The Impact of Metacognitive Instruction on ESL University Level Students’ Awareness and Use of the Reading Strategies

Previous studies related to metacognition and reading have extensively investigated metacognitive awareness of reading/reading strategies of students at a particular point in time and/or relation between metacognitive awareness and reading ability. However, research illustrating the changes in students’ awareness and use of the reading strategies over a period of time is strikingly absent from the literature. The present study was designed to explore qualitatively the impact of the metacognitive reading strategy instruction on students’ awareness and use of the reading strategies introduced over a period of time in university level ESL classrooms in Pakistan. It also offers a view of the students’ experience of such instruction. The study comprised of two action research cycles spanned over 4 months each. Eight undergraduate university students were selected as the participants of the study. Data collection instrument included interviews, think aloud protocols, learner diaries, end of class feedback, note-taking, researcher journal, and Survey of Reading Strategies (SORS) questionnaire. The findings revealed that metacognitive reading strategy instruction not only raised metacognitive awareness of the introduced reading strategies in students in a “real” classroom setting but also made them use the knowledge they gained during the study. Implications of the study are discussed.


Introduction and Overview
The ability to read is often associated with academic success (Ghaith & Sanyoura, 2019;Meniado, 2016;Van den Broek et al., 2007).Indeed, research indicates that ''there is strong interface between one's ability to read strategically and one's ability to excel academically'' (Mokhtari et al., 2008, p. 46).This implies that students must learn to read strategically.
Reading research has increasingly devoted considerable time and effort to figuring out how best to enable students to become strategic readers (Alexander & Jetton, 2000).A number of studies conducted in this regard indicate that students could be helped to become strategic readers by understanding how to use reading strategies so as to aid comprehension (Paris et al., 1983).However, research also points out that only teaching students reading strategies is not enough (Baker & Brown, 1984).Rather, it is vital to teach students to ''regulate'' or monitor the reading strategies they use in order to ensure success in reading comprehension (Baker & Brown, 1984;Griffith & Ruan, 2005).In other words, comprehension instruction should foster metacognition in students (Baker, 2002).
With this in view, the study reported here aimed to investigate if metacognitive reading strategies instruction could bring about any change in metacognitive awareness and use of reading strategies of students studying at the tertiary level in the Pakistani context.
The study explores the following research questions: Q.1 What changes, if any, does metacognitive instruction have on students' awareness of the reading strategies introduced during the study?
Q.2 What changes, if any, does metacognitive instruction have on students' use of the reading strategies introduced during the study?

Background
The literature review in this paper is divided into two sections.The first focuses on metacognition and discusses the key construct ''metacognitive awareness.''The second part reviews a selection of empirical studies which examine adult L2 readers' metacognitive awareness and strategy use of reading and suggest the gaps that are potentially filled by this study.

