INTRODUCTION
Background to the study
Humanity is progressing towards a learning society where the significance of reading in enhancing knowledge is clear In the context of the fourth industrial revolution, strong reading skills are essential for learners to navigate the overwhelming flow of information and to inspire creativity for future innovations Research consistently shows that increased reading correlates with greater academic success Moreover, reading is recognized as a fundamental skill in language learning that all students must master The awareness of the critical role of proficient reading for academic achievement has been emphasized in recent studies.
The benefits of reading are obvious; however, getting learners to read is one of the biggest challenges of teaching, especially at tertiary level (Hatteberg & Steffy,
Many students engage in reading primarily to fulfill course assignments, leading to feelings of boredom and discouragement despite recognizing the importance of reading in their education While some learners excel in mastering reading skills, others struggle not only in school but also during additional reading activities This disparity highlights the challenges faced by students who find it difficult to develop effective reading skills.
2 found to be less engaged in school activities and lower motivation, which made them leave schools (Joseph & Schisler, as cited in Piercey, 2013)
Facing difficulties in comprehension, the ultimate goal of reading (Hyland
Factors such as vocabulary and working memory significantly influence reading comprehension, with the impact of working memory varying based on the type of tasks involved Moreover, to achieve proficiency in reading comprehension, learners must navigate challenges that extend beyond linguistic features.
Understanding novel syntactic constructions, discourse organization, and linguistic markers is crucial for developing reasoning skills that enable individuals to decipher unfamiliar words and conceptual constructs, analyze text structures, and recognize inter-textual references (Goldman et al., 2016) Guthrie and Wigfield (1999) highlighted that without awareness and active engagement with the text, including the effort to construct knowledge, comprehension is significantly diminished.
The lack of effective reading strategies can hinder learners' comprehension (Ouellettee & Beers, 2010) According to Anderson (1991), successful strategic reading involves knowing how to employ various strategies effectively and coordinating their use This view is supported by Garner (1994) and Paris et al (1994), who assert that learners must understand when, where, and how to apply reading strategies to achieve automaticity Research indicates that utilizing reading strategies enhances reading speed, increases enjoyment, and boosts motivation, leading to greater reader autonomy (Izquierdo Castillo & Jiménez Bonilla, 2014) The reading process is active and demands a range of skills (Wiener & Bazerman, 1999), with skilled readers benefiting from specific metacognitive strategies before, during, and after reading.
3 reading in order to understand materials and to build relationships between themselves and texts (Pressley & Afflerbach, 1995)
Skilled readers engage in deliberate actions such as careful planning, flexible tactics, and consistent self-monitoring, which enhance their comprehension (Paris & Jacobs, 1984) They actively reflect on the subject matter, scan the text in both directions, and evaluate their understanding In contrast, poor readers often fail to utilize these effective strategies By employing and coordinating various reading strategies efficiently, skilled readers significantly improve their cognitive processes during reading (Sheorey & Mokhtari, 2001).
Empirical studies have confirmed the significant role of metacognitive reading strategies in language acquisition Teng (2019) demonstrated that young learners who underwent metacognitive intervention developed a greater awareness of metacognitive knowledge, leading to improved reading comprehension compared to their peers who did not receive such intervention Additionally, these learners exhibited enhanced understanding of reading requirements and increased confidence in managing reading tasks This aligns with Huynh's (2021) findings on the advantages of teaching metacognitive strategies to adult learners Moreover, it is essential to identify factors that can enhance the application of these strategies, with self-efficacy emerging as a positive predictor for both general language learning and specific metacognitive reading strategies (Hayat et al., 2020; Jiang, 2021; Tembo & Ngwira, 2016).
Self-efficacy is the confidence individuals have in their ability to perform specific behaviors, which is crucial for achieving personal goals According to Bandura (1982), higher self-efficacy leads to greater effort and persistence when confronting challenges or adverse situations This concept underscores the importance of self-belief in overcoming obstacles and striving for success.
Understanding the impact of thoughts and emotional responses to one's environment is crucial for academic success (Bandura, 1982) Alongside skills and strategies, learners must cultivate a strong belief in their capabilities (Ahmadian & Pasand, 2017) Extensive research has explored the connection between self-efficacy and learning strategies, highlighting its significance in educational outcomes (Bonyadi et al., 2012; Montaủo-Gonzỏlez & Cancino, 2020; Nosratinia et al., 2014).
Reading self-efficacy refers to an individual's belief in their ability to engage in reading activities (Morali, 2019) Research indicates that implementing effective reading strategies can significantly enhance learners' self-efficacy (McCrudden et al., 2005; Raissi & Roustaei, 2013; Tavakoli & Koosha, 2016) Recent studies have increasingly focused on the connection between reading self-efficacy and the use of metacognitive reading strategies (Ahmadian & Pasand, 2017; Alsuhaibani, 2019; Okyar, 2021; Shehzad et al., 2020) Despite this growing interest, the role of reading self-efficacy as a predictor in its relationship with metacognitive reading strategies remains underexplored.
A study by Le (2021) is among the first in Vietnam to explore the relationship between reading self-efficacy and reading strategies, revealing significant associations However, the findings were broad, as the research encompassed all reading strategies categorized by Oxford, including memory, cognitive, compensation, metacognitive, affective, and social strategies, while overlooking specific types of metacognitive strategies Additionally, the study did not break down reading self-efficacy into distinct dimensions and focused solely on high school students, leaving the connection between reading self-efficacy and metacognitive reading strategies among Vietnamese EFL adult learners unexplored.
Jiang's (2021) research represents a pioneering effort to explore the relationship between reading self-efficacy and metacognitive reading strategies among Chinese EFL students in grades 7 and 8 The study revealed a positive correlation, indicating that students with higher reading self-efficacy are more likely to engage in metacognitive reading strategies, which in turn enhances their motivation to read.
The study highlights five distinct metacognitive reading strategies, revealing that individuals with higher self-efficacy are more inclined to engage with these strategies enthusiastically Conversely, those with lower self-efficacy tend to show less interest in employing metacognitive techniques However, Jiang's research did not fully clarify the connections between various dimensions of reading self-efficacy and the different types of metacognitive reading strategies This gap emphasizes the necessity for further investigation into these relationships among adult EFL learners, which the current study aims to address.
Aim of the study
This study aims to investigate the relationship between English as a Foreign Language (EFL) learners' reading self-efficacy and their use of metacognitive reading strategies, addressing the need for further research in this area.
Research question
In order to fulfill the aim stated in the Introduction, one major question was posed as follows:
Are all dimensions of reading self-efficacy positively related to all types of metacognitive reading strategies among EFL learners?
Significance of the study
Metacognitive reading strategies are essential for enhancing text comprehension and reading performance, yet their usage among Vietnamese university EFL students remains low This highlights the necessity to explore factors that could encourage the adoption of these strategies to boost academic performance The current study investigates the role of reading self-efficacy as a predictor of students' use of metacognitive reading strategies and their academic success If reading self-efficacy is validated as beneficial, educators can modify their teaching approaches to foster improvements in students' reading strategy utilization and overall academic achievement.
6 reading self-efficacy which in turn encourages the use of some or all types of metacognitive reading strategies among their students
Research on the relationship between reading self-efficacy and metacognitive reading strategies in Vietnam, particularly among tertiary students, is limited Jiang (2021) emphasized the need for further studies in higher education to explore this relationship The current study aims to fill this gap by examining how reading self-efficacy influences the use of metacognitive reading strategies, contrasting with previous research that focused on how these strategies affect reading self-efficacy By applying Bandura’s (1986) social cognitive theory, this study uniquely investigates the connections between different dimensions of reading self-efficacy and various metacognitive reading strategies, moving beyond the unidimensional approach taken by Jiang (2021).
Scope of the study
Self-efficacy and metacognitive strategies are crucial for language learners, particularly in reading This study focuses on the relationship between reading self-efficacy and metacognitive reading strategies, utilizing self-report questionnaires to gather data on learners' perceived reading behaviors However, the absence of actual reading tests or interviews may introduce common method bias Additionally, the research is limited to second-year non-English major students at a university in Southern Vietnam, which may impact the generalizability of the findings.
