View PDF VersionPrevious ArticleNext Article
 
DOI: 10.1039/D1RP00342A (Paper) Chem. Educ. Res. Pract., 2022, 23, 408-421

The effect of metacognition on students’ chemistry identity: the chain mediating role of chemistry learning burnout and chemistry learning flow

Xipei Guo a, Wenbo Deng ab, Kaifu Hu ab, Weina Lei ab, Shuoqi Xiang a and Weiping Hu *ac
aMOE Key Laboratory of Modern Teaching Technology, Shaanxi Normal University, Xi’an, China. E-mail: weipinghu@163.com
bSchool of Education, Shaanxi Normal University, Xi’an, Shaanxi, China
cShaanxi Normal University Branch, Collaborative Innovation Center of Assessment toward Basic Education Quality at Beijing Normal University, China

Received 20th December 2021 , Accepted 17th January 2022

First published on 25th January 2022

Abstract

With the urgent goal of increasing student retention within science, technology, engineering, and mathematics (STEM) fields, STEM identity is highlighted as a powerful source of student persistence. Since chemistry is an important part of the STEM discipline, a growing body of research has focused on chemistry identity. However, we currently know very little about how to improve students’ chemistry identity. Therefore, the present study aimed to explore the mechanisms of metacognition, learning burnout, and learning flow in identity in the context of chemistry, further providing suggestions for the advancement of students’ chemistry identity. Based on previous studies, the current study hypothesized that chemistry learning burnout and flow would play a chain mediating role in the relationship between metacognition and chemistry identity. A sample of 594 tenth-grade students completed questionnaires for the assessment of the four main variables in this study. The results showed that (1) metacognition, chemistry learning burnout, and chemistry learning flow significantly predicted students’ chemistry identity after the effect of gender was controlled; (2) both chemistry learning burnout and chemistry learning flow played separate mediating roles in the relationship between metacognition and chemistry identity; and (3) the chain mediating effect of metacognition → chemistry learning burnout → chemistry learning flow → chemistry identity was significant. These findings imply that embedded metacognitive prompts, decreased learning burnout, and increased flow experience are vastly helpful in developing learners’ chemistry identity. Finally, we further highlight the educational implications of the findings of this study and propose lines of future research.

Introduction

In science education, there has been increased concern regarding science identity development. Science identity is seen as an important construct that positively influences learners’ academic decisions, academic achievement, and persistence in the major (Perez et al., 2014), and it is regarded as a critical outcome of experiential science, technology, engineering, and mathematics (STEM) programs (Huvard et al., 2020). Because chemistry is well accepted as a compulsory introductory subject within STEM discipline fields (Gonzalez and Paoloni, 2015), researchers have particularly highlighted chemistry identity development in recent decades. While the growth of identity is identified as essential for encouraging students’ chemistry learning (Skagen et al., 2018), current studies on chemistry identity focus principally on the development of measures of chemistry identity (Hosbein and Barbera, 2020b). Not much is known about how to develop students’ chemistry identity. For this reason, the present study aims to explore the underlying influencing factors influencing chemistry identity in high school students, providing further suggestions for the advancement of students’ chemistry identity.

According to relevant research, metacognition is an internal driver of students’ learning identity formation. Thus, we are interested in examining the relationship between metacognition and chemistry identity. Since learning burnout and flow have been reported to be both affected by metacognition and to influence science identity, this report explores whether learning burnout and learning flow play mediating roles in the association between metacognition and chemistry identity in the chemistry context.

Metacognition and chemistry identity

Identity is composed of self-view developed by an individual in the process of participating in an activity and an individual's recognition and classification of their role in a specific community or professional field (Vincent-Ruz and Schunn, 2018). Accordingly, chemistry identity refers to “recognized as a chemistry person” (Hosbein and Barbera, 2020a).

Identity status theory, which first theorizes identity formation as a measurable model (i.e., Marcia's identity model), is widely considered to provide a solid theoretical foundation for empirical studies of identity (Seaman et al., 2017). Exploration and commitment are defined as the core dimensions of identity formation in Marcia's identity model (Marcia and James, 1966). Exploration is the extent to which individuals discover multiple options for an identity element, as well as actively explore and evaluate the outcomes of different options (Luyckx et al., 2009). Commitment refers to making relatively firm choices regarding identity elements (e.g., their education and occupation), and using choices as a guide to meaningfully participate in relevant activities to consolidate their choices (Marcia and James, 1966). According to different degrees of exploration and commitment, individual development on a certain element of identity can be further divided into four kinds: diffusion, foreclosure, moratorium, and achievement (Seaman et al., 2017). Identity diffusion is the lowest level of identity state and refers to individuals who have neither experienced identity exploration nor made identity commitment. Identity achievement is the best state of identity development, representing identity exploration and commitment.

Building on Marcia's theory, Berzonsky's identity style model further explores the cognitive process of forming different identity states, as well as the different information processing methods used to deal with identity conflicts and problems from the perspective of social cognitive processing (Berzonsky and Kuk, 2000). Berzonsky emphasized that identity formation is a process of individual self-construction and reconstruction, that involves seeking and evaluating self-related information, formulating self-constructs (such as identity standards, goals and reference values), and continuously monitoring, questioning, evaluating and revising self-constructs (Berzonsky, 2008). According to different cognitive processing methods, Berzonsky divided identity formation into different types (i.e., identity processing style) (Seaman et al., 2017).

The development of metacognition is a necessary condition for accomplishing cognitive processing of identity (Welsh and Schmitt-Wilson, 2013). Metacognition, as a complex psychological construct, refers to “thinking about thinking” (Flavell, 1979; Lavi et al., 2019). On the one hand, metacognition represents the individual's knowledge and ability to manage information, plan, monitor, and evaluation cognitive processes, as well as reflect to regulate and modify their cognitive representations (Öz, 2016; Goupil and Kouider, 2019). Such metacognitive knowledge and abilities related to executive function are also necessary for individuals to form, monitor, evaluate, and revise their self-constructs in order to develop a stable identity (Phillips, 2008; Welsh and Schmitt-Wilson, 2013; Fleur et al., 2021). On the other hand, metacognition enables students to actively perceive their interests and talents, thus helping them to make identity decisions (Dignath and Büttner, 2008). Moreover, building on Marcia's and Berzonsky's theory, Welsh and Schmitt-Wilson (2013) investigated whether students’ metacognition could affect their identity status. The results indicated that metacognition was crucial to completing the exploration and commitment process in the development of identity. Specifically, lower levels of metacognition can cause lower identity achievement and increased identity diffusion (Welsh and Schmitt-Wilson, 2013).

The potential mediating effect of chemistry learning burnout

Learning burnout refers to an emotionally and physically exhausted state caused by overtaxing schoolwork that results in students feeling negative, incompetent, cynical, and detached regarding learning and having ongoing low academic self-efficacy (Schaufeli et al., 2002; Ling et al., 2014; Yang and Chen, 2015).

As argued by Semerari et al. (2012), emotional understanding and cognitive attribution are essential components of the metacognitive system, which includes identifying and controlling inner emotions, adjusting one's emotions to positively deal with challenges, and reasonably attributing learning results (Schraw, 1998). Naturally, the use of metacognitive strategies is conducive to regulating and managing emotion in the learning process to reduce burnout levels (Seibert et al., 2017; Pennequin et al., 2020). In addition, it has been demonstrated that the improvement of metacognitive awareness can lead to a significant reduction in a negative state of mind (Matthews et al., 1999; Al-Baddareen et al., 2015; Saricam, 2015; Öz, 2016), including learning burnout. Therefore, metacognitive awareness is a negative predictor of learning burnout (Saricam et al., 2017).

Regarding the relationship between learning burnout and chemistry identity, burnout has a negative impact on the formation process of students’ identity. As Berzonsky (2011) outlines, the identity formation process requires students to actively invest cognitive resources in exploration (e.g., processing identity-related information, resolving conflicts between different identities, identity reflection and revision, etc.). Students with high learning burnout cannot feel the pleasure and value of learning, so they minimize the cognitive resources invested in the identity exploration process and fail to form identity achievements (Wang et al., 2015). Several studies have proven that individual burnout in a specific field is negatively related to identity. For example, Erentaite et al. (2018) investigated how school experiences (including school engagement, and school burnout) affect students’ identity formation over time. The findings showed that students with high levels of school burnout were more likely to not actively explore in the identity formation process or to simply follow the advice of authority figures to make identity commitments. In contrast, students with high school engagement tended to be proactive in identity exploration, while school burnout could decrease school involvement (Erentaite et al., 2018). In addition, it has been proven that a higher level of burnout is related to a decline in academic well-being, interest, and engagement in learning (Wang et al., 2015; Rehman et al., 2020), whereas these effects have benefits for identity development (Trujillo and Tanner, 2014).

