervision, feedback, and explanation from the teacher (Klenowski, 2009).
In another school, JH began the reading class by asking the students to scan for difficult vocabularies and find their meanings in the dictionary. She then asked the students to read one paragraph silently and randomly appointed one student to explain the text's main idea. Without explanation of how to find the main idea, the teacher asked the students to read the whole text and answer the text's questions. A similar approach was used by LW, AS, and PW in their reading classes.
Generally, the teachers discussed the problematic or unusual vocabulary found in the reading texts, asked students to find the main ideas, and answered other questions related to the reading. There was no instruction on the strategies to read effectively. Although these teachers reported the belief about their role to facilitate students' reading comprehension, there is a gap in how this belief translates into their teaching practices. This problem can be attributed to many factors, such as their limited awareness of research-based reading instruction (Fuchs et al., 2019), the examination orientation which takes up too much space in the reading curriculum (Luo et al., 2019), and their limited understanding of practical strategies to teach reading effectively (Bakken & Lund, 2018). All these issues call for attention from teacher education programs to support teachers in improving their reading instruction.

CONCLUSIONS AND IMPLICATIONS
Despite admitting the importance of teaching reading in schools, the 31 English language teachers from six Indonesian provinces, who were the research participants, found teaching reading challenging. They identified several problems in reading instruction in their classrooms, including (a) lack of strategies to teach vocabulary; (b) limited repertoire of strategies to teach reading; (c) the use of teaching materials; (d) improving students' reading motivation and autonomy. Despite the teachers' views on the importance of reading, they did not employ methods for teaching reading that were able to develop the students' reading ability. The teachers tended only to test the students' reading ability rather than to teach them. This confirms a classic issue in ELT that reading is oftentimes a skill that is tested, not taught. The teachers were more inclined to test or assess students' competence than to teach them. The teachers asked students to complete numerous exercises about the content of reading texts, conducted orally (both individually and in groups), then the teachers checked and confirmed the accuracy of the answers. These findings imply that there needs to be an improvement in how reading instruction is conducted in Islamic secondary schools. This calls for the development of a teaching reading model that can effectively develop students' reading ability. This model should then be introduced to teachers at Islamic secondary schools in Indonesia to be considered for their schools.