Metacognition
Metacognition is most commonly understood as encompassing knowledge about, as well as the regulation of, cognition (Flavell, 1976).It is widely defined in terms of the knowledge and regulative components of cognition (Dinsmore et al., 2008;Guo, 2018;Schunk, 2008).In the literature, the knowledge and regulative components of metacognition are also referred to as ''awareness and control,'' or ''self-appraisal and self-management'' (Brown, 1987;Cross & Paris, 1988;Paris & Winograd, 1990).
The first component of metacognition, metacognitive knowledge ''is also labelled as metacognitive awareness'' (Pintrich et al., 2000, p. 45) in some models of metacognition.Metacognitive knowledge includes two components: knowledge about cognition and awareness of one's own cognition (Harris et al., 2010).Metacognitive knowledge is ''that portion of the total knowledge base that pertains to a given area of cognitive activity'' (Flavell, 1985cited in Garner, 1987).Metacognitive knowledge is usually referred to as statable information about cognition (Baker & Brown, 1984;Pintrich et al., 2000).It is statable since ''one can reflect on the cognitive processes involved and discuss them with others'' (Brown et al., 1983, p. 107).Flavell (1985cited in Garner 1987) and other theorists (e.g.Pintrich et al., 2000) point out that metacognitive knowledge can be retrieved and used during a cognitive enterprise either automatically or deliberately and its basis can be flawed just as the basis of other kinds of knowledge.Bransford et al. (1999) state that metacognitive knowledge gained in one context could be used in another.Research demonstrates that metacognitive knowledge usually improves performance (Schraw & Moshman, 1995).Reading research indicates a causal role of metacognitive knowledge in reading comprehension (Ozturk & Senaydin, 2019;Pressley et al., 1992).
Metacognitive knowledge is broken down into declarative, procedural, and conditional knowledge (Brown, 1987;Dunn, 2017;Garner, 1987;Jacobs & Paris, 1987;Paris et al., 1983;Schraw & Moshman, 1995;Soto et al., 2019).Harris et al. (2010) state that the effective use and coordination of all these three facets of metacognitive knowledge ensures academic development and performance.Declarative knowledge is the ''knowledge of the what of cognition'' (Pintrich et al., 2000, p. 48).In other words, it refers to ''knowing what or knowing that'' (Schmitt, 2005, p. 102).To be a good reader one must have declarative knowledge about self, task and strategy variables (Schmitt, 2005).With respect to reading strategies, declarative knowledge is the knowledge about what strategies are and ''can help readers in setting goals and adjusting actions to changing task conditions'' (Paris et al., 1983, p. 303).Procedural knowledge includes knowing the procedure for performance and use of the cognitive strategies (Pintrich et al., 2000).Procedural knowledge is important ''since one must know how to perform various strategies involved to be successful'' (Schmitt, 2005, p. 102, emphasis in original).Also, procedural knowledge is significant since it helps students perform task more automatically (Pressley et al., 1987) as well as use different strategies for problem solving (Glaser & Chi, 1988cited in Schraw, 1998).Paris et al. (1983, p. 303) state that procedural knowledge is often gained from explicit instruction or from experiences that are repeated.Since declarative and procedural knowledge deal with only the knowledge and skills that are needed for performance, Paris et al. (1983, p. 303) introduced a new term, conditional knowledge, to talk about the ''conditions under which one might wish to select or execute actions.''Conditionalknowledge includes ''knowing when and why to apply various actions'' (Paris et al., 1983, p. 303).Conditional knowledge helps in more effective use of strategies (Reynolds, 1992) as it provides a rationale for their use (Paris et al., 1983).Having a rationale for executing an action could make that action appear valuable to students who could then become motivated to execute it.As Paris et al. (1983, p. 312) put it: ''Conditional knowledge is the glue that holds skill and will together.''Conditional knowledge is often equated with executive control since it deals with online monitoring of cognition (Schmitt, 2005).It is well established in the literature that without conditional knowledge an individual having complete procedural knowledge is not able to adjust his/her behavior according to the demands of the changing task (Paris et al., 1983).
Metacognitive awareness is regarded as a critical element of skilled, strategic reading (Auerbach & Paxton, 1997;Carrell et al., 1989).Perhaps this is why increasing metacognitive knowledge of reading processes is considered to be ''an important first step towards becoming constructively responsive, strategic, and thoughtful readers.''Many researchers (e.g.Casanave, 1988;Mokhtari & Sheorey, 2008) have emphasized the centrality of metacognitive awareness in reading comprehension and reading strategies.In this regard, Pressley and Gaskins (2006, p. 103) state that ''good readers possess metacognitive knowledge about reading strategies.''According to Sheorey and Mokhtari (2001, p. 433), ''the reader's metacognitive knowledge about reading includes an awareness of a variety of reading strategies and that the cognitive enterprise of reading is influenced by this metacognitive awareness of reading strategies.''The knowledge about what, how, when, and why associated with the use of strategies plays a critical role in promoting the autonomous use of strategies (Pressley et al., 1989;Rahimi & Katal, 2012).