Organization of thesis chapters
The thesis will be organized as follows:
Chapter 1 provides an introduction to reading self-efficacy and metacognitive reading strategies Some theoretical aspects and empirical evidence are provided to identify research gaps that the thesis would fill out Also, the main aim of this study followed by one research question is presented to direct the paper The scope and contribution of the thesis are also mentioned in this section
Chapter 2 presents a review of literature related to reading self-efficacy and metacognitive reading strategies from which hypotheses were formulated to answer the research question The present study’s conceptual framework is also presented in this chapter
Chapter 3 describes the methodology of the study It begins with describing research design Then comes where and when the study was conducted, how the recruitment of the sample occurred, what background information represented the sample, what needed to ensure validity and reliability of the research instrument, and how data were analyzed
Chapter 4 presents how data screening process occurred, whether the internal consistency of research instrument was ensured and whether data were normally distributed This section also tests the reliability and validity of the framework to conclude the proposed hypotheses Essentially, results are discussed in light of the social cognitive theory by Bandura (1986) and in comparison to recent studies
Chapter 5 presents a summary of main findings and some pedagogical implications for EFL learners, teachers, and other stakeholders Limitations of the study and recommendations for further studies are also presented in this part
LITERATURE REVIEW
Reading self-efficacy
Reading is a dynamic process where individuals engage with written texts to comprehend and interpret the author's message (Williams, 1984; Grabe & Stoller, 2002) This interaction involves a two-way exchange of information between readers and authors (Brown & Briggs, 1989), as readers actively seek clues within the text to enhance their understanding (Peha, 2003) The subsequent discussion will explore the classification of reading types and their effects on reading comprehension.
Reading types and their impacts on reading comprehension
Reading can be categorized based on three key criteria Firstly, readers engage in academic reading or reading for entertainment, with research indicating that those who read for pleasure demonstrate greater improvements in reading comprehension compared to their academic counterparts Secondly, the manner in which readers approach texts can be distinguished between expeditious reading and careful reading.
Reading can be categorized into three main forms: search reading, which focuses on finding ideas about a specific topic; skimming, which allows for a quick overview of the text; and scanning, aimed at locating particular words or phrases (Khalifa & Weir, 2009) Careful reading is essential as it enables readers to absorb the majority of information presented in texts (Khalifa).
Research indicates that students who receive instruction in both expeditious reading and careful reading significantly improve their reading comprehension skills compared to those who do not participate in such lessons (Al-Qahtani & Lin, 2016).
Reading can be categorized into two modes: silent and aloud Research findings on the impact of these reading modes on comprehension have been inconsistent Some studies indicate no significant difference in comprehension between silent and aloud reading (Aisiah & Mahaputri, 2019; McCallum et al., 2004), while others suggest that silent reading leads to better understanding (Rupley et al., 2020) Conversely, certain authors argue that reading aloud enhances learners' ability to grasp information (Chen et al., 2019; Hale et al., 2007; Robinson et al., 2019).
Self-efficacy is a cognitive concept that encompasses individuals' judgments about their capabilities to perform specific behaviors, as well as their perceptions of the difficulties associated with those behaviors.
“individuals’ judgments of their abilities in organizing and conducting actions necessary to achieve a certain level of performance in an area” (Bandura, 1986, p
Psychologist Albert Bandura's definition of self-efficacy is widely recognized, describing it as an individual's belief in their ability to achieve desired outcomes through their actions This concept highlights the importance of self-belief in motivating individuals to pursue goals and overcome challenges.
Self-efficacy plays a crucial role in individuals' confidence to take necessary actions in future situations (Bandura, 1982) According to McCabe (2003), self-efficacy significantly influences motivation for specific behaviors Supporting this, Lin and Wang (2021) found that learners with higher creative self-efficacy experienced less anxiety when using virtual reality compared to those with lower self-efficacy To aid researchers in developing accurate self-efficacy measures, Bandura (2006) emphasized that assessment items should focus on “can do” rather than “will do,” as “can” reflects one's judgment of capability, while “will” pertains to intention.
Dimensions and sources of efficacy expectations
Self-efficacy theory posits that efficacy expectations are the beliefs individuals hold regarding their ability to successfully perform behaviors that lead to desired outcomes, encompassing three key dimensions Firstly, these expectations vary in magnitude, indicating the level of difficulty a person believes they can handle, from simple tasks to more complex challenges Secondly, efficacy expectations differ in generality, meaning that past experiences—whether successful or unsuccessful—can influence an individual's confidence in their abilities across different tasks Lastly, the strength of efficacy expectations can vary, with weak expectations often stemming from negative experiences, while strong expectations motivate individuals to persist in their efforts despite setbacks.
According to Bandura (1977a), there are four primary sources that shape personal efficacy expectations The most significant source is performance accomplishments, or "mastery experiences," as noted by Pajares (1997), which strongly influences self-efficacy expectations (Bandura, 1997; Usher & Pajares, 2008) Successes or failures in performing a behavior can significantly affect an individual's confidence in their future capabilities regarding that behavior (Bandura, 1977a) Additionally, emotional arousal serves as another internal source that can impact self-efficacy.
Emotions and physical sensations, such as anxiety and fatigue, significantly influence an individual's self-efficacy levels (Bandura, 1997; Usher & Pajares, 2008) For example, a learner's anxiety about reading aloud can gradually diminish her confidence in this skill over time.
Vicarious experience, a key source of self-efficacy, is influenced by one's surroundings and demonstrates that observing others successfully perform tasks can foster the belief that personal effort will lead to success and increased confidence Additionally, verbal persuasion from others, such as friends encouraging someone by sharing their own successful experiences, can significantly enhance an individual's confidence to undertake similar actions.
This article explores the intersection of reading and self-efficacy, presenting a comprehensive overview of reading self-efficacy It combines key characteristics of both concepts and introduces a scale designed to measure this important construct effectively.
As discussed earlier, self-efficacy is task- and domain-specific Hence, when it comes to the domain of reading, reading self-efficacy was defined as
Learners' perceptions of their reading abilities significantly influence their performance in various reading tasks (Li & Wang, 2010) Reading self-efficacy, as described by Shehzad et al (2020), refers to learners' confidence in their reading capabilities and their appraisals of their ability to complete specific reading tasks For instance, students majoring in social sciences may exhibit greater confidence in reading and discussing topics related to human rights compared to subjects like medicine or civil construction In this study, reading self-efficacy is defined as learners' evaluations of their skills in performing reading-related activities, ultimately aimed at enhancing their reading proficiency.
Roles of reading self-efficacy
Research indicates that reading self-efficacy plays a vital role in the success of learners as readers When students possess high reading self-efficacy, they are more confident in their ability to tackle challenging texts.
Metacognitive reading strategies
Language learning strategies are defined as conscious and purposeful actions taken by learners to enhance their acquisition and use of a second or foreign language According to Cohen (1998), these strategies involve deliberate choices aimed at strengthening learning processes Oxford (1989) describes them as steps that improve the acquisition, storage, retention, recall, and application of new language information Ellis (1994) emphasizes that these strategies encompass mental or behavioral activities linked to specific stages of language acquisition or use Similarly, Griffiths (2015) highlights that learners engage in actions—either deliberately or automatically—to facilitate their language learning and regulation.
Benefits of language learning strategies
Effective strategies are crucial for language learners as they promote active and self-directed engagement, enhancing communication skills (Oxford, 1990) By employing learning strategies, learners can improve their confidence and proficiency in language acquisition.
Learning strategies enhance the educational experience by making it easier, faster, and more enjoyable, while also promoting self-direction and effectiveness in various contexts (Oxford, 1990) They significantly influence learners' motivation and emotional states, as well as their approach to acquiring and organizing new knowledge (Weinstein & Mayer, 1986) By actively employing language learning strategies, students can take charge of their learning pace, develop their language abilities, and boost their confidence throughout the learning journey (Hong, 2017).
17 learners perform a variety of tasks from the very beginning of learning to the most advanced levels of proficiency in the target language (Cohen, 2011)
Oxford (1990) categorized language learning strategies into two main groups: direct strategies, which include memory, cognitive, and compensation techniques, and indirect strategies, encompassing metacognitive, affective, and social approaches Similarly, O’Malley and Chamot (1990) identified three types of language learning strategies: metacognitive, cognitive, and social/affective strategies They emphasized that metacognitive and cognitive strategies are closely linked to higher language proficiency and often work in tandem to enhance learning This article will delve deeper into metacognitive reading strategies, highlighting their significance in language acquisition.
Metacognition, a term introduced by Flavell in 1976, refers to the awareness and understanding of one’s own cognitive processes Researchers agree that it encompasses knowledge about cognitive states and abilities that can be shared among individuals, while also integrating affective and motivational aspects of thinking (Paris & Winograd, 1990) This cognitive self-regulation enables individuals to monitor tasks effectively and engage in reflective thinking, which combines various cognitive processes (Anderson, 2002) By reflecting on past experiences and feelings, metacognition facilitates critical evaluation of one’s thinking, potentially leading to significant changes in language acquisition strategies (Anderson, 2003).