The potential mediating effect of chemistry learning flow

Learning flow is characterized as a cognitive-affective state experienced in the learning process (Waterman et al., 2003; Bose, 2008). Specifically, while experiencing this psychological state, students are completely involved in an interest-oriented activity or some activity accompanied by enjoyment and pleasant feelings (Csikszentmihalyi et al., 2005). Three basic preconditions for achieving flow have been identified: clear goals, immediate feedback, and a balance between challenges and skills (Csikszentmihalyi, 1996; Bachen et al., 2016). To achieve flow, students must be allowed to set learning goals, choose cognitive strategies, make plans, select learning strategies, actively monitor cognitive processes, and continuously adjust and coordinate cognitive processes immediately based on feedback information in metacognitive processing (Schraw and Dennison, 1994; Gonzalez and Paoloni, 2015). It follows that metacognitive awareness is highly likely to be a necessary precondition to students’ learning flow experience. It has been demonstrated that metacognition has a positive effect on students’ learning flow (Han and Kim, 2017).

As to the link between learning flow and identity, it could be supported by eudaimonic identity theory (EIT). Based on Marcia's identity model, Waterman (1990) further proposed EIT, which tried to determine how people make the better choice among many identity choices from the perspective of eudaimonistic philosophy and existential philosophy. According to EIT, what is important for the identity development is that individuals could identify their potential and value in a domain-specific area and are willing to put in the effort to realize their self-value (Waterman, 2004). The subjective experience resulting from intrinsically motivated activities is closely related to the degree of self-value identification and realization. Flow, on the other hand, is identified as a subjective experience that arises from intrinsically motivated stimulation activities (Waterman, 2011). In brief, flow experience, as one of the subjective experiences included in identity-related activities (Coatsworth et al., 2006; Waterman, 2011), could predict the development of identity (Mao et al., 2016). In addition, previous studies have confirmed that learning flow can improve students’ academic achievement (Joo et al., 2012) and that academic achievement is an important indicator for predicting students’ identity development (Pop et al., 2016).

The mediating effect of chemistry learning burnout and chemistry learning flow

Furthermore, we need to clarify the relationship between burnout and flow. According to Wu et al. (2021), positive affect (i.e., the frequency of experiencing positive emotions) could activate students’ thinking, trigger creative and exploratory ideas, further give action to ideas, and thus could provide resources for students learning flow. In other words, positive affect and positive emotions are prerequisites for students to experience learning flow. It follows that chemistry learning burnout as a negative emotion may have a negative impact on learning flow.

In addition, flow theory suggests that people can experience flow only when they are intrinsically motivated (Rodríguez-Ardura and Meseguer-Artola, 2017). Therefore, learning flow is also seen as a psychological state experienced by students when they are engaged in activities that are intrinsically motivated (Kowal and Fortier, 2000; Waterman, 2011). However, previous studies have demonstrated that burnout significantly and negatively predicts intrinsic motivation (Ghanizadeh and Jahedizadeh, 2017; Rehman et al., 2020). Given that increased burnout reduces students’ self-efficacy and intrinsic motivation, while self-efficacy and intrinsic motivation can facilitate students’ learning flow experiences, we can infer that the higher chemistry learning burnout suffered, the lower the possibility that they experienced chemistry learning flow.

The present study

The above literature review shows that metacognition will affect the cognitive process of identity formation and ultimately affect the result of identity status. While identity is domain-specific (e.g., some students have higher chemistry identity but lower language identity), no studies were located that examined the relationship between metacognition and chemistry identity (Berzonsky and Kuk, 2000). Therefore, the first aim of the present study is to explore the relationship between metacognition and the development of students’ chemistry identity in high school.

Further, given that the enhancement in metacognitive awareness can reduce students’ learning burnout, the reduction in learning burnout is conducive to promoting students’ identity formation processes, it could be hypothesized that metacognition can influence students’ learning identity through learning burnout. Since both burnout and identity are domain-specific, the present study hypothesizes that the impact of metacognition on chemistry identity is mediated by chemistry learning burnout.

In addition, because metacognition could promote learning flow, which is one of the important conditions for developing identity, we assume that chemistry learning flow also plays a mediating role in the association between metacognition and chemistry identity. Taking it one step further, since both learning burnout and learning immersion may mediate the effects of metacognition on chemistry identity, we assume that metacognition influences chemistry identity through the chain mediating effect of chemistry learning burnout and chemistry learning flow.

Taken together, the hypotheses of the present study are as follows:

Hypothesis 1: There is a positive relationship between perceptions of metacognition and chemistry identity.

Hypothesis 2: Chemistry learning burnout has a mediating effect between metacognition and chemistry identity.

Hypothesis 3: Chemistry learning flow acts as a mediator between metacognition and chemistry identity.

Hypothesis 4: Chemistry learning burnout and chemistry learning flow play a chain mediating role in the associations between metacognition and chemistry identity.

In sum, the present study first analyzed relevant theoretical and empirical studies to develop possible hypotheses for the relationship between metacognition, chemistry learning burnout, chemistry learning flow, and chemistry identity. Subsequently, we would select the appropriate measurement instruments and adapt some of them to fit the research topic of this study. Given the adaptation of some of the instruments, exploratory factor analysis (EFA) and confirmatory factor analysis (CFA), and the reliability by Cronbach's Alpha (α) were used to confirm the reliability of the adapted instruments. Immediately afterward, serial multiple mediation analyses were conducted to verify the hypotheses proposed. Finally, the findings were combined with the initial theoretical analysis to provide feasible suggestions for enhancing the chemistry identity of high school students.

Method

Participants

The present study selected 594 students (47% males; 53% females) in Grade 10 from six high schools in China by the stratified purposive sampling method. Firstly, according to the economic development situation, we selected two cities from China's first-tier cities and second-tier cities respectively, and one city from third-tier cities and fourth-tier cities respectively, making a total of six cities (first-tier cities have the best economic development) (Xu et al., 2021). Secondly, we selected an ordinary high school in each of the six cities, divided the students into three levels the high, medium, and low based on their academic performance, as well as selected about 30 students from each level. The age range of the participants was 15 to 17 years old and the average age of the participants was 16.5 years old.

All participants have completed six years of elementary school science courses (which includes chemistry) and have experienced two years of chemistry subject matter courses. As a result, they already have some understanding of the subject of chemistry. And they are about to decide whether or not to continue with a chemistry major course in Grade 11. Therefore, chemistry identity development is crucial for Grade 10 students in China (i.e., the participants of this study).

Survey administration

The survey was completed in computer class through an electronic questionnaire. As the survey was implemented in six different schools, it took a total of one week to collect the questionnaires from all six schools. Permission has been obtained from all appropriate authorities and students’ parents before proceeding with the data gathered. In addition, this study gained ethical approval from the Ethics Committee of the Academic Committee of the Ministry of Education of Key Laboratory of Modern Teaching Technology, Shaanxi Normal University in China. Before the investigation began, the researchers explained to all participants the purpose of the study and the use of the data, as well as promised to keep participants’ personal information strictly confidential. All the students volunteered to participate in the study with no additional reward.

Survey instruments

The original measurement tools used for metacognition and chemistry identity were in English. To avoid measurement errors caused by students’ inability to understand English, the original English tools were translated in this study (the specific original text and translation are included in the introduction of each measurement tool). There are three translators, all of whom have doctorate degrees in education from China. Their native language is Chinese, and they are proficient in English. All three have in-depth knowledge of several variables and pedagogy-related content in this study. Each of the three translators translated the English version of the measurement tool into Chinese, then compared and discussed the final version of the translation together. Then, we used translation software to translate Chinese back to English. The aim was to make sure the meaning had not changed. Next, five students in Grade 10 were approached to read and comment on the questionnaire to ensure that the students could understand correctly what was expressed in the items.
Measures of metacognition. The Junior Metacognitive Awareness Inventory (Jr. MAI) developed by Sperling et al. (2002) was used to measure students’ metacognition. The Jr. MAI includes two subscales: metacognitive knowledge (i.e., awareness and understanding of cognitive processes) with 9 items and metacognitive regulation (i.e., abilities to monitor and control cognitive processes) with 9 items (Ning, 2017). All items were measured on a 5-point Likert scale (1 = Never, image file: d1rp00342a-u1.tif; 2 = Seldom, image file: d1rp00342a-u2.tif; 3 = Sometimes, image file: d1rp00342a-u3.tif; 4 = Often, image file: d1rp00342a-u4.tif; 5 = Always, —image file: d1rp00342a-u5.tif). In the present study, because the test objects are Chinese students, we translated the original English scale into Chinese. The original text, translation (in Chinese), and back translation information of the two items are presented in Table 1 as examples.
Table 1 Chinese and English versions of example items for metacognition
Item1 Item 2
Original text I think of several ways to solve a problem and then choose the best one I draw pictures or diagrams to help me understand while learning
Translation (in Chinese) image file: d1rp00342a-u6.tif image file: d1rp00342a-u7.tif
Back translation Facing a chemical question, I think about several different solutions, and then choose the best method In the process of learning chemistry, I draw pictures or diagrams to help me understand

Measures of chemistry identity. The measurement instrument for chemistry identity was adapted based on Student Science Identity (SSI) questionnaire and Engineering Identity Measures (EIM) (Godwin, 2016; Chen and Wei, 2020). The SSI is designed to measure secondary school students’ science identity (chemistry is included in science), close to the research content of the present study. However, the measurement structure and structure connotation of SSI are not fully consistent with the present study, so some items need to be deleted and corrected.