Studies of Metacognitive Awareness and Strategy Use
Reading research started documenting findings related to L1 children's metacognitive awareness in 1970s (El-Hindi, 1993).Since the 1980s, a number of studies related to L2 adult metacognition and reading has been conducted (e.g.Anderson, 1991;Carrell, 1989;Knight et al., 1985).
Related to the first aspect of metacognition, knowledge of cognition, an early study was carried out by Devine (1984) who investigated adult ESL readers' conceptualizations of the reading process and the impact of their conceptualizations on their reading performance.The participants in the study were 20 beginning/low intermediate level students studying in a communitybased ESL program in US.Their ages ranged from 19 to 40.Data for the study was collected by means of an oral reading interview, a sample of oral reading and a retelling of oral reading.Results showed that ESL readers hold sound-, word-, or meaning-oriented model of reading which they brought with them to the reading classroom.The study also revealed that meaning centered readers showed good to excellent comprehension on the retelling task, while sound-centered readers demonstrated poor or very poor comprehension.The comprehension pattern of the word-centered readers showed that apart from one reader who apparently comprehended the text very well all the other readers demonstrated poor or fair comprehension.The writer suggests that the teachers of L2 may help students in adopting a meaning-centered approach to reading as it seemed to predict more successful comprehension.This study generated much interest in the models of reading held by adult L2 readers.However, surprisingly, there is not much research activity on this aspect of metacognitive knowledge of adult L2 readers in the literature.
Much reading research carried out on adult L2 students has focused on the relation between their metacognitive awareness about reading strategies and reading achievement.An early study that investigated the correlation among reading comprehension, the use of strategy and perceived strategy use was carried out by Barnett (1988).The participants in this large scale quantitative study were 272 students learning fourth semester French in a university in US.They read an unfamiliar passage, answered questions before reading the text, wrote a recall protocol as well as the ending of a text in English.A questionnaire based on reading strategies that students perceived they use while reading was also filled in by the participants.The study found a strong correlation among reading comprehension, strategy use and the perception of strategy use in the university level students.Carrell (1989) also reports the metacognitive awareness of various types of reading strategies by two groups of second language learners in both their L1 and L2, and the relationship between their awareness and reading comprehension.The first group included 45 native speakers of Spanish at intermediate to advanced levels of English enrolled in an ESL program at an American University.The participants in the second group were 75 English native speakers studying intermediate to advanced levels of Spanish.The study employed a metacognitive questionnaire that tapped the participants' metacognitive awareness about the strategies they use while reading in both their L1 and L2.The participants also read a text in both their first and second languages and answered multiple choice questions related to the text.The study results were reminiscent of Devine's (1984) study, as they showed that in general the better readers in both groups as compared to the lower-level students did not focus on local or bottom-up processing.These studies join several others earlier studies within the literature that tended to explore the relation between students' metacognitive awareness of reading strategies and their reading ability.Investigation of the relation between metacognitive awareness and reading ability has remained an active and important area of inquiry in L2 studies.Recently, for instance, Dhanapala (2010) investigated metacognitive awareness of 169 Sri Lankan university students with respect to their reading proficiency levels.For data collection, a reading comprehension test is used, accompanied by Metacognitive Awareness of Reading Strategies Inventory (MARSI).Like earlier studies, the writer found a positive linear relationship between students' metacognitive awareness of reading strategies and their text comprehension.
Some recent research has gone beyond demonstration of correlation between metacognition and reading ability and has explored differences in metacognitive awareness and perceived use of reading strategies among native and non-native readers of English.Sheorey and Mokhtari (2001) in a large scale quantitative study, for instance, investigated the differences in the metacognitive awareness and perceived use of reading strategies among 105 native English speakers and 152 ESL students studying in a university in US while reading academic texts.For this purpose, they used the Survey of Reading Strategies (SORS) questionnaire.The findings revealed, among other things, that both US and ESL students were aware of almost all strategies that were listed in the SORS.In addition, in both group the students with high reading ability reported higher reading strategy use than the students with the low reading ability in the respective groups.
Mokhtari and Reichard ( 2004) also investigated whether significant differences existed in the metacognitive awareness and perceived use of specific strategies when reading academic texts in English between first and second language readers.In more detail, they assessed metacognitive awareness of reading strategies of 141 American and 209 Moroccan students studying in an American and Moroccan university respectively.They administered MARSI.The findings revealed that American as well as Moroccan students demonstrated similar patterns of awareness and use of strategies although they studied in different socio-cultural environments.Specifically, both groups exhibited a moderate to high level of awareness in terms of reading strategy use.Concerning the perceived strategy use, there were certain strategies that Moroccan students stated to be using comparatively more than American students.
More recently, one study has branched out into a new direction by examining the differences in metacognitive awareness of reading strategies between foreign and second English language learners.This was the major purpose of the Karbalaei's (2010) study that investigated possible significant differences between 96 Iranian and 93 Indian college students' metacognitive awareness of reading strategies.The participants completed the Metacognitive Awareness of Reading Strategies Inventory (MARSI) after taking a reading comprehension test.Results showed a similar pattern of strategy awareness while reading academic texts between the students of both groups.Results also revealed that Indian students comparatively demonstrated a higher metacognitive awareness and strategy use than Iranian students.
These studies have offered rich insights regarding adult ESL students' metacognitive awareness of reading strategies but arguably, they captured only two aspects of metacognitive awareness.Essentially, the studies either investigated metacognitive awareness of reading/ reading strategies of students at a particular point in time and/or relation between metacognitive awareness and reading ability.The change in metacognitive awareness and use of reading strategies in adult ESL students over a period of time is a topic that remains to be explored in the literature as identified by Griffiths (2013, p. 83) who pointed out ''the vast majority of strategy research has looked at a sample of learners at a particular point in time rather than trying to measure changes over a period of time.''Researching the latter topic is further compelled by the fact that in the area of metacognitive awareness researchers ''are still concerned with causal questions'' (Baker, 2008, p. 37), although longitudinal designs could be valuable for studying temporal development of metacognitive awareness and use.The current study extends the research in this area by illustrating the changes in adult ESL students' awareness and use of reading strategies over a period of time in a ''real'' classroom setting.It is important to do so also because Oxford (2017, p. 44) contends that ''the intensity with which quantitative strategy research has been pursued has probably delayed the world's understanding of contextualization of learning strategies.''The study also sheds light on the change in metacognitive awareness and use of reading strategies of Pakistani university level students, which has not been reported in the literature to date.Baker and Cerro (2000, p. 129) indicates that assessment is ''still far from having adequate tools for measuring metacognition.''The literature therefore suggests that researchers should use as many research methods as possible to obtain converging evidence regarding students' metacognition.In this regard, for instance, White (1988:, p. 74) wrote, ''though each method is weak, the constellation of evidence from them will be more reliable and valid than each alone.''Being mindful of this, this study employed multiple methods for data collection.The interviews, think-aloud protocols (TAP), learner diaries and researcher journal were the main data sources in the study.Interviews and TAPs were carried out at the start, middle and end of the study in both action research cycles.Additional information was collected by means of end-of-class feedback of students and the SORS questionnaire which was used primarily for pedagogical purposes at the start and end of the study.Notes were also taken by the researcher during the fieldwork.The strategies that were introduced during the lessons over a period of 4 months in each cycle were prediction, activating prior knowledge, identifying the main ideas, skimming, and scanning.