18 recognize what she knows and does and what she does not know and does not do (Anderson, 2003)
Metacognitive strategies, extensively studied and applied in language learning over the past three decades, are essential for effective learning (Cer, 2019; de Boer et al., 2018; Mohseni et al., 2020) These strategies, rooted in metacognition, enable learners to go beyond basic cognitive tools, allowing them to adjust their learning processes (Oxford, 1990) They empower learners to manage and evaluate their cognitive activities, ensuring active and efficient learning (Wenden, 1991) Without these strategies, learners struggle to organize their studies, track progress, and assess their achievements and future goals (O’Malley & Chamot, 1990) Vandergrift (2002) emphasizes the importance of metacognitive strategies in guiding and directing language learning activities, as they encourage learners to reflect on their learning journey.
Frequent use of metacognitive strategies is linked to enhanced learning outcomes (Anam & Stracke, 2016) These strategies enable learners to evaluate their problem-solving efforts, plan future actions, monitor their success, and refine their learning techniques (Baker & Brown, 1984) Research indicates that learners actively reflect on their thinking during reading, demonstrating their ability to plan, monitor, evaluate, and apply information effectively (Wade et al., 1990) Without these metacognitive strategies, learners may lack direction and the ability to assess their progress and future goals (O’Malley & Chamot, 1990).
Distinction between metacognitive strategies and metacognitive knowledge
Metacognitive strategies and metacognitive knowledge are distinct yet interconnected elements of metacognition, with the former focusing on the regulation of learning through skills such as planning, monitoring, and evaluating, while the latter, or metacognitive awareness, encompasses the information learners gain about their learning processes Both components are essential in language learning, as they contribute to the development of strategic and self-regulated learners.
Oxford (1990) highlighted the importance of metacognitive strategies in empowering learners to take control of their cognitive processes These strategies include centering learning, which involves connecting prior knowledge with new information and prioritizing listening over speaking in both formal and informal contexts Additionally, learners can enhance their language acquisition by organizing their studies, setting clear goals, defining the purposes of language tasks, and actively seeking practice opportunities, such as creating a study schedule for English.
1990) Evaluating includes self-monitoring and self-evaluating such as thinking about one’s own progress in learning English (Oxford, 1990)
Metacognitive strategies, as defined by O’Malley and Chamot (1990), involve reflecting on one’s own thinking processes These strategies encourage learners to consider what they are learning, monitor their learning methods, and assess their overall learning experiences to improve comprehension.
1990) Initially, O’Malley and Chamot (1990) classified metacognitive strategies into four main groups as exhibited in Table 2.1
Table 2.1 Initially proposed types of metacognitive strategies by O’Malley and Chamot (1990)
Concentrating on in-advance decisions to pay attention to special aspects of particular tasks such as planning to look for keywords and phrases
Planning Planning for the organization of either written or spoken discourse
Monitoring Reviewing attention to tasks or monitoring comprehension of information needed to be remembered or monitoring production being taken place
Evaluating Checking comprehension after completing receptive language activities or evaluate language production after it occurs
O'Malley and Chamot (1990) refined their classification of planning strategies by incorporating selective attention and four additional sub-types, culminating in a finalized framework as shown in Table 2.2.
Table 2.2 Modified types of metacognitive strategies by O’Malley and Chamot (1990)
Preparing before reading by skimming materials for organizing principles, main ideas, and concepts that need to be learnt
Being ready to notice details of entire materials
Deciding to engage in general to learning tasks or to ignore irrelevant distractors
Planning for and rehearsing linguistic elements needed to conduct upcoming tasks
Understanding necessary conditions for learning and arranging for those conditions
Reviewing attention to tasks or monitoring comprehension of information needed to be remembered or monitoring production while being taken place
Checking comprehension after completing receptive language activities or evaluating language production after it occurs
This study utilizes the classification of metacognitive reading strategies proposed by O’Malley and Chamot (1990) due to its comprehensive nature and clarity regarding the roles of each strategy in the learning process Specifically, it focuses on advance organizer strategies and selective attention strategies as the primary planning strategies, which together represent nearly 50 percent of all metacognitive strategies, with advance organizers accounting for 21.4 percent and selective attention for 22.3 percent The remaining strategies encompass monitoring and evaluating strategies In total, this research examines four types of metacognitive strategies—advance organizers, selective attention, monitoring, and evaluating—within the context of reading The following sections will explore these metacognitive reading strategies and the assessment measures used to evaluate their application.
Metacognition, introduced by Brown (1980) in the context of reading, emphasizes that reading is a strategic process involving both knowledge and action Garner (1987) defines reading strategies as intentional activities employed by proactive learners to address cognitive challenges Afflerbach et al (2008) further elaborate that these strategies represent readers' conscious efforts to manage and adapt their reading processes to achieve specific goals.
Metacognitive reading strategies play a crucial role in helping learners decode text, understand words, and construct meaning These strategies are characterized by goal-driven behaviors and conscious awareness As noted by Varga (2017), actively selecting metacognitive reading strategies enables learners to observe, communicate, and deepen their understanding of their reading processes In this context, metacognitive reading strategies are defined as intentional activities that learners engage in to effectively manage their cognitive processes during reading tasks.
A selected measure of metacognitive reading strategies for the present study
The Metacognitive Awareness of Reading Strategies Inventory (MARSI), developed by Mokhtari and Reichard in 2002, is a widely recognized tool for assessing metacognitive awareness in reading However, it is not appropriate for the current study for several reasons Firstly, MARSI focuses on learners' awareness of metacognitive strategies rather than the strategies themselves, which is the primary focus here Secondly, the items in MARSI do not align with the four specific types of metacognitive reading strategies being investigated: advance organizer strategies, selective attention strategies, monitoring strategies, and evaluating strategies Additionally, MARSI was designed for native English speakers in grades 6 to 12, and a study by Guan et al (2011) indicated that its factor structure does not apply to proficient adult readers, making it unsuitable for EFL adult learners Lastly, MARSI primarily addresses reading related to academic materials, while this study encompasses a broader range of reading activities, including both academic and recreational reading.
Following O’Malley and Chamot (1990), a measure of metacognitive reading strategies was developed by Jiang (2021) encompassing four types First is advance organizer strategies which reflect preparation before reading by
Relationship between reading self-efficacy and metacognitive reading
This study aims to explore the relationship between reading self-efficacy and metacognitive reading strategies, grounded in behavioral theory Early 20th-century theories focused on human behavior as a response to environmental stimuli or an effort to manage desires and impulses (Piercey, 2013) Bandura (1997) emphasized the importance of understanding these dynamics in the context of self-efficacy and personal agency in learning processes.
Individuals actively shape their life experiences rather than merely reacting to external influences Various theories have been developed to understand behavior formation, including the theory of planned behavior, the transtheoretical model, and the behavior change wheel Among these, the social cognitive theory, rooted in self-efficacy, provides insights into the interplay between self-efficacy and behavior This study employs social cognitive theory to explore the connections between reading self-efficacy dimensions and different metacognitive reading strategies.
Social cognitive theory, as proposed by Bandura (1986), highlights the direct link between self-efficacy and behavioral changes, emphasizing that human behavior results from the reciprocal interactions among cognitive, environmental, and behavioral factors Cognitive factors encompass knowledge, expectations, and attitudes, while environmental factors involve social norms, community access, and peer influences Behavioral factors represent skills, practice, and self-efficacy, illustrating the complex interplay that shapes individual behavior.
Figure 2.1 The triangular relationship between the factors in social cognitive theory (Bandura, 1986)
According to Bandura (2000), self-efficacy is a central component of social cognitive theory, influencing the development of desired behaviors both directly and indirectly He demonstrated that self-efficacy interacts with outcome expectations, goal setting, and social-structural support, ultimately shaping behavior formation Supporting this theory, Guilloteaux (2007) found that learners with positive outcome expectations and an understanding of task values are more likely to engage in those tasks.
Figure 2.2 The structure of a causal model based on social cognitive theory (Bandura, 2000)
Bandura (2000) introduced two essential constructs—goals and social-structural factors—that are crucial for understanding the adoption of new behaviors (Lazaro, 2019) Within the framework of social cognitive theory, self-efficacy interacts reciprocally with these constructs, indicating that goal setting can also influence an individual's self-efficacy.