Four dimensions were included in the SSI questionnaire: performance, competence, recognition, and interest. This is also the initial identity constructs as defined by Hazari et al. (2010). However, Hazari et al. (2010) suggested that competence and performance beliefs should belong to the same dimension after investigation. Other studies also have suggested that competence/performance beliefs, interest, recognition work together for domain-specific identity shaping (Hosbein and Barbera, 2020b; Verdín, 2021). Based on the above, the measurement of chemistry identity in the present study is according to the three-factor structure. Competence/performance beliefs mean students’ beliefs in how well they can perform during completing the tasks and mastering the knowledge. The recognition measures an individual's perception of whether oneself or others consider them as a chemistry person. “Interest” is defined as an interest in understanding and learning about a subject, as well as engaging in relevant inquiry activities (Hazari et al., 2010), whereas the SSI measures interest more broadly than Hazari's defined. Interest in the SSI encompasses career interests and choices (Chen and Wei, 2020).

Combining the two points above, we need to select appropriate items in the SSI to integrate the competence and performance dimensions and focus on the constructs connotations of chemistry identity followed in this study. The measurement constructs and constructs connotations of EIM are consistent with this study and have been applied to other studies (Godwin, 2016; Verdín, 2021). Therefore, we selected 14 items in the SSI for testing chemistry identity according to Godwin's EIM with 11 items. All items on chemistry identity were presented in a Likert-type scale ranging from 1 (Strongly disagree, image file: d1rp00342a-u8.tif) to 5 (Strongly agree, image file: d1rp00342a-u9.tif). Two items were selected as examples presented in Table 2, including original text, translation (in Chinese), and back translation information.

Table 2 Chinese and English versions of example items for chemistry identity
Item1 Item 2
Original text I can understand scientific laws and principles well I will learn more about science knowledge through a variety of sources
Translation (in Chinese) image file: d1rp00342a-u10.tif image file: d1rp00342a-u11.tif
Back translation I can understand the laws and principles contained in chemistry very well I am willing to learn more chemistry knowledge through various channels

Measures of chemistry learning burnout. The Maslach Burnout Inventory-Student Survey (MBI-SS) is the most widely used measure instrument for learning burnout, which includes three sub-dimensions: exhaustion, cynicism, low sense of achievement (Li et al., 2021). Exhaustion implies that students feel frustrated and uncomfortable with learning. Cynicism refers to an indifferent, distant attitude toward learning. Low achievement characterizes the degree to which students feel disappointed, dissatisfied, or meaningless about their learning (Luo et al., 2016).

Based on the MBI-SS, Xing and Chen (2010) developed a learning burnout questionnaire suitable taking into account the burnout characteristics of Chinese students. Although Xing and Chen (2010) added a dimension of interpersonal alienation based on the three dimensions of the MBI-SS, the dimension describes students’ feelings about learning as a whole (e.g., I had almost no good friends in the class) and thus cannot be focused within the chemistry discipline. Therefore, we selected appropriate items (14 in total) related to the three dimensions of the MBI-SS from Xing and Chen's research tools to test chemistry learning burnout. The rating of this scale ranged from not meet at all agree (image file: d1rp00342a-u12.tif) to full compliance (image file: d1rp00342a-u13.tif). In addition, items about the low sense of achievement in the questionnaire were reverse-scored (e.g., item 1). We converted the scores of these items before calculating the total scores, so higher scores reflected greater burnout. Next, some examples of the scales were displayed (translated from Chinese):

Item 1: “image file: d1rp00342a-u14.tif, image file: d1rp00342a-u15.tif, image file: d1rp00342a-u16.tif”.(The more difficult the chemical problem is, the more interested I am in it and the harder I will try to accomplish it).

Item 2: image file: d1rp00342a-u17.tif. (I rarely listen carefully in chemistry class).

Measures of chemistry learning flow. The chemistry learning flow was measured using the Scale for Adolescents’ Flow State in Learning (SAFSL), which was a self-report instrument developed by Lei et al. (2012) in Chinese aimed at domain-specific learning activities. According to nine characteristics of flow pointed out by Csikszentmihalyi, combined with Flow State Scale (FSS) and Dispositional Flow Scale (DFS) scale, the SAFSL consists of four dimensions: “clear goals of learning” means that students have clear goals in the learning process and know how well they are performing; “concentration on the task and enjoying” means that the student feels pleasure and satisfaction in the learning process; “loss of self–consciousness” refers to a temporary loss of self-awareness; and “distortion of time perception” refers to the temporary distortion of time perception (Lei et al., 2012; Wu et al., 2021). Each item of the SAFSL was rated on a 5-point scale (1 = Never experienced, image file: d1rp00342a-u18.tif; 5 = Always experienced, image file: d1rp00342a-u19.tif). Example items from SAFSL were as follows (translated from Chinese):

Item 1: “image file: d1rp00342a-u20.tif, image file: d1rp00342a-u21.tif”. (In learning chemistry, I don’t care what others will think of me).

Item 2: “image file: d1rp00342a-u22.tif, image file: d1rp00342a-u23.tif”. (When I was engaged in learning chemistry, I felt that time was passing at a different pace than usual).

Data analysis procedures and tools

Specifically, SPSS 21.0 and AMOS 23.0 were used to conduct statistical analyses in this study (Browne and Cudeck, 1993; Hair, 2006; Hayes, 2012). Research procedures are as follows:

(1) Some variables involved in the education research field are not directly measurable (e.g., identity and burnout in this study) and are referred to as latent variables or factors. In research, latent variables (factors) are usually measured using instruments such as Likert scales, and the items (or indicators) in the scale are referred to as observable variables. When using such an instrument, it is critical to test structural validity of the measurement instruments using an exploratory factor analysis (EFA) and confirmatory factor analysis (CFA), as well as the reliability by Cronbach's Alpha (α).

Notably, EFA has various factor extraction for analyses, including Principal Components Analysis (PCA), Principal Axis Factoring (PAF), Maximum Likelihood (ML), Unweighted Least Squares, and similar (Koyuncu and Kılıç, 2019). Among them, PAF and ML are considered to be the two methods that give the best results. ML is mainly applied to normally distributed data, and PAF is better when the data distribution deviates from normal (Costello and Osborne, 2005). After extracting the factors, a suitable factor rotation method needs to be selected to specify the loading of each item in each factor. Oblique rotation is usually used when factors are correlated with each other, and orthogonal when the opposite is true. Oblique rotations include Oblimin and Promax, the latter being more suitable for larger and more complex data (Koyuncu and Kılıç, 2019). Based on the above, EFA was conducted using maximum likelihood with Oblimin rotation in the present study.

Further, we randomly divided all samples into two parts. EFA was first administered to Sample 1 (n = 297) to test whether the factor structure of each measurement instrument was consistent with the original predefined. Next, we implemented a CFA on Sample 2 (n = 297). The aim was to examine whether the correspondence between the measured factors and the scale items was as expected (Velayutham and Aldridge, 2012). It is worth noting that we performed a second-order CFA to test whether the first-order factors are sub-dimensions of a higher-order construct (Ning, 2017). The appropriateness of the second-order factors helped us to clarify if we can use the total score of each instrument in the subsequent mediation analysis process. The cutoff criteria for fit indexes in EFA and CFA are presented in the appendix.

(2) We conducted serial multiple mediation analyses to analyze the underlying mechanism of metacognitive influence on chemistry identity with control of the effect of gender (Hayes, 2012). To be able to simultaneously examine all the hypothetical mediation paths and control over the Type I error rates, this study used PROCESS v.3.3 SPSS macro to conduct multiple linear regressions (Preacher and Hayes, 2008; Hayes, 2013).

The PROCESS tests the mediating effect based on ordinary least-squares regression and the bias-corrected percentile bootstrap method (Yıldız, 2016), and the bias-corrected percentile bootstrap method has been proved to be one of the best methods to test the mediating effect and suitable for non-normal data (Edwards and Lambert, 2007; Preacher and Hayes, 2008; Fang et al., 2014). This study conducted mediation analyses using 5000 bootstrap samples along with 95% confidence intervals. The effect is significant if the 95% bootstrap confidence interval (95% CI) does not overlap with zero (i.e., both Boot LLCI and Boot ULCI in the data presented are positive or negative at the same time) and the criteria for significance was set as p < 0.05.

According to Hayes’ model templates 6, we proposed a hypothetical path diagram as shown in Fig. 1. Hayes’ model templates 6 is serial multiple mediator models, the feature of which is that it assumes a causal relationship between mediation variables. Therefore, the independent variable (X) influences the dependent variable (Y) through four different paths in the serial multiple mediator models, including three indirect effects from X to Y (X → M1 → Y, X → M2 → Y, X → M1 → M2 → Y) and one remaining direct effect of X to Y does not pass through any mediators (Hayes, 2013; Hong et al., 2019).


image file: d1rp00342a-f1.tif
Fig. 1 Hypothesized relationships among metacognition, chemistry learning burnout, chemistry learning flow, and chemistry identity.