Methodology
In the current study, different techniques drawn from qualitative approaches were used to provide validity checks on the action research data so as to strengthen the trustworthiness of this research.In more detail, internal validity in this research was enhanced through triangulation, an approach commonly used in qualitative research for checking validity (Burns, 1999, p. 163).Triangulation ''entails using more than one method or source of data in the study of social phenomena'' (Bryman, 2004, p. 275).Burns (1999, p. 163) points out that ''action researchers use multiple methods and the perspectives of different participants in order to gain a richer and less subjective picture than they obtain by relying on a single data gathering technique.''Using different research methods also enables researchers to ensure that what is reported is well-supported and evidenced across different sources of information.This study, conducted during a period of 4 months in each cycle, therefore used multiple data collection methods as mentioned earlier.In addition, to enhance the internal validity of this research member checking was carried out by getting feedback or comments from the research participants on interview data and TAP data.Moreover, a bilingual English teacher who did her Masters in ELT from a university in the UK was involved in checking translations.To enhance external validity in this study, interview scripts, research journal, end-of-class feedback were used for providing detailed accounts of the participants' standpoint to readers.Burns (2010, p. 135) points out that in an action research study the data analysis ''does not have to wait until the end.' ' Braun and Clarke (2006) also indicate that in qualitative research the process of analysis can start during data collection when the researcher notice and identify interesting features in the data.In this study, the process of analysis started in the act of teaching in both cycles of the study.The teacher-researcher adopted thematic analysis method for analyzing interviews, thinkaloud protocols (TAP), learner diaries, researcher journal and end-of-class feedback of students.Thematic analysis involves identifying, organizing and reporting themes and sub-themes in a text that could result in ''rich and detailed, yet complex, account of data'' (Braun & Clarke, 2006, p. 78)

Participants
The study was carried out in a public sector university of Pakistan.It comprised of two action research cycles spanned over 4 months each.The participants of the both cycles comprised of first year Bachelor of Arts (BA) students from a department from the Faculty of Science.The group for the first cycle comprised of 30 Pakistani students, 7 male and 23 female, while the second group had 26 students altogether, 5 male and 21 female.The entire groups that ranged in age 19 to 20 were invited to take part in the current action research study.However, four participating students from each group were selected for the in-depth data collection.These students represented different proficiency levels in reading.
Reading proficiency was operationalized in terms of the grades obtained on the University of Warwick's English Language Test (WELT, 2005).In the first cycle, Saba scored A grade, Furqan scored B grade, Khadija scored C grade, and Ali scored D grade.In the second cycle, Marium scored B grade, Rida scored C grade, Nida scored D grade, and Sarah scored E grade.For interpretation of scores see Appendix A (this paper uses pseudonyms to protect the anonymity of the students).