According to Lazaro (2020), insufficiently challenging goals can undermine self-efficacy, as highlighted by Bandura (1986) This relationship underscores the significance of goals in strategic planning and suggests that self-efficacy plays a crucial role in shaping behaviors.
Social-structural factors significantly impact individuals' self-efficacy, influencing their ability to confront challenges A strong belief in one's capabilities encourages persistent efforts to overcome obstacles (Lazaro, 2020), while self-doubt can lead to reduced effort and premature resignation (Bandura, 2000) Research demonstrates that self-efficacy affects various behaviors, including physical activity (Dutton et al., 2009), food choices (Luszczynska et al., 2007), technology adoption (Kulviwat et al., 2014), and learning strategies (Drysdale & McBeath, 2018; Tembo & Ngwira, 2016) Additionally, there is substantial empirical evidence supporting the connection between reading self-efficacy and the use of metacognitive reading strategies.
2.3.2 Empirical evidence for the relationship between reading self- efficacy and metacognitive reading strategies
Recent studies have shown a growing interest in the connection between reading self-efficacy and metacognitive reading strategies (Ahmadian & Pasand, 2017; Jiang, 2021; Le, 2021; Li & Wang, 2010; Naseri & Zaferanieh, 2012; Shehzad et al., 2020), consistently indicating significant relationships between these two variables However, differences exist between these studies and the current research, as highlighted by Alsuhaibani (2019) and Okyar (2021), which did not explore this relationship.
30 how each dimension of reading self-efficacy was related to each type of metacognitive reading strategies
In Vietnam, Le's (2021) mixed-methods research is pioneering in exploring the connection between reading self-efficacy and reading strategies among EFL senior high school learners The study found a significant relationship between reading self-efficacy and the utilization of reading strategies However, it is important to note that Le's research examined reading strategies in a general context, rather than focusing on specific metacognitive reading strategies such as advance organizers, selective attention, monitoring, and evaluating strategies, which are the focus of the present study Despite these differences, Le's findings underscore the existing link between reading self-efficacy and reading strategies.
Shehzad et al (2020) explored the connection between reading self-efficacy and metacognitive reading strategies, focusing on the mediating role of self-efficacy in the relationship between four sources of self-efficacy and metacognitive strategies among Saudi EFL tertiary students Their findings revealed significant correlations between the self-efficacy sources and reading self-efficacy beliefs, as well as between these beliefs and metacognitive reading strategies This highlights the importance of reading self-efficacy as a mediator in these associations Additionally, while Shehzad et al expanded their framework to include the four sources of reading self-efficacy, the current study aims to investigate the specific relationships between each dimension of reading self-efficacy and various types of metacognitive reading strategies.
The present study closely aligns with Jiang's (2021) research on Chinese EFL junior high school learners, particularly regarding dimensions of reading self-efficacy and metacognitive reading strategies Jiang's findings indicate that reading self-efficacy serves as a predictor for the utilization of metacognitive reading strategies.
Jiang's study highlights a connection between reading self-efficacy and metacognitive reading strategies among learners, yet it fails to explore the specific relationships between each dimension of reading self-efficacy and various types of metacognitive strategies Furthermore, the research does not utilize Bandura's social cognitive theory to analyze and clarify the interplay between these two constructs.
Recent studies have suggested a connection between self-efficacy and reading strategies, yet the specific relationships between dimensions of reading self-efficacy and various metacognitive reading strategies remain underexplored, particularly within the framework of Bandura’s (1986) social cognitive theory Notably, Jiang (2021) emphasized the need to investigate these relationships among tertiary students to enhance generalizability, a topic that has been largely overlooked in the Vietnamese context This gap motivated the current research, which shifts the focus from the impact of reading strategies on self-efficacy to the role of reading self-efficacy in predicting the use of metacognitive reading strategies.
Table 2.3 Review of previous studies
Study Research issue Methodology Participant Result
The relationship between the use of reading strategies and reading self- efficacy
Reading strategy questionnaire adapted from Mokhtari and Sheorey
Reading self-efficacy questionnaire: 7 items selected from Rahimi and Atiyah (2009), Sun (Sun,
2010), Ghonsooly and Majid (2011); and 8 newly- developed items
191 female Saudi EFL university freshmen
Learners’ use of reading strategies and sense of reading self-efficacy were at moderate levels
A positive association existed between reading strategy and reading self-efficacy
The factors affecting learners’ reading self-efficacy included teachers' teaching methodology, encouragement, and feedback, learners’ exam grades and their ability of reading comprehension
Reading strategy use and its relation to reading self-efficacy and gender
Questionnaire on reading strategy use adopted from Iyitoglu’s (2011) Turkish
211 male Turkish EFL university students of different academic majors
Learners' use of reading strategies was at a moderate level
Study Research issue Methodology Participant Result version translated from Mokhtari and Sheorey (2002)
Reading self-efficacy adapted from Yanar and Bumen (2012)
Female learners used reading strategies more frequently than their male counterparts
Learners’ use of reading strategies was positively related to reading self- efficacy
Le (2021) The relationship between reading self-efficacy and reading strategies
A questionnaire including reading self-efficacy edited from Tobing (2013) and reading strategies modified from Barón (2013) and Shang (2010)
300 Vietnamese EFL students in grades 10, 11, and
Learners’ level of reading self-efficacy was moderate
The main sources of reading self- efficacy were performance accomplishment and vicarious experience
Reading self-efficacy had a considerable relationship with reading strategy use
The association between four self- efficacy sources and metacognitive reading strategies using reading self-
Questionnaire on sources of reading self-efficacy adapted from Usher and Pajares
188 Saudi EFL learners from five public universities
Four self-efficacy sources were significantly related to reading self- efficacy
Study Research issue Methodology Participant Result efficacy beliefs as a mediator
Questionnaire on reading self-efficacy modified from Tobing (2013)
Questionnaire on metacognitive reading strategies adopted from Mokhtari and Sheorey
Reading self-efficacy was significantly associated with metacognitive reading strategies
Reading self-efficacy mediated the association between four self-efficacy sources and metacognitive reading strategies
Jiang (2021) The relationship between reading self-efficacy and the use of metacognitive reading strategies
Reading self-efficacy questionnaire adapted from Piercey (2013)
Metacognitive reading strategies questionnaire constructed based on O’Malley and Chamot
100 Chinese EFL learners in grades
Learners felt rather confident in their English reading capabilities
Learners used metacognitive reading strategies quite frequently
Reading self-efficacy was an influential factor that affected the learners’ use of metacognitive reading strategies
Conceptual framework
This study explores the relationship between dimensions of reading self-efficacy and metacognitive reading strategies, grounded in Bandura's social cognitive theory (1986) Recent research has highlighted a connection between these two constructs, prompting an investigation into how four dimensions of reading self-efficacy correlate with four types of metacognitive reading strategies By extending Bandura's framework, the study aims to provide insights into the interplay between self-efficacy in reading and the strategic approaches employed by readers.
Hence, the following hypotheses were formulated:
Hypothesis 1: Reading skill self-efficacy is positively related to advance organizer strategies (H1a), selective attention strategies (H1b), monitoring strategies (H1c), and evaluating strategies (H1d)
Hypothesis 2: Reading test self-efficacy is positively related to advance organizer strategies (H2a), selective attention strategies (H2b), monitoring strategies (H2c), and evaluating strategies (H2d)
Hypothesis 3: Reading self-efficacy for regulation is positively related to advance organizer strategies (H3a), selective attention strategies (H3b), monitoring strategies (H3c), and evaluating strategies (H3d)
Hypothesis 4: General reading self-efficacy is positively related to advance organizer strategies (H4a), selective attention strategies (H4b), monitoring strategies (H4c), and evaluating strategies (H4d)
The study's conceptual framework, illustrated in Figure 2.3, identifies reading self-efficacy as independent variables, while various types of metacognitive reading strategies serve as dependent variables.
Figure 2.3 Conceptual framework of the present study
METHODOLOGY
Research question
In order to pursue the research aim set in the Introduction, the study addresses the following question:
Are all dimensions of reading self-efficacy positively related to all types of metacognitive reading strategies among EFL learners?
Research design
To address the research question, a quantitative approach was employed, allowing for findings to be generalized across a population (Creswell, 2012) A cross-sectional survey design was implemented to collect data on various variables at a single point in time from a sufficiently large sample Statistical analysis was conducted to describe trends and measure the associations between these variables (Creswell, 2012) This methodology facilitated predictions regarding whether reading self-efficacy influenced the use of metacognitive reading strategies Furthermore, quantitative research enabled the interpretation of results by comparing them with prior predictions and existing studies (Creswell, 2012).