Suitability for measurement instruments

Metacognition. Consistent with expectations, two factors were extracted by EFA analysis for Jr. MAI. After three rounds of EFA analysis, three items were deleted in turn. The reason is that two items had high loadings on two factors and one item's factor loading was lower than 0.4. The fourth round of EFA was performed on the remaining 15 items with a KMO value of 0.916 and Bartlett's test of sphericity indicated χ2 (105) = 1960.584 (p < 0.001). The cumulative variance explained by the two factors was 54.23%, and the factor loadings of all 15 items were within the range of 0.450–0.924. The second-order CFA results indicated that the revised Jr. MAI model fits well: χ2/df = 2.520; SRMR = 0.056; CFI = 0.927; IFI = 0.928; GFI = 0.910; RMSEA = 0.072. The Cronbach's α for the whole questionnaire was 0.902, as well as for the two sub-dimensions were 0.847 and 0.878.
Chemistry identity. As mentioned earlier, a total of 14 items were initially used to measure chemistry identity. After EFA analysis, we removed one item because its factor loading was below 0.4. EFA analysis was performed again on the remaining 13 items. The Kaiser–Meyer–Olking (KMO = 0.899) and the Barlett's test of sphericity (χ2(66) = 2295.966, p < 0.001) showed the adequacy of the data set to perform the factor analysis; the total explained variance of the three factors was 73.29% and the factor loadings of each item were ranged from 0.615–0.997. The second-order CFA results were as follows: χ2/df = 2.386; SRMR = 0.050; CFI = 0.966; TLI = 0.958; GFI = 0.927; RMSEA = 0.068. The CFA results supported that the second-order model was well-fitted model structures for chemistry identity. In addition, the Cronbach's α coefficient for the whole questionnaire and the three sub-dimensions were 0.915, 0.869, 0.930, 0.845 respectively, indicating that the questionnaire for chemistry identity had high reliability.
Chemistry learning burnout. EFA was performed on sample 1 with a KMO of 0.895 and Bartlett's spherical test showed χ2(91) = 1880.326 and p < 0.001, indicating suitability for EFA. Consistent with the prediction, three factors were extracted and explained 62.98% of the total variance. The factor loadings for all items were in the range of 0.435–0.838. The results of the second-order CFA showed that χ2/df = 2.624; SRMR = 0.064; CFI = 0.924; TLI = 0.903; GFI = 0.920; RMSEA = 0.074. In addition, Cronbach's α coefficient was 0.861.
Chemistry learning flow. Since no changes were made to the SAFSL in this study, as well as reliability and validity for the content and construction of the SAFSL have been reported in Lei et al. (2012), we only did CFA and Cronbach analysis in present study. The internal consistency of Cronbach's α coefficient for this study was calculated to 0.929. To ensure the appropriateness of the total score, we conducted a second-order CFA and the results showed a good model fit: χ2/df = 2.505; SRMR = 0.042; GFI =0.935; CFI = 0.968; TLI = 0.957; RMSEA = 0.071.

Common method deviation test

Harman's single factor was used to estimate the common source variance (Podsakoff et al., 2003). The results showed that 9 factors’ initial eigenvalues were higher than 1, and the explanatory rate of the first common factor was 12.97%, which was lower than 40%. And the cumulative interpretation total variance was 63.56%. The first factor explained the variance was less than half of the cumulative total variance. Thus, no common method bias effect was observed between the main variables (Mo et al., 2019).

Results

Descriptive statistics and bivariate correlations

The mean, standard deviation, and correlation coefficient correlations of all the variables were displayed in Table 3. The skewness and kurtosis results indicate that the data are normally distributed. Bivariate correlations results showed that chemistry identity positively correlated with metacognition (r = 0.52, p < 0.001) and chemistry learning flow (r = 0.71, p < 0.001), as well as negatively correlated with burnout (r = −0.58, p < 0.001). The metacognition had a significantly negative correlation with chemistry learning burnout (r = −0.35, p < 0.001) and a positive correlation with chemistry learning flow (r = 0.54, p < 0.001); as well as burnout was significantly negatively correlated with chemistry learning flow (r = −0.57, p < 0.001).
Table 3 Descriptive statistics and bivariate correlations
Variable Mean SD Min Max Skew Kurt 1 2 3
Notes: *p < 0.05; **p < 0.01; ***p < 0.001. 1, 2, 3 represent metacognition, chemistry learning burnout, and chemistry learning flow respectively.
Metacognition 50.70 8.69 15 75 0.35 1.01 —
Chemistry learning burnout 35.36 9.02 14 59 −0.13 −0.39 −0.35*** —
Chemistry learning flow 37.90 8.73 12 60 0.05 0.81 0.54*** −0.57*** —
Chemistry learning identity 37.96 9.48 13 65 −0.11 −0.38 0.52*** −0.58*** 0.71***

The chain mediating effects analyses

Considering some previous studies have shown that there may be gender differences in students’ identity, gender was used as a control variable in this study (Robinson et al., 2019). After controlling for the effect of gender, the multiple linear regression results are displayed in Table 4 and Fig. 2. Multiple linear regression tested whether the effects between the different variables were positive or negative, as well as whether the effect sizes were significant. Specifically, if the 95% bootstrap confidence interval (Boot LLCI → Boot ULCI) does not overlap with zero, the effect of the predictor variable on the outcome variable is significant. Beta (β) indicates the standardized regression coefficient, and its value is positive, indicating that the predictive variable has a positive impact on the outcome variable; otherwise, the effect is negative.
Table 4 Results of multiple linear regression after controlling for the effect of gender (n = 594)
Predictor variable Outcome variable R R 2 F β t Bootstrap 95% CI
Boot LLCI Boot ULCI
Notes: All variables used in this table have been standardized. The gender variable was virtualized. Boot LLCI and Boot ULCI refer to the lower and upper limits of the 95% confidence interval estimated by the deviation-corrected percentile Bootstrap method. Bootstrap sample size = 5000.
Eqn (1)
Metacognition Burnout 0.35 0.12 41.20 −0.34 −8.84 −0.43 −0.25
Gender −0.07 −0.94 −0.22 0.07
Eqn (2)
Metacognition Flow 0.70 0.49 188.04 0.37 11.69 0.28 0.45
Burnout −0.43 −13.77 −0.52 −0.35
Gender 0.34 5.65 0.22 0.45
Eqn (3)
Metacognition Identity 0.75 0.57 194.05 0.18 5.71 0.12 0.25
Burnout −0.25 −7.54 −0.33 −0.18
Flow 0.45 11.75 0.37 0.53
Gender 0.14 2.44 0.02 0.25


image file: d1rp00342a-f2.tif
Fig. 2 The chain mediating effect of metacognition and chemistry identity.

Firstly, the results of eqn (1) indicated that metacognition had a negative significant effect on chemistry learning burnout (β = −0.34, p < 0.001). Secondly, after including chemistry learning flow into the regression equation, eqn (2) showed that metacognition had a significant positive effect on chemistry learning flow (β = 0.37, p < 0.001), while burnout has a negative effect on chemistry learning flow (β = −0.43, p < 0.001). Thirdly, after adding chemistry identity to the regression equation, eqn (3) indicated that both the level of metacognition and chemistry learning flow had a direct and significant effect on the performance of chemistry identity (β = 0.18, p < 0.001; β = 0.45, p < 0.001); chemistry learning burnout had a significantly negative effect on chemistry identity (β = −0.25, p < 0.001).

After clarifying the significant effects among the variables, we tested whether the direct and indirect effects (i.e., mediating effects) of metacognition on chemistry identity were significant. Table 5 presents the specific results of mediation analysis, including the total effect, the direct effect, and the three mediating effects. Similarly, if the 95% confidence interval (95% CI) of the coefficients of each pathway did not include 0, it indicates a significant mediating effect of the indirect pathway.

Table 5 Indirect effect of chemistry learning burnout and chemistry learning flow after controlling for the effect of gender
Effect Boot SE Boot LLCI Boot ULCI Ratio of indirect to total effect (%)
Notes: Indirect effect 1 was metacognition → chemistry learning burnout → chemistry identity. Indirect effect 2 was metacognition → chemistry learning flow → chemistry identity. Indirect effect 3 was metacognition → chemistry learning burnout → chemistry learning flow → chemistry identity (the variable processing is the same as in the previous table).
Total effect 0.50 0.03 0.43 0.57 —
Direct effect 0.18 0.03 0.12 0.25 —
Total indirect effect 0.32 0.03 0.26 0.37 64
Indirect effect 1 0.09 0.02 0.05 0.13 18
Indirect effect 2 0.16 0.03 0.11 0.21 32
Indirect effect 3 0.07 0.01 0.04 0.09 14

As can be seen in Table 5, the total effect of metacognitive awareness on chemistry identity was 0.50, with a direct effect of 0.18 (95% CI is 0.12–0.25), supporting hypothesis 1. The indirect standardized mediated effect was significant with a total value of 0.32 (95% CI is 0.26–0.37), accounting for 64% of the total effect. Specifically, metacognition influences students’ chemistry identity through three indirect pathways: the indirect effect of metacognition → chemistry learning burnout → chemistry identity was 0.09 (95% CI is 0.05–0.13), supporting hypothesis 2; the indirect effect of metacognition → chemistry learning flow → chemistry identity was 0.16 (95% CI is 0.11–0.22), supporting hypothesis 3; the indirect effect of metacognition → chemistry learning burnout → chemistry learning flow → chemistry identity was 0.07 (95% CI is 0.04–0.09), supporting hypothesis 4. The effects of direct and indirect paths were significant, which indicates that the mediating variable plays a partial mediating role.