Results
Data gathered during the study yielded evidence of change that occurred in the participating students' metacognitive awareness and use of the reading strategies that were introduced to and practiced during the study.
Increased students' awareness of the reading strategies they already use.The study reveals that three participating students in each cycle, namely, Furqan, Khadija and Ali in Cycle 1 and Rida, Nida and Saba in Cycle 2, who at the start of the study exhibited no awareness of the reading strategies they had used in the past before reading an academic text, demonstrated an awareness of them within 2 months of instruction.The findings, moreover, reveal that one participating student of each cycle realized within 2 months of instruction that they had less than fully comprehensive metacognitive knowledge of the reading strategies they already used.More specifically, Saba in Cycle 1 reported realizing that she lacked both declarative and conditional knowledge of the strategies of skimming and scanning: I am so surprised that I always used to do skimming and scanning and I never knew what's the basic difference between them.Like our teachers used to do skim and scan . . .when people use to ask me I also very boldly used to tell them that this means this and this means that.And we learnt in class that day that oh my god this is the difference between them.Purposes are different why we use them even that purpose is different.For what we use them even that is different.So I was really surprised that what I know is superficial.I don't know in detail.(Extract from Saba's Interview) Likewise, Nida in Cycle 2 reported that she lacked declarative and conditional knowledge of the other two strategies, namely prediction and activating prior knowledge as illustrated from the following extract: Researcher: Were you aware of the strategies such as prediction, activating prior knowledge that we studied in the classes?Nida: No maam.I mean a little ( . . . ) Out of these I used to use some.But I was not aware why I was using them, sometimes I used them sometimes I did not.
And I was not aware of why I was using them.And I was not aware of what I was doing.Researcher: Alright.So which strategies were you aware of prior to the start of the semester?Nida: I knew that by looking at the title I used to guess what is in the text, used to activate prior knowledge.I used to do these but I did not know if I should do this or not.(Extract from Nida's interview) What is noteworthy is that in this paper the extracts have not been punctuated, rather using the convention of ''sic'' attention to students' errors have been drawn.''Sic'' has not been used partially because students made quite a few errors related to pronunciation and sentence structure as they tried to communicate their ideas and experiences in English, and also because this study does not aim to draw readers' attention to the errors students made during speaking or writing.
Change in students' awareness and use of the reading strategies that were practiced during the lessons Interviews and learners' diaries.A common finding from the interviews and learners' diaries is that all four participating students of each cycle reported gaining declarative knowledge of the introduced strategies during the study.For instance, Sarah in Cycle 2 reported at the end of the study that she became aware of several strategies during the lessons: Initially I did not know that any strategy is used during reading.I did not know any such thing.But then I learnt gradually about strategies.First I learnt about pre-reading strategies like activating prior knowledge, prediction.Then I learnt how they will be used.I learnt about skimming, then scanning.(Extract from Sarah's Interview) This extract above indicates that Sarah had no declarative knowledge of the strategies of ''activating prior knowledge,'' ''prediction,'' ''skimming,'' and ''scanning at the start of the study.''However, it appears that she gradually ''learnt about'' the introduced strategies.In addition, she also ''learnt how they will be used.''Another common finding from interviews and learners' diaries is that all participating students' gained conditional knowledge of using the strategies introduced during the study.However, the data also shows that those participating students of the first cycle who had very good or good reading ability developed a more comprehensive understanding of ''why'' and ''when'' the strategies are used as compared to those students who had average or low reading ability.For instance, Furqan in Cycle 1 who had good reading ability reported in his diary ''why'' in his opinion applying the strategy of skimming is helpful.He also reported ''when'' and ''how'' he intended to apply it: I will always do skimming every time before I read something (unless it's a narrative), it gives an overall flavor of the text, clear the mind and gives us an idea that what is in the text so that we could expect more.Skimming also helps putting interest in the text.It saves time off course.I will now skim text using the strategy of reading the first and the last paragraph and first line of each paragraph because that works for me quite efficiently.(Extract from Furqan's diary) On the contrary, Ali in Cycle 1, who had low reading ability, did not seem to have a conscious awareness of the fact that the strategy of activating prior knowledge is used both before and while reading, although during the thinkaloud protocol he had employed the strategy of activating prior knowledge during reading as well.Ali's lack of conditional knowledge became evident when he reported at the end of the study that he mostly realized after starting to read that he had not activated his prior knowledge: Ali: I have used the strategy of activate prior knowledge.