This study aligns with recent research that utilizes a quantitative approach to explore the relationship between self-efficacy beliefs and metacognitive strategies, as evidenced by the works of Alsuhaibani (2019), Okyar (2021), and Shehzad et al (2020).
Research site
Vietnam has become a focal point for English language learning due to significant reforms initiated in 1986 and its accession to the World Trade Organization in 2007, which heightened the demand for English proficiency among professionals and students alike As a result, numerous English schools and centers have emerged to cater to this growing need The 2018 General Education Program emphasizes the importance of English, designating it as a compulsory subject from grade 3 to 12, while also requiring non-English majors in higher education to meet foreign language criteria for graduation Although various studies have explored metacognitive reading strategies and their correlation with reading self-efficacy, there remains a gap in understanding the specific relationship between these two factors, highlighting the necessity for further research in this area.
This study focused on Ho Chi Minh City, a vibrant metropolis attracting individuals from various social and educational backgrounds seeking employment and educational opportunities The high demand for English learning in this region of Vietnam significantly aided the researcher in recruiting participants The findings of this research are expected to enhance English teaching and learning in this dynamic urban environment The study was conducted at a university in Ho Chi Minh City, which offers bachelor programs across both social and scientific disciplines.
39 which gave the researcher a chance to approach participants from different fields of study to extend the generalizability of the present study
Jiang (2021) highlighted the importance of metacognitive reading strategies and reading self-efficacy in university students, particularly those who predominantly select English as their foreign language At this university, students must complete three mandatory English courses or present an English language certificate equivalent to at least level 3 of the Vietnamese Standardized Test of English Proficiency (Ministry of Education and Training, 2015) Therefore, it is crucial for students to understand and develop metacognitive reading strategies to enhance their reading efficiency and fulfill academic requirements.
Sampling and sampling procedures
This study focused on second-year non-English major students for three key reasons Firstly, freshmen were excluded due to their transitional phase from high school to higher education, which often leads to differing expectations and learning styles that have not been adequately addressed (Zimmerman, 1998) Additionally, first-year students may struggle to identify issues in their learning strategies, as they likely lacked explicit training and opportunities to implement these strategies during high school (McDaniel et al., 2021; Rachal et al., 2007; Ryan).
High school students typically engage in passive learning during class, while in higher education, they are expected to take on self-regulation and self-assessment of their studies (Glenn, 2004; Nordell).
In 2009, it was noted that reading strategies were often overlooked in general education, leading high school students to rely on dictionaries for unfamiliar words (Phan, 2014) Consequently, many learners entered university without a clear understanding of metacognitive reading strategies (Huynh, 2021) As a result, students typically become acclimated to the university environment by their second year, allowing them to adapt their learning approaches effectively.
This article outlines 40 strategies, focusing specifically on reading strategies, to help students effectively regulate their reading practices By understanding and applying these strategies, students can respond more accurately to research instruments.
In the research site, second-year students opting to complete three English courses instead of submitting a required English certificate likely aimed to finish these courses alongside their peers, avoiding delays until their third or fourth years This decision, unless influenced by unforeseen circumstances like registration failures or personal issues, highlights the importance of metacognitive reading strategies in enhancing English reading performance during their second year Consequently, the researcher focused exclusively on second-year students, aligning with previous studies on metacognitive reading strategies.
The researcher faced challenges in approaching target participants due to not being a lecturer at the research site To overcome this, she leveraged existing relationships to connect with five Deans from the faculties of Physics, Literature, Chemistry, History, and Information Technology This approach aligns with previous studies, such as Okyar (2021) Once approval was obtained from the faculties, the researcher collaborated with administrative staff to gather schedules for second-year students during the 2021-2022 academic year Utilizing convenience sampling, a non-probability method, allowed for participant engagement despite potential representation issues This sampling method is favored in educational research because it captures participant characteristics, such as gender and field of study, that can reflect the broader population (McMillan & Schumacher, 2001; Mesgar et al., 2012) Additionally, it ensures that participants are willing and available for the study (Cresswell, 2012, p.67).
Participant involvement is crucial for research success The use of convenience sampling enables researchers to gather a larger number of responses quickly and cost-effectively (Creswell, 2012) Furthermore, easy accessibility of participants during data collection is essential for effective research outcomes (Etikan et al., 2016).
For the main study, a second-year class initially selected for piloting the research instrument was excluded from data collection The sample size was determined using the N:q ratio, where N represents the number of participants and q denotes the number of questionnaire items, following the 5:1 ratio recommended by Suhr (2006) Given that the questionnaire includes 45 items across two main variables, the minimum sample size required is 225 students To ensure a robust data collection, the researcher planned to gather responses from additional classes to exceed this minimum and account for potential invalid responses.
Research instrument
This study employed a quantitative approach, necessitating the involvement of a significant number of participants for generalizability (Creswell, 2012) To effectively address the research question, a questionnaire survey was utilized, allowing for rapid engagement with a large participant pool and facilitating the collection of substantial data (Mackey & Gass).
The questionnaire developed in 2005 comprises three main sections, with Part A focusing on demographic profiles by including two items that inquire about students' genders and faculties The choice of these control variables aligns with previous research on reading self-efficacy and metacognitive reading strategies, as evidenced by studies such as Bećirovic et al (2017), Mesgar et al (2012), and Shehzad et al (2019) Participants are given options for gender (Male/Female) and faculty (History/Literature/Physics/Chemistry).
Information Technology) were provided in advance The participants were required to tick only one option for each item
Part B of the study consists of 25 items designed to measure reading self-efficacy among participants, categorized into four dimensions The first dimension, reading skill self-efficacy, includes six items that assess students' confidence in performing specific reading tasks The second dimension, reading test self-efficacy, comprises three items that reflect students' beliefs in their ability to achieve good reading performance The third dimension, reading self-efficacy for regulation, consists of five items that evaluate students' confidence in managing their reading processes Lastly, general reading self-efficacy, which includes 11 items, addresses students' overall reading abilities, with two items reversely coded to ensure respondent attention (Schmitt & Stults, 1985) All items are rated on a five-point Likert scale, where participants indicate their confidence level from 1 ("strongly disagree") to 5 ("strongly agree") Mean scores are interpreted as low (less than 3.39), moderate (3.40 to 3.79), or high (greater than 3.80) (Pihie & Akmaliah, 2009).
Part C of the study consists of 20 items designed to assess participants' frequency of using various metacognitive strategies during English reading These strategies are categorized into four types: Advance Organizer strategies, which include four items focused on learners' preparation before reading; Selective Attention strategies, comprising seven items that reflect learners' focus on specific aspects of the text; Monitoring strategies, with four items addressing the real-time oversight of reading; and Evaluating strategies, which include five items that assess learners' reflections on their reading performance in relation to their pre-set goals and identify any issues encountered Participants rated each item using a five-point Likert scale, ranging from 1 ("I never or almost never do this") to 5 ("I always or almost always do this"), selecting the option that best represented their views The study is grounded in three levels of strategy use as proposed in the literature.
Mokhtari and Reichard (2002) and Oxford (1990) to interpret the data results, namely, “low” (mean scores at 2.4 and less), “medium” (mean scores from 2.5 to 3.4), and “high” (mean scores at 3.5 and above)
Piloting is essential for validating and ensuring the reliability of research instruments (Srinivasan & Lohith, 2017) To achieve content validity and reliability, this study's questionnaire underwent two rounds of piloting, which included expert interviews and feedback from a class of second-year EFL students (Naz et al., 2021).
Content validity of the questionnaire
Expert interviews were conducted to validate the appropriateness of items used to assess conceptual representations of reading self-efficacy and metacognitive reading strategies (MacKenzie et al., 2011) Five bilingual academics, including two Ph.D holders and three master's graduates in Teaching English to Speakers of Other Languages, participated in the study All experts had extensive experience, ranging from 18 to 35 years, in teaching English, particularly reading skills to tertiary students During the interviews, each item was read aloud, prompting discussions on its relevance and suitability for the study's context, as well as its ability to accurately reflect the dimensions of reading self-efficacy and metacognitive strategies.
Experts recommend modifying item 5 from Jiang's (2021) measure of general reading self-efficacy, while keeping the other 44 items intact The original content of item 5 states, “Seeing adults who are good at English reading motivates me to be a better reader.” However, since the current study involves tertiary students who are adults, experts suggest replacing the term “adults” with “others” to better align with the participants' age group.