Discussion and conclusion

As mentioned earlier, chemistry identity is vital for students’ persistence within chemistry and even STEM fields. To determine teaching interventions that could be effective in improving student chemistry identity, this research investigated how chemistry identity developed in high school students is associated with metacognition, chemistry learning burnout, and chemistry learning flow.

The relationship between metacognition and chemistry identity

The first major contribution of this study is the finding that metacognition has an impact on the development of chemistry identity. Many theories and much evidence have directly or indirectly suggested the close relationship between metacognitive awareness and identity. For example, Berzonsky's identity style model emphasizes the central role of cognitive processing in identity formation (Berzonsky, 2008). Metacognition involves knowledge and the ability to manage information; plan, monitor, and evaluate cognitive processes; regulate and modify cognitive representations (Goupil and Kouider, 2019), which are crucial to the formation and modification of the cognitive process of identity (Phillips, 2008; Welsh and Schmitt-Wilson, 2013). In addition, previous research has suggested that metacognition is strongly associated with students’ positive psychological structures, such as persistence in learning and interest (Tsai et al., 2018), which are important components of identity (Hazari et al., 2010; Huvard et al., 2020). Although studies have demonstrated that metacognition could affect identity formation (Welsh and Schmitt-Wilson, 2013), there is no direct empirical evidence that metacognitive awareness has a positive effect on students’ science identity, such as their chemistry identity. The results of the present study provide direct empirical evidence that metacognition has a significant positive effect on students’ chemistry identity (direct effect = 0.18).

The mediating roles of chemistry learning burnout and chemistry learning flow

The second contribution of this study concerned the exploration of the mediating effects of chemistry learning burnout and chemistry learning flow on the association between metacognition and chemistry identity. The direct and indirect effects between metacognition on chemistry identity were all significant, suggesting that both chemistry learning burnout and chemistry learning flow played partial mediating roles in the relationship between metacognition and chemistry identity (total indirect effect = 0.32).

As stated by EIT, although talent and abilities are crucial to the development of adolescents’ identity, participation in activities provides space for individuals to explore their talents and abilities, and subjective experiences generated in activities can further effectively promote the development of adolescents’ identity (Sharp et al., 2007). Therefore, the positive association between metacognition and chemistry identity was mostly mediated through chemistry learning flow in the present study (β = 0.16): metacognition (i.e., skill) affected chemistry learning flow (i.e., subjective experiences), while flow experiences in the chemistry learning activity had a strong impact on the development of chemistry identity. In addition, this result is in accordance with the study of Bakker and Woerkom (2017), which demonstrated that the use of metacognitive strategies such as self-setting goals and self-feedback met the three basic needs of psychological development (autonomy, competence, and relatedness) (Guardia and Jennifer, 2009); promoted the experience of flow; and then produced desirable outcomes, such as the development of identity, good academic performance, and the improvement in well-being.

In addition, the 95% CI of the conditional indirect effects of chemistry learning burnout did not include 0, showing that chemistry learning burnout had a significant mediating effect between metacognition and chemistry identity (β = 0.09). The development of metacognitive awareness can help students immediately perceive, understand and adjust their negative emotions to effectively reduce their sense of burnout in the process of chemistry learning (Seibert et al., 2017; Pennequin et al., 2020). An increase in learning burnout causes students to lose interest in learning and fail to perceive the meaning of learning (Hu and Yeo, 2020), affects students’ identity processing style, and has a negative impact on students’ self-perception of academic ability (Erentaite et al., 2018; Verhoeven et al., 2018); thus, chemistry learning burnout negatively predicts students’ chemistry identity (Verhoeven et al., 2018).

The chain mediating effect of chemistry learning burnout and chemistry learning flow

As expected, the chain mediating effect of metacognition → chemistry learning burnout → chemistry learning flow → chemistry identity was significant (β = 0.07). This finding suggests that chemistry learning burnout plays a mediating role in the relationship between metacognition and chemistry learning flow, whereas chemistry learning flow acts as a mediator between learning burnout and chemistry identity. This finding is consistent with many studies showing that the improvement in metacognitive awareness can reduce students’ learning burnout level (Saricam et al., 2017; Wang, 2019), and that learning burnout is negatively related to flow experience (Kasa and Hassan, 2016). Furthermore, previous studies have shown that students’ flow experiences encourage them to persist in learning, stimulate their intrinsic learning motivation, and improve their learning satisfaction (Joo et al., 2012; Mesurado et al., 2016), which is conducive to the development of identity.

Implications for practice

To develop students’ chemistry identity, teachers should enable students to participate in autonomous learning with embedded metacognitive prompts. According to the results of the present study, improving students’ metacognition is feasible for the development of students’ chemistry identity. Successful autonomous learning requires activation of and engagement in metacognitive awareness (Papamitsiou and Economides, 2019). Therefore, autonomous learning can provide opportunities for students to improve their metacognition (Zhang et al., 2015). Metacognitive prompts are seen as a key strategy to enable learners to understand and appropriately apply metacognitive knowledge and metacognitive strategies, to act autonomously and to become central leaders in the learning process (Lavi et al., 2019; Papamitsiou and Economides, 2019). Thus, metacognitive prompts are useful for improving metacognitive awareness and autonomous learning performance. Autonomous learning has also been widely accepted as having an impact on the development of students’ identity (Benson, 2007; Martinhansen, 2018).

Furthermore, the results of this study show that metacognition also has an indirect impact on chemistry identity through chemistry learning flow. Flow experience relies greatly on perceived autonomy and motivation (Kowal and Fortier, 1999). According to self-determination theory, autonomous learning emphasizes the provision of autonomy support for students, and stimulates students’ intrinsic motivation (Papamitsiou and Economides, 2019). Previous studies also have shown that autonomy-supportive teaching behaviour with metacognitive strategies (Ohtani and Hisasaka, 2018) could lead to significant differences in students’ intrinsic motivation and learning flow (Aurah, 2013). Thus, teachers should especially provide students with guidance for understanding and applying metacognitive strategies to support their autonomous learning, thereby achieving the purpose of strengthening students’ chemistry identity.

Context-based learning (CBL), which is currently an important educational concept, should be stressed in chemistry learning to enhance students’ chemistry identity. The key features of context-based learning are the use of real situations as the starting point and anchor for science learning (Prins et al., 2018), and the requirement of students to take the initiative to apply their existing knowledge to interpret new information to further establish connections between old and new knowledge (Dori et al., 2018). Therefore, context-based learning serves as a platform for students to activate their metacognition (Dori et al., 2018). In addition to metacognition, this study emphasizes the key roles of chemistry learning flow and the level of chemistry learning burnout in the development of chemistry identity. Because CBL allows students to establish an understanding of the relevance of scientific concepts in the real world (Prins et al., 2018), it helps students perceive the meaning of learning, stimulates their learning motivation (Overman et al., 2014), increases their interest and enjoyment in chemistry learning, further reduces their chemistry learning burnout (Sevian et al., 2018) and increases their experience of learning flow.

In addition, as argued by Vincent-Ruz and Schunn (2018), apart from linking classroom science learning to real-world science, a sense of community and affiliation also drives science identity. Considering the important role of context-based learning and autonomous learning in promoting the development of students’ chemistry identity, we further suggest that socioscientific issues (SSIs) should be incorporated into chemistry classroom teaching. One reason is that SSIs, as complex issues arising from the complex interaction between science and society (Sadler and Zeidler, 2005), can provide an opportunity for learners to apply scientific knowledge to situations closely related to daily life. In this way, students can perceive and give practical significance to scientific content and learning process, which could thereby enhance their learning motivation and interest (Sevian et al., 2018). A second reason for the incoporation of SSIs into chemistry classroom teaching is that a key characteristic of the use of SSIs is the combination of scientific knowledge, moral judgements, and other social factors for reasoning, argumentation, and social decisions (Xiao and Sandoval, 2017). By debating, discussing, and evaluating each other's ideas with a group of people who hold the same or different positions, students form a sense of community and affiliation.

Limitations and future directions

The present study was conducted with students in the high school first grade and mainly focused on the underlying effect mechanisms between metacognition and chemistry identity. Nevertheless, there are some limitations of this study that need further research in the future. First, the present study is cross-sectional and can only illustrate the correlations between several variables, as well as possible causal relationships (Rozgonjuk et al., 2020). The study does not provide an absolute model of causal relationships between variables but provides evidence to support the causal relationships described in existing theories (Hosbein and Barbera, 2020b). Future longitudinal studies are needed to confirm the causal relationships between several variables.