But mostly after starting to read I remember that I have not activated my prior knowledge.Researcher: So what do you do?Ali: Rather than reading from the beginning again I decide that I would inshallah remember using it the next time if I would read the text again.(Extract from Ali's interview) The foregoing finding that Furqan with very good or good reading ability better gained conditional knowledge of the strategies introduced had implications for the second cycle of the study.It suggested that the teacherresearcher should spend more time discussing the ''when'' of the strategy use during the second cycle so that students of all abilities could gain conditional knowledge of the strategies equally well.This was kept in view in the second cycle.The data from the second cycle showed that all four participating students gained conditional knowledge of the introduced strategies.
The data also shows that all participating students of each cycle not only developed metacognitive knowledge of the reading strategies introduced, but also became aware of the knowledge they gained during the study.At the end of the study, for instance, Marium in Cycle 2 reported that as compared to before she has gained knowledge of reading and the reading strategies: Now I know what is reading.I did not know before what is reading.And it was beyond possibility that I would know of reading strategies.Now I know reading strategies and also how to use them.(Extract from Marium's Interview) Related to students' awareness of the reading strategies is the finding that students started using the strategies they became aware of on the texts they read during the study.However, some students used all strategies effectively during reading as compared to others.For instance, Saba in Cycle 1 who had very good reading ability employed strategies well during reading.The following extract illustrates her effective use of the activating prior knowledge and prediction strategies while reading the text ''Will Climate Change Lead to Conflict or Cooperation?''I used my prior knowledge and prediction that if climate change is talked upon, usually global warming will be focused on as previously we were asked to evaluate changes in weather if any and changes are because of global warming.Secondly, if conflicts is mentioned then maybe lack of resources and fight over resources is discussed in the text as I know that it is proposed that the third world war is supposed to break out on scarcity of water.Thirdly if cooperation is discussed then it must be regarding achieving sustainability.I want to find out what are the risks of conflict and advantages of cooperation so that I can frame my mind.(Extract from Saba's diary) Evidence from this show that Saba did not simply ''dive into a text'' (Pressley & Afflerbach, 1995, p. 32).Rather, she predicted what the text would be about and had a ''goal in processing the text'' (ibid.).
Think aloud protocols.The further confirmation of the changes in students' awareness and use of reading strategies was found during TAPs.The data reveals that all eight participating students, for instance, made predictions in the second and third think aloud, although none of them used the strategy of prediction during the first think aloud.In more detail, in the second and third think aloud students made prediction about the text by reading its title.As an example, in the second think aloud after reading the title of the text Rida in Cycle 2 predicted that the text would discuss the problems of the education system of Pakistan: On reading its title ''Pakistan education emergency: failing its future'' I could see that it might be about education system of Pakistan, that it has a few flaws and in future it could fail.(Translation from Rida's TAP) Besides that, the data shows that during the third think aloud out of the eight, two participating students predicted what the paragraph would discuss next on the basis of the text they were reading.For instance, Marium in Cycle 2 while reading the text ''The Debate about the Semester System'' tried to predict the opinions of those who favor the semester system: It is claimed that teachers are subjective in their approach and go by personal whims by teaching and awarding mix to the students.On the other hand the supporters of the semester system have their own opinions.What are their opinions?They favour the, might they favour the semester system because they used this system.(Note that in the abstract taken from TAP, the text in bold reflects the text read by the student.The text in italics show the spoken thought of the participant.)(Extract from Marium's TAP) The use of the strategy of prediction by all eight participating students indicates a clearly identifiable change in students' procedural knowledge of this strategy.
The data from TAPs also reveals that all eight participating students activated their prior knowledge during the second and third think aloud, just as they did in the first think aloud.However, unlike the first think aloud, in these think alouds students were aware that they had used the strategy of activating prior knowledge during reading.The following example from Nida in Cycle 2 illustrates how she reported her use of the strategy of activating prior knowledge when elicited about her reading process during the third think aloud: Researcher: What did you do to understand the second paragraph?Nida: (Lines deleted).I have activated my prior knowledge as we are experiencing semester system in the university and have experienced annual system in intermediate.(Extract from Nida's TAP) Besides that, the data shows that two of the participating students, Rida and Nida respectively, identified the main idea of two paragraphs during the third TAP.An instance of Rida's use of the strategy of identifying the main idea is evident from the following extract: Quizzes, assignments and presentations keep the students on their toes.All these activities develop confidence in the students.Prompt feedback by the teachers makes learning an active activity.The focus in the semester system is on learning and developing skills rather than preparation for examination.There is very little room for cramming in semester system.The main idea of this paragraph is developing skills of the students by the teachers.(Translation from Rida's TAP) The finding that in addition to the strategy of prediction, Rida and Nida started using the strategy of identifying the main idea by the end of the study indicates a significant growth in their procedural knowledge of the strategy since at the start of the study they did not use this strategy during TAP.