The original measures of reading self-efficacy and metacognitive reading strategies were translated from English to Chinese by Jiang (2021) to ensure participant comprehension and enhance response validity (Griffee, 2001; Tsang et al., 2017) During this study, there was no Vietnamese version of the questionnaire available Consequently, after selecting the relevant items, the questionnaire was translated into Vietnamese, followed by a back-translation procedure as recommended by Brislin (1980) For instance, the monitoring strategy item “While reading, I will check whether I understand what I have read” was translated into Vietnamese as “Trong khi đọc, tôi sẽ kiểm tra xem liệu tôi có hiểu điều tôi đang đọc hay không.” However, a typing error was identified during back-translation, where “đang đọc” should have been “đã đọc,” and the phrase “hay không” was deemed redundant The final revised Vietnamese version of this item is: “Trong khi đọc, tôi sẽ kiểm tra xem liệu tôi có hiểu điều tôi đã đọc.” (Refer to Appendix 2 for the complete revision of the questionnaire items).
The researcher administered the final Vietnamese version of the questionnaire to a class of 34 second-year students to identify any necessary amendments in word usage Students were invited to provide feedback directly on the questionnaire or seek clarification from the researcher for any ambiguous items After excluding participants who did not select English as their foreign language or who opted out of the study, the results were compiled.
A total of 30 responses were collected from participants, as detailed in Table 3.1, which presents descriptive statistics The students took approximately 13 minutes to complete the printed questionnaires, and all distributed questionnaires were returned without any concerns raised, either in writing or verbally To ensure clarity, the researcher verbally confirmed with the learners about the questionnaire items, and all participants expressed that the questions were sufficiently clear.
Table 3.1 Descriptive statistics of the participants in piloting
The researcher meticulously reviewed student responses, identifying invalid entries based on two negatively coded items and unmarked statements, ultimately confirming that there were no missing data in the 30 responses collected All students correctly answered the attention check items related to general reading self-efficacy, leading to the acceptance of all responses for reliability testing Data were analyzed using SPSS version 22, with coding for gender as male (1) and female (2), while all participants from the faculty of History were uniformly coded as 1 Responses to reading self-efficacy items were assigned values from 1 (Strongly Disagree) to 5 (Strongly Agree), and similar coding was applied to metacognitive reading strategy items, ranging from "I never or almost never do this" to "I sometimes do this."
“I usually do this.”, and “I always or almost always do this.”, were coded as 1, 2, 3,
4, and 5, respectively Items 1 and 2 in general reading self-efficacy were reversedly scored by SPSS 22
To ensure questionnaire reliability, Cronbach’s Alpha coefficients should exceed 7, with Corrected Total-Item Correlation coefficients greater than 3 (Hair et al., 2018; Nunnally, 1978) The results indicated that the Cronbach’s Alpha values for the four dimensions of reading self-efficacy—reading skill self-efficacy (.864), reading test self-efficacy (.807), reading self-efficacy for regulation (.877), and general reading self-efficacy (.848)—demonstrated strong reliability.
The reliability of the reading strategies questionnaire, which included 46 strategies categorized as advance organizer, selective attention, monitoring, and evaluating, was confirmed with coefficients of 854, 953, 888, and 896, respectively Additionally, the Corrected Total-Item Correlation coefficients for all 45 items exceeded 3, indicating strong reliability (refer to Appendix 3 for detailed results) These findings validated the questionnaire's effectiveness in measuring reading self-efficacy and metacognitive reading strategies among Vietnamese tertiary students, allowing for its distribution to the sample A summary of the questionnaire items is presented in Table 3.2, with English and Vietnamese versions available in Appendices 4 and 5.
Table 3.2 Summary of questionnaire items
Part B: English reading self-efficacy
Reading skill self-efficacy 6 items Ordinal scale
Reading test self-efficacy 3 items
Reading self-efficacy for regulation 5 items
11 items (The first two items were reversely coded.)
Advance organizer strategies 4 items Ordinal scale
Data collection procedure
In late April 2022, the researcher reached out to administrative staff to obtain timetables in preparation for participant recruitment for both the pilot and main study, with data collection scheduled for May.
Before the Covid-19 pandemic, data collection typically occurred at the end of the semester, when students were preparing for final exams through self-study However, due to the pandemic's impact, some courses during the 2021-2022 academic year required an additional five to seven weeks to complete, as indicated by the updated timetables To enhance the chances of effectively reaching the target students, the researcher selected classes that were scheduled for the extended seven-week period.
In the first week of May 2022, the researcher conducted a pilot study by distributing questionnaires in person during an offline class, as the university was offering both online and offline courses Following the pilot, data collection for the main study was scheduled for the second and third weeks of May 2022 To accommodate online classes, a Google Form version of the questionnaire was created Questionnaires were distributed online to participants in the Physics and Information Technology faculties, while printed versions were provided to those in Chemistry Additionally, a combination of both printed and online questionnaires was utilized for the Literature and History faculties This article provides a detailed overview of the data collection activities for both the pilot and main study.
The researcher conducted the study in offline classrooms and was granted access to online classes by the responsible teachers To minimize the Hawthorne effect, participants were approached during break time without their teachers present Students had the option to select from various university-provided languages for credit, and those who did not choose English were excluded from the study For those who opted for English, the researcher clarified that participation was voluntary and assured them of the confidentiality of the collected data The questionnaire was designed to be anonymous, ensuring that no personal information, such as names, emails, or student codes, was gathered.
Data analysis scheme for the main study
Data collection utilized both Google Forms and printed questionnaires, with raw data from Google Forms exported to an Excel file, while printed responses were manually entered Each item was coded consistently with the pilot phase, categorizing faculties as History (1), Literature (2), Physics (3), Chemistry (4), and Information Technology (5) History and Literature were combined into a social category (coded as 1), while the remaining faculties were classified as science (coded as 2) The complete dataset was then imported into SPSS 22 for statistical analysis.
Data screening was initially performed to remove univariate outliers, followed by internal consistency tests where Cronbach’s Alpha values above 7 and Corrected Item-Total Correlation values above 3 were deemed acceptable (Hair et al., 2018; Nunnally, 1978) Subsequently, the mean, standard deviations, skewness, and kurtosis coefficients of all items were analyzed to assess normality, with cut-off values for skewness and kurtosis set between -2 and +2, and -7 and +7, respectively (Hair et al., 2018) Pearson correlations were then conducted on the dimensions of reading self-efficacy and metacognitive reading strategies, examining the absolute correlation coefficients among the four dimensions of reading self-efficacy for potential multicollinearity, with values below 8 indicating low risk (Senaviratna & Cooray, 2019) To further verify the absence of multicollinearity, Tolerance and Variance Inflation Factor values were evaluated in SPSS, ensuring they met the criteria of greater than 2 and less than 10, respectively (Field, 2009).
After multicollinearity was confirmed not to be a big issue, factor analyses, including both exploratory and confirmatory types, were run to confirm whether
The study focused on the relationship between reading self-efficacy and metacognitive reading strategies, classifying items within these dimensions In accordance with exploratory factor analysis (EFA) guidelines, reading self-efficacy was treated as an independent variable while metacognitive reading strategies were viewed as a dependent variable Items that passed reliability tests were analyzed separately for EFA, adhering to the recommendations of Hair et al (2010) against mixing dependent and independent variables in a single analysis Hair et al (2015) further supported this approach, affirming that EFA should be conducted on independent or dependent variables separately to ensure accurate results.
In the exploratory factor analysis (EFA), the principal components method with varimax rotation was utilized, setting the factor loading threshold at 5 to confirm the statistical significance of the observed variables (Hair et al., 2018) Additionally, the Kaiser-Meyer-Olkin (KMO) measure ranged from 5 to 1.0, indicating adequate sampling for the analysis (Kaiser).
To determine the appropriateness of factor analysis, cut-off values were established, including a significance level of less than 05 for Bartlett’s test of sphericity (Hair et al., 2018) Additionally, eigenvalues exceeding 1 and a cumulative variance greater than 50% indicate a robust factor structure (Hair et al., 2018) It is crucial that all items align correctly within the predefined dimensions of reading self-efficacy and various metacognitive reading strategies.