Second, the findings of this study are limited by the sample and are only representative of how metacognition is related to chemistry identity formation among secondary school students in most regions of China. To improve the universality and representative of the findings, future studies should examine students at different developmental stages and in different cultural contexts to enhance the generalizability and representativeness of the findings.

Third, the survey of all variables in this study used self-reported data, which has limitations. Although self-reports reflect the reality of the individual, a few students may tend to answer what is acceptable to them rather than what is completely true. Therefore, future surveys of students, parents, and teachers need to be conducted simultaneously. The validity of the study can be further enhanced by comparing the findings of all three.

Fourth, previous studies indicate that both internal and external factors influence students’ science identity (Martin-Hansen, 2018), while this research only explored the influence of internal factors on chemistry identity. For example, teaching style, the openness of the learning environment, school environment, interactions among students, and parents’ occupations may have an impact on students’ chemistry identity. Therefore, future research can also investigate how external factors shape students’ chemistry identity. Considering both external and internal influences together facilitates us to come up with a more comprehensive and effective solution to enhance students’ chemistry identity.

Funding

This research was supported by the Fundamental Research Funds for the Central Universities (2019TS131), National Natural Science Foundation of China (31871118), Research Program Funds of the Collaborative Innovation Center of Assessment toward Basic Education Quality at Beijing Normal University (2019-05-002 BZPK01; 2020-05-002-BZPK01) and Interdisciplinary Project of Learning Science of Shaanxi Normal University.

Conflicts of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Appendix: cutoff criteria for fit indexes in EFA and CFA

For EFA, it is first necessary to clarify whether the data are suitable for EFA analysis based on KMO and Bartlett's sphericity test. Second, the appropriate factors are extracted based on the factor eigenvalues. Finally, inappropriate items are removed based on factor loadings (Costello and Osborne, 2005). Regarding the CFA, the goodness of model fit is based on a set of indices. And the fit indices and acceptable ranges for EFA and CFA are shown in Table 6 (Browne and Cudeck, 1993; Byrne, 2010; Field, 2000).
Table 6 Cutoff criteria for fit indexes in EFA and CFA
Indicator Acceptable range
EFA
Kaiser–Meyer–Olkin (KMO) Greater than 0.6
Bartlett's sphericity test p less than 0.05
Eigenvalue Greater than 1
Factor loading Greater than 0.4
CFA
χ 2/df Less than 5
Standardized root mean square residual (SRMR) Close to or below 0.08
Root mean square error of approximation (RMSEA) Less than 0.08
Goodness-of-fit index (GFI) Greater than 0.9
Comparative fit index (CFI) Greater than 0.9
Tucker–Lewis index (TLI) Greater than 0.9