Questionnaire:
Survey of Reading Strategies (SORS).Further evidence of change in students' awareness and use of the strategies of prediction, activating prior knowledge and skimming is evident through a comparison of the SORS questionnaire that students filled in on the first and the last day of the study in both action research cycles.Specifically, as shown in Table 1 below more students of the both groups started using the strategies of prediction, activating prior knowledge and skimming by the end of the study.
The above Table shows that the number of students who reported that they ''always or almost always'' used the strategies of activating prior knowledge and prediction almost doubled at the end of the study.Moreover, a large number of students reported that they ''usually'' used the strategy of prediction by the end of the study.The data indicates that students' enhanced use of the strategies introduced by the end of the study is also evident from the fact that while the number of students who ''never or almost never'' used the strategy of skimming from ten decreased to zero out of 56, it decreased from 10 to 1 in the case of the strategy of prediction.

Discussion
In keeping with the research questions, this section examines the core findings related to the changes in the participating students' awareness and use of the reading strategies within the context of the literature on metacognitive reading strategy instruction.In doing so it also illustrates the development of awareness and use of reading strategies in participating students over a period of time.
In terms of change that metacognitive instruction has on students' awareness of the reading strategies (research question 1), there appears to be little doubt of an overall gain in participating students' awareness of the reading strategies during the study.Particularly, as mentioned earlier, the study reveals that six participating students reported awareness of the reading strategies they had used in the past.The gains in students' awareness of the reading strategies that were already in their repertoire of use was of particular importance to the study as the literature stresses that ''helping learners become metacognitive about the use of strategies in their current repertoire is more effective than asking them to learn to use different and new strategies'' (Griffith & Ruan, 2005, p. 13).The findings related to students' gain in declarative, procedural and conditional knowledge also contribute to the existing research in the field since change in students' metacognitive knowledge of reading strategies over a period of time is a topic that has not been explored in the literature, as mentioned before.
The results also reveal that the participating students became aware of the reading strategies that were focused during the study.Evidence of this can be seen in the fact that almost all students gained declarative, conditional and procedural knowledge of the introduced reading strategies.More specifically, the findings reveal that all the participating students in both cycles exhibited gaining declarative knowledge of the strategies of prediction, activating prior knowledge, skimming, scanning, and identifying the main idea.The participating students' conditional and procedural knowledge of the reading strategies introduced were also bolstered during the study.However, in Cycle 1 two participating students, namely Saba and Furqan, demonstrated a thorough understanding of the why, when and how to use the strategies introduced.In contrast, in Cycle 2, all four participating students exhibited comprehensive conditional and procedural knowledge of the introduced strategies.Apparently, there are two possible drivers of greater gains of procedural and conditional knowledge of reading strategies in all the participating students in Cycle 2. First as compared to Cycle 1, in Cycle 2 the teacherresearcher spent more time discussing when and why to use the introduced strategies to develop students' conditional knowledge, in keeping with the decision taken at the end of Cycle 1.Second, the researcher provided students in Cycle 2 with more opportunities as compared to students in Cycle 1 to reflect and discuss with others how they used the introduced reading strategy.These factors could have accounted for enhanced conditional and 8 SAGE Open procedural knowledge in students in Cycle 2 as the literature claims that ''students will transfer a strategy to their tasks if they possess the 'how to employ' or procedural knowledge of that strategy and the 'why and when to use' or the conditional knowledge'' (Simpson et al., 2004, p. 3).Overall, the finding that the participating students became aware of the strategies introduced corresponds to Kutlut€ urk's (2020) finding that indicated that university level students showed awareness of the strategies taught as a result of instruction.The finding is also consistent with Huang and Newbern's (2012) study that showed that adult ESL students with limited English and literacy skills showed awareness of the strategies taught after receiving metacognitive reading strategy instruction for 4 months.They reported that the students in their study could name the strategies and could explain when and why they used the strategies, just like students of this study.
What is noteworthy is that the foregoing findings also show that the participating students' reading abilities were not necessarily related to their reported awareness of the reading strategies.This is evident from the fact that Furqan in Cycle 1, who had good reading ability demonstrated no awareness of the strategies he used before reading an academic text prior to the start of the study.This finding contrasts with the existing literature that shows that efficient readers exhibit an awareness of strategies when reading in English and other languages.For instance, Hamiddin and Saukah (2020) study showed that good ESL readers encompassed more metacognitive awareness as compared to the less successful readers.Likewise, Sheorey and Mokhtari's (2001) study showed that good L1 and ESL readers of English were aware of the reading strategies, and knew how to utilize them.
The findings of the study also reveal that Saba in Cycle 1, who had very good reading ability, lacked declarative and conditional knowledge of the strategies of skimming and scanning.(e.g.Forrest & Waller, 1980;Ghaith & El-Sanyoura, 2019;Myers & Paris, 1978;Phakiti, 2003).To put it another way, the literature points out that only ''reading novices have some metacognitive deficiencies in them.These deficiencies appear in both declarative and procedural knowledge categories'' (Garner, 1987, p. 36).In one study, for instance, Gambrell and Heathington (1981) investigated good and poor adult readers' metacognitive differences in strategy awareness.They found that as compared to good adult readers, poor adult readers reported awareness of fewer strategies and were not aware of how and when to use them.Wu (2002) also investigated if students having high proficiency differed from students having low proficiency in terms of their reading awareness.His study showed that more proficient readers appeared to have more awareness of metacognitive skills as compared to low proficient readers.I suspect that one reason for the lack of relation in students' reading ability and their metacognitive awareness of the reading strategies may lie in the participating students' prior educational experiences.The participating students were taught to read each and every word during reading.This could have developed a word-centered theoretical orientation of reading in students.This in turn could have developed a deficiency in students' metacognitive knowledge of reading strategies even though some of them learnt to read well.This assumption is in line with the literature that suggests that some in-school reading experiences could actually produce the knowledge problems in students.For instance, Garner and Kraus (1981) indicate that students' perception of the reading process is strongly affected by instruction.In their study they claimed that the poor readers in their study did not have knowledge about appropriate comprehension strategies because the instruction they received emphasized oral reading and decoding, at the expense of comprehension.Along similar lines, Pintrich (2002, p. 221) stated that ''a teacher may encourage the use of certain strategies for reading.''Likewise, Sheorey and Mokhtari (2001, p. 433) point out that ''the reader's metacognitive knowledge about reading may be influenced by a number of factors, including previous experiences and culture-specific instructional practices.'' The findings also show that all four participating students from both cycles not only developed metacognitive awareness of the introduced reading strategies but also became aware of the knowledge they gained during the study.This indicates that the students began to develop metacognitive knowledge of the two components identified by the literature, namely, knowledge about cognition and awareness of one's own cognition (Harris et al., 2010).
With regard to Research Question 2, the findings reveal that the participating students from both cycles used the reading strategies they got introduced to them during the lessons.This was evident in the findings from the various data sets.For instance, the findings from the TAP show that in Cycle 1 all four participating students deployed the strategies of prediction and activating prior knowledge.In addition, two participating students used the strategy of identifying the main idea as well during the TAP.Students used these strategies within 2 months of instruction although none of them reported using them at the start of the study.Similarly, in Cycle 2 all four participating students made predictions in the second and third TAP.Moreover, two participating students also used the strategy of identifying the main idea in the third TAP.This indicates that the participating students in Cycle 2 also started using the strategies introduced within 2 months of the instructions, although during the first think aloud they only used one strategy, namely, activating prior knowledge.These findings are consistent with the existing research (e.g.Ali & Razali, 2019;Carrell, 1985;Carrell et al., 1989;Kern, 1989;Zhang, 2008) that shows that metacognitive reading strategy training brought about changes in ESL students' use of reading strategies in short term.
Overall, the findings related to change in students' metacognitive awareness and use of reading strategies support and reinforce the findings of the existing literature that indicate that strategies could be taught and students' awareness of strategies increases as a result of instruction (Aghaie & Zhang, 2012;Burchard & Swerdzewski, 2009;Chalmers & Fuller, 2009;Chularut & DeBacker, 2004;Huang & Newbern, 2012;Ko, 2007;Kutlut€ urkm 2020;Nunan, 1997;Mizumoto & Takeuchi, 2009;Tinnesz et al., 2006).Moreover, the findings support Auerbach and Paxton's study (1997) that indicates that metacognitive instruction can increase students' awareness and understanding of reading strategies.