EFA results were cross-validated using confirmatory factor analysis (CFA) This step was performed on Analysis of Moment Structures version 24 (AMOS
The study examines model fit indices and construct validity, emphasizing the importance of standardized regression weight coefficients exceeding 5 (Tabachnick & Fidell, 2001) Key fit indices utilized include the chi-square divided by degrees of freedom (χ2/df), Tucker-Lewis fit index (TLI), incremental fit index (IFI), comparative fit index (CFI), standardized root mean square residual (SRMR), and root mean square error of approximation (RMSEA) The established cutoff value for χ2/df is less than 2 (Schermelleh-Engel et al.).
2003), TLI and IFI and CFI > 90 (Tabachnick & Fidell, 2001), SRMR < 08, RMSEA< 06 (Hu & Bentler, 1999) Regarding construct validity, it is indicated
Convergent validity is established when construct reliabilities exceed 7 and the average variance extracted is greater than 4, as noted by Bagozzi & Yi (1988) and Brunelle & Lapierre (2007) Additionally, constructs demonstrate discriminant validity if their average variance extracted values surpass the maximum shared variance and if the square root of the average variance extracted is greater than their correlations with other constructs, following the criteria set by Fornell & Larcker (1981).
In this study, a comprehensive structural equation modeling (SEM) approach was utilized to assess all items retained from the confirmatory factor analysis (CFA) phase SEM, as defined by Hoyle (1995), is a robust statistical method for testing hypotheses regarding relationships among observed and latent variables This method allows researchers to control for measurement errors and analyze interdependencies, providing insights into the covariance patterns among observed variables and maximizing the explanation of their variance within the proposed model (Kline, 1998) The covariance-based structural equation modeling (CB-SEM) was performed using AMOS 24, with model fit indices evaluated against the CFA phase's established cut-off values to ensure adequate goodness of fit Significant associations between dimensions of reading self-efficacy and types of metacognitive reading strategies were indicated by probability values below 05 To validate the CB-SEM findings, bootstrap sampling with 500 replications was employed; if the 95% confidence interval values contained zero, it suggested no associations between the variables (Efron & Tibshirani, 1993; Streukens & Leroi-werelds, 2016; Xu, 2016).
A one-way analysis of variance (one-way ANOVA) was conducted using SPSS 22 to investigate the differences in reading self-efficacy and metacognitive reading strategies based on control variables such as gender and fields of study (Armstrong et al., 2000) If Levene’s test shows a probability greater than 05, data transformation is unnecessary due to homogeneity Furthermore, a smaller probability value of the F-ratio indicates significant differences among the groups.
If the probability value of the F-ratio is less than 05, it indicates that control variables significantly influence the main variables Conversely, a probability value greater than 05 suggests no impact from the control variables To analyze differences among students from various faculties, post-hoc tests were conducted The criteria for confirming these differences include a Levene test probability greater than 05, an F-ratio probability less than 05, and a multiple comparison probability also less than 05 (Armstrong et al., 2000).
RESULTS AND DISCUSSION
Descriptive statistics
The study initially aimed to gather a minimum of 225 responses, accounting for potential invalid entries, but ultimately collected 335 responses After screening the data using SPSS 22, 67 responses were excluded, resulting in a final valid sample of 268, exceeding the required minimum Among these, 131 responses (48.9%) were from male participants and 137 (51.1%) from female participants The five selected faculties were categorized into two fields of study: social and science.
Table 4.1 Distribution of genders with regard to faculties
Table 4.2 Distribution of genders with regard to fields of study
Male Female Fields of study
Reliability of the questionnaire
The internal consistency of the questionnaire was assessed using the remaining 268 responses, revealing that all dimensions of reading self-efficacy and metacognitive reading strategies had Cronbach’s Alpha values exceeding 7 However, certain items showed Corrected Item-Total Correlations below 3, specifically item 1 in reading skill self-efficacy, and items 3, 8, and 9 in general reading self-efficacy, as well as items 1 and 4 in selective attention strategies Following the removal of these items, a subsequent reliability test indicated that the dimensions of reading skill self-efficacy and selective attention strategies met the reliability criteria Notably, item 10 in general reading self-efficacy had a Corrected Item-Total Correlation of 431 in the initial assessment.
> 3 was down to 277 < 3 in the second round (see Appendix 7), which required the researcher to remove item 10 in general reading self-efficacy and conduct the
The third round of testing demonstrated a Cronbach’s Alpha of 919 for general reading self-efficacy, exceeding the acceptable threshold of 7 Additionally, all items exhibited Corrected Item-Total Correlations greater than 3, confirming the reliability of this dimension for subsequent statistical analyses.
In addition, all the items that were kept in the reliability tests had absolute values of skewness and kurtosis smaller than 2.0 and 7.0 respectively (see Appendix
The data confirmed a normal distribution, as outlined by Hair et al (2018) Additional insights into the convergent and discriminant validity of the questionnaire will be discussed in the factor analysis sub-section.
Testing multicollinearity
The analysis presented in Table 4.3 reveals that the correlation coefficients among the four dimensions of reading self-efficacy are all below 8, indicating no multicollinearity (Senaviratna & Cooray, 2019) Furthermore, as shown in Table 4.4, all Tolerance values exceed 2 and VIF values are below 10, further confirming the absence of multicollinearity (Field, 2009).
Table 4.3 Correlations between dimensions of reading self-efficacy and types of metacognitive reading strategies
RSE_sk RSE_te RSE_re RSE_ge AOS SAS MS ES
Note * Correlation is significant at the 05 level (2-tailed)
** Correlation is significant at the 01 level (2-tailed)
RSE_sk = Reading skill self-efficacy; RSE_te = Reading test self-efficacy
RSE_re = Reading self-efficacy for regulation; RSE_ge = General reading self-efficacy
AOS = Advance organizer strategies; SAS = Selective attention strategies
MS = Monitoring strategies; ES = Evaluating strategies
Reading self-efficacy for regulation 927 1.079
Dependent Variable: Advance organizer strategies.
Testing the convergence of questionnaire items
An exploratory factor analysis (EFA) was conducted on 20 items measuring reading self-efficacy, yielding a KMO value of 836, which indicates the suitability of factor analysis Bartlett’s test of sphericity, with a p-value less than 05, confirmed significant relationships among the items Four factors with eigenvalues greater than 1 were extracted, accounting for a cumulative variance of 66.313%, demonstrating a robust structure Notably, all 20 items aligned correctly with the four dimensions outlined in the conceptual framework.
Table 4.5 KMO and Bartlett's test of reading self-efficacy
Kaiser-Meyer-Olkin Measure of Sampling Adequacy .837 Bartlett's Test of
Table 4.6 Total variance explained of reading self-efficacy
Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings Total
Extraction Method: Principal Component Analysis
Table 4.7 Rotated component matrix a of reading self-efficacy
Extraction Method: Principal Component Analysis
Rotation Method: Varimax with Kaiser Normalization a Rotation converged in 5 iterations
Note RSE_sk = Reading skill self-efficacy
RSE_te = Reading test self-efficacy
RSE_re = Reading self-efficacy for regulation
RSE_ge = General reading self-efficacy
An exploratory factor analysis (EFA) was conducted on 18 items assessing metacognitive reading strategies The analysis demonstrated a high appropriateness for factor analysis, evidenced by a KMO value of 859 Additionally, Bartlett’s test of sphericity confirmed significant relationships among the items, supporting the validity of the analysis.
In a study by Hair et al (2018), four factors with eigenvalues exceeding 1 were identified and rotated, resulting in a cumulative variance of 59.356%, which demonstrates a strong structural validity Notably, all 18 items successfully converged, reinforcing the reliability of the findings (Hair et al., 2018).
59 correctly in the pre-determined types of metacognitive reading strategies (see Table 4.10)
Table 4.8 KMO and Bartlett's test of metacognitive reading strategies
Kaiser-Meyer-Olkin Measure of Sampling Adequacy .859 Bartlett's Test of
Table 4.9 Total variance explained of metacognitive reading strategies
Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings Total
Extraction Method: Principal Component Analysis
Table 4.10 Rotated component matrix a of metacognitive reading strategies
Extraction Method: Principal Component Analysis
Rotation Method: Varimax with Kaiser Normalization a Rotation converged in 5 iterations
Note AOS = Advance organizer strategies; SAS = Selective attention strategies
MS = Monitoring strategies; ES = Evaluating strategies
All items confirmed in EFA were then imported into AMOS 24 for CFA The examination of fit indices showed satisfying results (χ2/df=1.518 < 2, TLI
= 923 > 90, IFI = 931 > 90, and CFI = 930 > 90, SRMR = 0545 < 08, RMSEA
In the analysis, two items were identified with standardized regression weight coefficients below the acceptable threshold of 5, specifically item 3 related to reading skill self-efficacy at 46 and item 5 concerning evaluating strategies at 44 Consequently, these items were removed from the study, and the remaining items were re-evaluated for further scrutiny.