References

  1. Al-Baddareen G., Ghaith S. and Akour M., (2015), Self-efficacy, achievement goals, and metacognition as predicators of academic motivation, Procedia – Soc. Behav. Sci., 191, 2068–2073 DOI:10.1016/j.sbspro.2015.04.345.
  2. Aurah M. C., (2013), The effects of self-efficacy beliefs and metacognition on academic performance: A mixed method study, Am. J. Educ. Res., 1(8), 334–343 DOI:10.12691/education-1-8-11.
  3. Bachen C. M., Hernndezramos P., Raphael C. and Waldron A., (2016), How do presence, flow, and character identification affect players empathy and interest in learning from a serious computer game? Comput. Hum. Behav., 64, 77–87 DOI:10.1016/j.chb.2016.06.043.
  4. Bakker A. B. and Woerkom M. V., (2017), Flow at work: A self-determination perspective, Occup. Health Sci., 1(1), 47–65 DOI:10.1007/s41542-017-0003-3.
  5. Benson P., (2007), Autonomy in language teaching and learning, Lang. Teach., 40(1), 21–40 DOI:10.1017/S0261444806003958.
  6. Berzonsky M. D., (2008), Identity formation: The role of identity processing style and cognitive processes, Person. Individual Differ., 44(3), 645–655 DOI:10.1016/j.paid.2007.09.024.
  7. Berzonsky M. D., (2011), Social-cognitive perspective on identity construction, in Schwartz S. J., Luyckx K. and Vignoles V. L. (ed.), Handbook of identity theory and research, New York, NY: Springer, pp. 55–76.
  8. Berzonsky M. and Kuk L. S., (2000), Identity status, identity processing style, and the transition to university, J. Adolesc. Res., 15(1), 81–98 DOI:10.1177/0743558400151005.
  9. Bose D. K., (2008), Review of flow: The psychology of optimal experience, Inf. Des. J., 16(1), 75 DOI:10.1075/idj.16.1.11bos.
  10. Browne M. W. and Cudeck R., (1993), Alternative ways of assessing model fit, in Bollen K. A. and Long J. S. (ed.), Testing structural equation models, Sage Publications, pp. 136–162.
  11. Byrne B., (2010), Structural equation modeling with AMOS. Basic concepts, applications, and programming, New York: Routledge.
  12. Chen S. and Wei B., (2020), Development and validation of an instrument to measure high school students’ science identity in science learning, Res. Sci. Educ. DOI:10.1007/s11165-020-09932-y.
  13. Coatsworth J. D., Palen L.-A., Sharp E. H. and Ferrer-Wreder L., (2006), Self-defining activities, expressive identity, and adolescent wellness, Appl. Dev. Sci., 10(3), 157–170 DOI:10.1207/s1532480xads1003_5.
  14. Costello A. B. and Osborne J. W., (2005), Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis, Pract. Assess. Res. Eval., 10, 1–9.
  15. Csikszentmihalyi M., (1996), Creativity: Flow and the psychology of discovery and invention, New York, NY: Harper and Collins DOI:10.5860/CHOICE.28-0597.
  16. Csikszentmihalyi M., Abuhamdeh S. and Nakamura J., (2005), Flow, in Elliot A. J. and Dweck C. S. (ed.), Handbook of competence and motivation, New York, NY: Guilford Press, pp. 598–608.
  17. Dignath C. and Büttner G., (2008), Components of fostering self-regulated learning among students. A meta-analysis on intervention studies at primary and secondary school level, Metacogn. Learn., 3(3), 231–264 DOI:10.1007/s11409-008-9029-x.
  18. Dori Y. J., Avargil S., Kohen Z. and Saar L., (2018), Context-based learning and metacognitive prompts for enhancing scientific text comprehension, Int. J. Sci. Educ., 40(10), 1198–1220 DOI:10.1080/09500693.2018.1470351.
  19. Edwards J. R. and Lambert L. S., (2007), Methods for integrating moderation and mediation: a general analytical framework using moderated path analysis, Psychol. Methods, 12(1), 1–22, DOI:10.1037/1082-989X.12.1.1.
  20. Erentaite R., Vosylis R., Gabrialaviciute I. and Raiziene S., (2018), How does school experience relate to adolescent identity formation over time? Cross-lagged associations between school engagement, school burnout and identity processing styles, J. Youth Adolesc., 47(4), 760–774 DOI:10.1007/s10964-017-0806-1.
  21. Fang J, Wen Z, Zhang M, Sun P., (2014), The analyses of multiple mediation effects based on structural equation modeling, J. Psychol. Sci., 37(3), 735–741 DOI:10.16719/j.cnki.1671-6981.2014.03.001.
  22. Field A., (2000), Discovering statistics using SPSS for windows: Advanced techniques for the beginner, London: Sage.
  23. Flavell J. H., (1979), Metacognition and cognitive monitoring: A new area of cognitive–developmental inquiry, Am. Psychol., 34(10), 906.
  24. Fleur D. S., Bredeweg B. and van den Bos W., (2021), Metacognition: Ideas and insights from neuro- and educational sciences, NPJ Sci. Learn., 6(1), 13 DOI:10.1038/s41539-021-00089-5.
  25. Ghanizadeh A. and Jahedizadeh S., (2017), The Nexus between emotional, metacognitive, and motivational facets of academic achievement among Iranian university students, J. Appl. Res. High. Educ., 9(4), 598–615 DOI:10.1108/jarhe-05-2017-0060.
  26. Godwin A., (2016), The development of a measure of engineering identity, in Paper presented at 2016 ASEE annual conference and exposition, New Orleans, Louisiana DOI:10.18260/p.26122.
  27. Gonzalez A. and Paoloni P., (2015), Perceived autonomy-support, expectancy, value, metacognitive strategies and performance in chemistry: A structural equation model in undergraduates, Chem. Educ. Res. Pract., 16(3), 640–653.
  28. Goupil L. and Kouider S., (2019), Developing a reflective mind: From core metacognition to explicit self-reflection, Curr. Dir. Psychol. Sci., 28(4), 403–408 DOI:10.1177/0963721419848672.
  29. Guardia L. A. and Jennifer G., (2009), Developing who I am: A self-determination theory approach to the establishment of healthy identities, Educ. Psychol., 44(2), 90–104 DOI:10.1080/00461520902832350.
  30. Hair J. F., (2006), Multivariate data analysis, Pearson Education India.
  31. Han J.-R. and Kim J.-M., (2017), The mediating effects of self-efficacy between metacognition and learning flow in college students in healthcare field, J. Digit. Converg., 15(6), 273–282 DOI:10.14400/JDC.2017.15.6.273.
  32. Hayes A. F., (2012), PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling, http://www.afhayes.com.
  33. Hayes A. F., (2013), Introduction to mediation, moderation, and conditional process analysis: A regression-based approach, New York: Guilford Press.
  34. Hazari Z., Sonnert G., Sadler P. M. and Shanahan M.-C., (2010), Connecting high school physics experiences, outcome expectations, physics identity, and physics career choice: A gender study, J. Res. Sci. Teach. DOI:10.1002/tea.20363.
  35. Hong Y., Lee T. and Kim J. S., (2019), Serial multiple mediation analyses: How to enhance individual public health emergency preparedness and response to environmental disasters, Int. J. Environ. Res. Publ. Health, 16(2) DOI:10.3390/ijerph16020223.
  36. Hosbein K. N. and Barbera J., (2020a), Alignment of theoretically grounded constructs for the measurement of science and chemistry identity, Chem. Educ. Res. Pract., 21(1), 371–386 10.1039/c9rp00193j.
  37. Hosbein K. N. and Barbera J., (2020b), Development and evaluation of novel science and chemistry identity measures, Chem. Educ. Res. Pract., 21(3), 852–877 10.1039/c9rp00223e.
  38. Hu X. and Yeo G. B., (2020), Emotional exhaustion and reduced self-efficacy: The mediating role of deep and surface learning strategies, Motiv. Emot., 44(5), 785–795 DOI:10.1007/s11031-020-09846-2.
  39. Huvard H., Talbot R. M., Mason H., Thompson A. N., Ferrara M. and Wee B., (2020), Science identity and metacognitive development in undergraduate mentor-teachers, Int. J. STEM Educ., 7(1) DOI:10.1186/s40594-020-00231-6.
  40. Joo Y. J., Joung S. and Kim J., (2012), Structural relationships among self-regulated learning, learning flow, satisfaction, and learning persistence in cyber universities. Interactive Learning Environments, 22(6), 752–770 DOI:10.1080/10494820.2012.745421.
  41. Kasa M. and Hassan Z., (2016), Burnout and flow with moderating effect of individualism/collectivism. A study in Malaysian Hotel industry. Procedia – Soc. Behav. Sci., 219, 416–423 DOI:10.1016/j.sbspro.2016.05.063.
  42. Kowal J. and Fortier M. S., (1999), Motivational Determinants of Flow: Contributions From Self-Determination Theory. The Journal of Social Psychology, 139(3), 355–368 DOI:10.1080/00224549909598391.
  43. Kowal J. and Fortier M. S., (2000), Testing relationships from the hierarchical model of intrinsic and extrinsic motivation using flow as a motivational consequence, Res. Q. Exerc. Sport, 71(2), 171–181 DOI:10.1080/02701367.2000.10608895.
  44. Koyuncu İ. and Kılıç A. F., (2019), The Use of Exploratory and Confirmatory Factor Analyses: A Document Analysis, Ted EĞİtİm Ve Bİlİm, DOI:10.15390/eb.2019.7665.
  45. Lavi R., Shwartz G. and Dori Y. J., (2019), Metacognition in chemistry education: A literature review, Israel J. Chem. DOI:10.1002/ijch.201800087.
  46. Lei L., Ma X. and Wang W., (2012), Development and application of the scale for adolescents’ flow state in learning, Psychol. Res., 5(1), 45–48.
  47. Li C., Zhang L. J. and Jiang G., (2021), Conceptualisation and measurement of foreign language learning burnout among Chinese EFL students, J. Multiling. Multicult. Dev., 1–15 DOI:10.1080/01434632.2021.1931246.
  48. Ling L., Qin S. and Shen L.-F., (2014), An investigation about learning burnout in medical college students and its influencing factors, Int. J. Nurs. Sci., 1(1), 117–120 DOI:10.1016/j.ijnss.2014.02.005.
  49. Luo Y., Wang Z., Zhang H. and Chen A., (2016), The influence of family socio-economic status on learning burnout in adolescents: Mediating and moderating effects, J. Child Family Stud., 25(7), 2111–2119 DOI:10.1007/s10826-016-0400-2.
  50. Luyckx K., Vansteenkiste M., Goossens L. and Duriez B., (2009), Basic need satisfaction and identity formation: Bridging self-determination theory and process-oriented identity research, J. Counsel. Psychol., 56(2), 276–288 DOI:10.1037/a0015349.
  51. Mao Y., Roberts S., Pagliaro S., Csikszentmihalyi M. and Bonaiuto M., (2016), Optimal experience and optimal identity: A multinational study of the associations between flow and social identity, Front. Psychol., 7, 67 DOI:10.3389/fpsyg.2016.00067.
  52. Marcia and James, E., (1966), Development and validation of ego-identity status, J. Person. Soc. Psychol., 3(5), 551 DOI:10.1037/h0023281.
  53. Martin-Hansen L., (2018), Examining ways to meaningfully support students in STEM, Int. J. STEM Educ., 5(1), 53 DOI:10.1186/s40594-018-0150-3.
  54. Martinhansen L., (2018), Examining ways to meaningfully support students in STEM, Int. J. STEM Educ., 5(1), 1–6 DOI:10.1186/s40594-018-0150-3.
  55. Matthews G., Hillyard E. J. and Campbell S. E., (1999), Metacognition and maladaptive coping as components of test anxiety, Clin. Psychol. Psychother., 6(2), 111–125 DOI:10.1002/(SICI)1099-0879(199905)6:2<111::AID-CPP192>3.0.CO;2-4.