Conclusion
The current study investigated if metacognitive reading strategies instruction could bring about any change in metacognitive awareness and use of reading strategies of university level ESL students in the Pakistani context.From the study, it emerged that metacognitive reading strategy instruction not only developed metacognitive awareness and use of the introduced reading strategies in students but also made them aware of and use the knowledge they gained during the study.Thus this study suggests that metacognitive reading strategies instruction could be realized in tertiary level Pakistani classrooms as this can aid in increasing metacognitive awareness and use of reading strategies in students.The result findings may offer some pedagogical implications for the ESL language instruction.For instance, language teachers should gain knowledge regarding teaching of reading strategies and should try to stimulate the students to reflect on their processes of reading so as to improve their metacognitive awareness and use of reading strategies.Moreover, teachers should reflect on their own teaching practices during metacognitive strategy instruction also by seeking students' perspective on their teaching.
For future research, it would be useful to study what help students decide what strategies to deploy during reading from a repertoire of available strategies.This will be a valuable contribution to the field since in the literature on metacognition and reading comprehension ''there is no clear understanding of if and how some strategies are chosen over others'' by good readers (Griffith & Ruan 2005: 15), let alone by students.

University of Warwick English Language Test (WELT)
Marking and Interpretation of WELT Scores (Revised Version 2005) 1.In papers G3, one mark is given for each correct item.The possible top mark for each paper is as follows: Paper G3 30 marks

Conversion to Grades
When each paper has been marked, the score should be converted to a grade, as follows: Paper G3: ACADEMIC READING

Interpretation of Scores
In broad terms, the Grades can be interpreted as follows: Grade A This indicates that the candidate reads English with very good general comprehension at a reasonable speed.Grade B This indicates that the candidate reads English well, with few misunderstandings.Given more time, further practice and support, he/she should soon become a very efficient reader.Grade C This indicates that the candidate is a reasonably competent reader of English, who should be able to cope in the university context, but who still reads rather slowly and with some misunderstanding in examination conditions.The candidate would be advised to read widely in English and work on increasing his/her vocabulary.Grade D This indicates that the candidate's misunderstanding of the texts is serious and that he/she is likely to be an extremely slow reader in English.The candidate needs further language study and more reading practice.Grade E This indicates that the candidate cannot read with any useful understanding in English.

Table 1 .
Comparison of the Number of Students in Cycle 1 and Cycle 2 Who Used the Strategies Before the Start of the Study and at the End of the Intervention (n = 56).