2009) As shown in Figure 4.1, after the removal of those two items, the standardized regression weight coefficients of the remaining items were greater
62 than 5 and fit indices indicated a good fit between the hypothesized model and the data (χ2/df = 1.516 < 2, TLI = 929 > 90, IFI = 937 > 90, CFI = 936 > 90, SRMR
= 0544 < 08, RMSEA = 044 < 06) (Hu & Bentler, 1999; Schermelleh-Engel et al., 2003; Tabachnick & Fidell, 2001)
Note RSE_sk = Reading skill self-efficacy; RSE_te = Reading test self-efficacy
RSE_re = Reading self-efficacy for regulation
RSE_ge = General reading self-efficacy
AOS = Advance organizer strategies; SAS = Selective attention strategies
MS = Monitoring strategies; ES = Evaluating strategies
Figure 4.1 Convergence of questionnaire items
The assessment of construct validity included both convergent and discriminant validity, as detailed in Table 4.11 The composite reliabilities ranged from 767 for reading skill self-efficacy to 920 for general reading self-efficacy, exceeding the recommended threshold of 7 (Bagozzi & Yi, 1988) Additionally, the average variance extracted varied from 428 for selective attention strategies to 768 for reading test self-efficacy, indicating acceptable convergent validity (Brunelle & Lapierre, 2007; Menguc & Auh, 2006; Zhou et al., 2005) Furthermore, all eight constructs demonstrated discriminant validity, as their average variance extracted values were higher than the maximum shared variance, and the square root of the average variance extracted surpassed their correlations with other constructs (Fornell & Larcker, 1981).
Table 4.11 Composite reliability, average variance extracted, and correlation matrix
CR AVE MSV RSE_ge RSE_re SAS MS ES RSE_sk RSE_te AOS
CR = Composite reliability; AVE = Average variance extracted; MSV = Maximum Shared Variance
RSE_sk = Reading skill self-efficacy; RSE_te = Reading test self-efficacy
RSE_re = Reading self-efficacy for regulation; RSE_ge = General reading self-efficacy
AOS = Advance organizer strategies; SAS = Selective attention strategies
MS = Monitoring strategies; ES = Evaluating strategies
Hypotheses testing
Following the CFA results that confirmed an acceptable model, CB-SEM was utilized to evaluate the study's hypotheses, as illustrated in Figure 4.2 The model demonstrated a satisfactory fit to the data, with a χ2/df ratio of 1.742, which is below the threshold of 2 (Schermelleh-Engel et al., 2003), and goodness-of-fit indices showing IFI at 908 and CFI at 907, both exceeding the 90 benchmark (Tabachnick).
The study's findings indicate a tolerable degree of misfit, with an RMSEA of 053, which is below the 06 threshold established by Hu & Bentler (1999) Although the TLI value of 898 is slightly below the acceptable 90 benchmark (Kim et al., 2016; Naeghel et al., 2012), and the SRMR at 0864 exceeds the 08 criterion (Schmidt & Retelsdorf, 2016), these results support the conclusion of significant relationships among the variables.
Hypothesis 1: Reading skill self-efficacy is positively related to advance organizer strategies (H1a), selective attention strategies (H1b), monitoring strategies (H1c), and evaluating strategies (H1d)
The analysis of the relationships between reading self-efficacy dimensions and metacognitive reading strategies using CB-SEM revealed that reading skill self-efficacy showed insignificant correlations with all four metacognitive strategies: advance organizer strategies (β = -.126, p = 096), selective attention strategies (β = -.068, p = 402), monitoring strategies (β = 069, p = 359), and evaluating strategies (β = -.005, p = 950) Bootstrap sampling further confirmed these findings, with confidence intervals containing zero for all strategies, leading to the rejection of hypothesis 1.
Note RSE_sk = Reading skill self-efficacy; RSE_te = Reading test self-efficacy; RSE_re = Reading self-efficacy for regulation
RSE_ge = General reading self-efficacy; AOS = Advance organizer strategies; SAS = Selective attention strategies
MS = Monitoring strategies; ES = Evaluating strategies
Figure 4.2 CB-SEM results between dimensions of reading self-efficacy and types of metacognitive reading strategies
Students may exhibit confidence in their reading skills primarily due to their strong vocabulary, which is essential for comprehension A broad vocabulary enables better understanding of texts, as evidenced by the questionnaire items in the study that focus on vocabulary-related aspects, such as deciphering word meanings and identifying key information The limited use of metacognitive reading strategies among students may stem from a lack of awareness of these strategies rather than a deficiency in reading or vocabulary confidence Consequently, it is unpredictable that those with high self-efficacy in reading skills would frequently employ metacognitive strategies, resulting in insignificant correlations between these variables.
In this hypothesis, metacognitive reading strategies are influenced indirectly by reading skill self-efficacy through goals and social-structural factors, as illustrated in Bandura's social cognitive theory (2000) Research by Shih and Huang (2019) supports that students' application of metacognitive strategies is shaped by their learning outcomes and peer interactions Learning outcomes serve as motivational goals, prompting learners to aim for specific scores in quizzes, assignments, and reading tests For example, students strategically allocate time to research, comprehend, and utilize information effectively to produce high-quality assignments Additionally, these outcomes reflect learners' aspirations for self-fulfillment in English proficiency, such as achieving necessary levels for future graduate studies.
Table 4.12 Results of the relationships between dimensions of reading self-efficacy and metacognitive reading strategies
Monitoring strategies 136 048 136 [-.022, 293] 107 H2d Reading test self-efficacy
H3a Reading self-efficacy for regulation
H3b Reading self-efficacy for regulation
H3c Reading self-efficacy for regulation
H3d Reading self-efficacy for regulation
Note CB-SEM = Structural equation modeling; N = Number of bootstrap samples; β = Standardized coefficient
P = Probability value; CI = Confidence interval; ***: p < 001
Hypothesis 2: Reading test self-efficacy is positively related to advance organizer strategies (H2a), selective attention strategies (H2b), monitoring strategies (H2c), and evaluating strategies (H2d)
CB-SEM path coefficients, as presented in Table 4.12, indicated that reading test self-efficacy was not associated with advance organizer strategies (β
The analysis revealed that various strategies were evaluated, with significant findings indicating that reading test self-efficacy influences the use of monitoring strategies The bootstrap method, which included 500 replications, showed confidence intervals (CIs) ranging from -.069 to 234, -.145 to 201, and -.044 to 274, all containing zero, suggesting non-significant results (p-values of 251, 849, and 152 respectively) These findings confirm the initial results regarding the impact of self-efficacy on strategy use.
The analysis yielded a p-value of 048, indicating a borderline significance in the prediction of monitoring strategy use by reading test self-efficacy However, bootstrap results confirmed that no significant relationship exists between these variables, as evidenced by the 95% confidence interval.
IC ranged from -.022 to 293, containing 0 (p = 107 > 05) Eventually, hypothesis
The study revealed that reading test self-efficacy does not predict the use of metacognitive reading strategies, suggesting that the low application of these strategies may stem from a lack of training rather than low self-efficacy Students might also experience stress during tests, which could hinder their ability to utilize these strategies effectively Additionally, this finding may be influenced by the students' academic goals; if these goals are not sufficiently challenging, motivation to engage in strategic behaviors may diminish It is important to note that English was not the students' major, potentially contributing to a mindset of complacency regarding their performance To better understand these dynamics, it is essential to reference social cognitive theory, which posits that behavior formation is influenced by both self-efficacy and goal-setting.
71 with borderline scores needed to pass their English courses rather than trying metacognitive reading strategies to improve their reading performances
Hypothesis 3: Reading self-efficacy for regulation is positively related to advance organizer strategies (H3a), selective attention strategies (H3b), monitoring strategies (H3c), and evaluating strategies (H3d)
The CB-SEM path coefficients revealed that reading self-efficacy for regulation has significant positive associations with all four types of metacognitive reading strategies Specifically, advance organizer strategies showed a strong correlation (β = 417, p < 001), while selective attention strategies also demonstrated a notable relationship (β = 251, p < 001).
= 002 < 05) (H3b), monitoring strategies (β = 260, p < 001) (H3c), and evaluating strategies (β = 383, p =