  56. Mesurado B., Richaud M. C. and Mateo N. J., (2016), Engagement, flow, self-efficacy, and eustress of university students: A cross-national comparison between the Philippines and Argentina, J. Psychol., 150(3), 281–299 DOI:10.1080/00223980.2015.1024595.
  57. Mo J., Wang C., Niu X., Jia X., Liu T. and Lin L., (2019), The relationship between impulsivity and self-injury in Chinese undergraduates: The chain mediating role of stressful life events and negative affect, J. Affect. Disord., 256, 259–266 DOI:10.1016/j.jad.2019.05.074.
  58. Ning H. K., (2017), The Bifactor Model of the Junior Metacognitive Awareness Inventory (Jr. MAI), Curr. Psychol., 38(2), 367–375, DOI:10.1007/s12144-017-9619-3.
  59. Ohtani K. and Hisasaka T., (2018), Beyond intelligence: A meta-analytic review of the relationship among metacognition, intelligence, and academic performance, Metacogn. Learn., 13(2), 179–212 DOI:10.1007/s11409-018-9183-8.
  60. Overman M., Vermunt J. D., Meijer P. C., Bulte A. M. W. and Brekelmans M., (2014), Students' perceptions of teaching in context-based and traditional chemistry classrooms: Comparing content, learning activities, and interpersonal perspectives, Int. J. Sci. Educ., 36(11), 1871–1901 DOI:10.1080/09500693.2013.880004.
  61. Öz H., (2016), Metacognitive awareness and academic motivation: A cross-sectional study in teacher education context of Turkey, Procedia – Soc. Behav. Sci., 232, 109–121 DOI:10.1016/j.sbspro.2016.10.035.
  62. Papamitsiou Z. and Economides A. A., (2019), Exploring autonomous learning capacity from a self-regulated learning perspective using learning analytics, Br. J. Educ. Technol., 50(6), 3138–3155 DOI:10.1111/bjet.12747.
  63. Pennequin V., Questel F., Delaville E., Delugre M. and Maintenant C., (2020), Metacognition and emotional regulation in children from 8 to 12 years old, Br. J. Educ. Psychol., 90 Suppl 1, 1–16 DOI:10.1111/bjep.12305.
  64. Perez T., Cromley J. G. and Kaplan A., (2014), The role of identity development, values, and costs in college STEM retention, J. Educ. Psychol., 106(1), 315–329.
  65. Phillips T. M., (2008), Age-related differences in identity style: A cross-sectional analysis, Curr. Psychol., 27(3), 205–215 DOI:10.1007/s12144-008-9035-9.
  66. Podsakoff P. M., MacKenzie S. B., Lee J.-Y. and Podsakoff N. P., (2003), Common method biases in behavioral research: A critical review of the literature and recommended remedies, J. Appl. Psychol., 88(5), 879–903 DOI:10.1037/0021-9010.88.5.879.
  67. Pop E. I., Negru-Subtirica O., Crocetti E., Opre A. and Meeus W., (2016), On the interplay between academic achievement and educational identity: A longitudinal study, J Adolesc, 47, 135–144 DOI:10.1016/j.adolescence.2015.11.004.
  68. Preacher K. J. and Hayes A. F., (2008), Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models, Behav. Res. Methods, 40(3), 879–891 DOI:10.3758/brm.40.3.879.
  69. Prins G. T., Bulte A. M. W. and Pilot A., (2018), Designing context-based teaching materials by transforming authentic scientific modelling practices in chemistry, Int. J. Sci. Educ., 40(10), 1108–1135 DOI:10.1080/09500693.2018.1470347.
  70. Rehman A. U., Bhuttah T. M. and You X., (2020), Linking burnout to psychological well-being: The mediating role of social support and learning motivation, Psychol. Res. Behav. Manag., 13, 545–554 DOI:10.2147/PRBM.S250961.
  71. Robinson K. A., Perez T., Carmel J. H. and Linnenbrink-Garcia L., (2019), Science identity development trajectories in a gateway college chemistry course: Predictors and relations to achievement and STEM pursuit, Contemp. Educ. Psychol., 56, 180–192 DOI:10.1016/j.cedpsych.2019.01.004.
  72. Rodríguez-Ardura I. and Meseguer-Artola A., (2017), Flow in e-learning: What drives it and why it matters, Br. J. Educ. Technol., 48(4), 899–915 DOI:10.1111/bjet.12480.
  73. Rozgonjuk D., Kraav T., Mikkor K., Orav-Puurand K. and Täht K., (2020), Mathematics anxiety among STEM and social sciences students: the roles of mathematics self-efficacy, and deep and surface approach to learning, Int. J. STEM Educ., 7(1), DOI:10.1186/s40594-020-00246-z.
  74. Sadler T. D. and Zeidler D. L., (2005), The significance of content knowledge for informal reasoning regarding socioscientific issues: Applying genetics knowledge to genetic engineering issues, Sci. Educ., 89(1), 71–93 DOI:10.1002/sce.20023.
  75. Saricam H., (2015), Metacognition and happiness: The mediating role of perceived stress, Stud. Psychol., 57(4), 271–283 DOI:10.21909/sp.2015.03.699.
  76. Saricam H., Celik I. and Sakiz H., (2017), Mediator role of metacognitive awareness in the relationship between educational stress and school burnout among adolescents, J. Educ. Fut., 11, 159–175.
  77. Schaufeli W. B., Martinez I. M., Pinto A. M., Salanova M. and Bakker A. B., (2002), Burnout and engagement in university students a cross-national study, J. Cross-Cult. Psychol., 33(5), 464–481 DOI:10.1177/0022022102033005003.
  78. Schraw G., (1998), Promoting general metacognitive awareness, Instruct. Sci., 26(1), 3–16 DOI:10.1007/978-94-017-2243-8_1.
  79. Schraw G. and Dennison R. S., (1994), Assessing metacognitive awareness, Contemp. Educ. Psychol., 19(4), 460–475 DOI:10.1006/ceps.1994.1033.
  80. Seaman J., Sharp E. H. and Coppens A. D., (2017), A dialectical approach to theoretical integration in developmental-contextual identity research. Dev Psychol, 53(11), 2023–2035 DOI:10.1037/dev0000383.
  81. Seibert G. S., Bauer K. N., May R. W. and Fincham F. D., (2017), Emotion regulation and academic underperformance: The role of school burnout, Learn. Individ. Differ., 60, 1–9 DOI:10.1016/j.lindif.2017.10.001.
  82. Semerari A., Cucchi M., Dimaggio G., Cavadini D., Carcione A., Battelli V., Nicolo G., Pedone R., Siccardi T. and Angerio S. D., (2012), The development of the metacognition assessment interview: Instrument description, factor structure and reliability in a non-clinical sample, Psychiatr. Res. Neuroimaging, 200(2), 890–895 DOI:10.1016/j.psychres.2012.07.015.
  83. Sevian H., Dori Y. J. and Parchmann I., (2018), How does STEM context-based learning work: What we know and what we still do not know, Int. J. Sci. Educ., 40(10), 1095–1107 DOI:10.1080/09500693.2018.1470346.
  84. Sharp E. H., Coatsworth J. D., Darling N., Cumsille P. and Ranieri S., (2007), Gender differences in the self-defining activities and identity experiences of adolescents and emerging adults, J. Adolesc., 30(2), 251–269 DOI:10.1016/j.adolescence.2006.02.006.
  85. Skagen D., McCollum B., Morsch L. and Shokoples B., (2018), Developing communication confidence and professional identity in chemistry through international online collaborative learning, Chem. Educ. Res. Pract., 19(2), 567–582 10.1039/c7rp00220c.
  86. Sperling R. A., Howard B. C., Miller L. A. and Murphy C., (2002), Measures of children's knowledge and regulation of cognition, Contemp. Educ. Psychol., 27(1), 51–79 DOI:10.1006/ceps.2001.1091.
  87. Trujillo G. and Tanner K. D., (2014), Considering the role of affect in learning: Monitoring students' self-efficacy, sense of belonging, and science identity, CBE – Life Sci. Educ., 13(1), 6–15 DOI:10.1187/cbe.13-12-0241.
  88. Tsai Y.-h., Lin C.-h., Hong J.-c. and Tai K.-h., (2018), The effects of metacognition on online learning interest and continuance to learn with MOOCs, Comput. Educ., 121, 18–29 DOI:10.1016/j.compedu.2018.02.011.
  89. Velayutham S. and Aldridge J. M., (2012), Influence of psychosocial classroom environment on students’ motivation and self-regulation in science learning: A structural equation modeling approach, Res. Sci. Educ., 43(2), 507–527 DOI:10.1007/s11165-011-9273-y.
  90. Verdín D., (2021), The power of interest: Minoritized women's interest in engineering fosters persistence beliefs beyond belongingness and engineering identity, Int. J. STEM Educ., 8(1) DOI:10.1186/s40594-021-00292-1.
  91. Verhoeven M., Poorthuis A. M. G. and Volman M., (2018), The role of school in adolescents’ identity development. A literature review, Educ. Psychol. Rev., 31(1), 35–63 DOI:10.1007/s10648-018-9457-3.
  92. Vincent-Ruz P. and Schunn C. D., (2018), The nature of science identity and its role as the driver of student choices, Int. J. STEM Educ., 5(1), 48 DOI:10.1186/s40594-018-0140-5.
  93. Wang M., Chow A., Hofkens T. and Salmelaaro K., (2015), The trajectories of student emotional engagement and school burnout with academic and psychological development: Findings from Finnish adolescents, Learn. Instruct., 36, 57–65 DOI:10.1016/j.learninstruc.2014.11.004.
  94. Wang Y., (2019), The Relationship Among Non-English Majors' Autonomous Learning Ability, Learning Burnout and English Grades: A Structural Equation Modeling Study, Chinese J. Appl. Linguistics, 42(1), 79–91, DOI:10.1515/cjal-2019-0005.
  95. Waterman A. S., (1990), Personal expressiveness: Philosophical and psychological foundations, J. Mind Behav., 11(1), 47–74 DOI:10.1177/0022167890301007.
  96. Waterman A. S., (2004), Finding someone to be: Studies on the role of intrinsic motivation in identity formation, Identity, 4(3), 209–228 DOI:10.1207/s1532706xid0403_1.
  97. Waterman A. S., (2011) Eudaimonic identity theory: Identity as self-discovery, in Schwartz S., Luyckx K. and Vignoles V. (ed.), Handbook of Identity Theory and Research, Springer, New York, NY DOI:10.1007/978-1-4419-7988-9_16.
  98. Waterman A. S., Schwartz S. J., Goldbacher E., Green H., Miller C. and Philip S., (2003), Predicting the subjective experience of intrinsic motivation: The roles of self-determination, the balance of challenges and skills, and self-realization values, Pers. Soc. Psychol. Bull., 29(11), 1447–1458 DOI:10.1177/0146167203256907.
  99. Welsh M. and Schmitt-Wilson S., (2013), Executive function, identity, and career decision-making in college students, SAGE Open, 3(4), 215824401350575 DOI:10.1177/2158244013505755.
  100. Wu R., Huebner E. S., Zhou J. and Tian L., (2021), Relations among positivity, positive affect in school, and learning flow in elementary school students: A longitudinal mediation model, Br. J. Educ. Psychol., 91(4), 1310–1332 DOI:10.1111/bjep.12418.
  101. Xiao S. and Sandoval W. A., (2017), Associations between attitudes towards science and children's evaluation of information about socioscientific issues, Sci. Educ., 26(3–4), 247–269 DOI:10.1007/s11191-017-9888-0.
  102. Xing Q. and Chen D. D., (2010), Research on learning burnout of primary and middle school students: Present situation and prospect, J. Educ. Dev., 9, 29–33 DOI:10.16215/j.cnki.cn44-1371/g4.2010.09.018.
  103. Xu L., Wang L. and Nygaard C., (2021), Locational decisions and subjective well-being: An empirical study of Chinese urban migrants, Appl. Econ., 1–16 DOI:10.1080/00036846.2021.2005237.
  104. Yang H. and Chen J., (2015), Learning perfectionism and learning burnout in a primary school student sample: A test of a learning-stress mediation model, J. Child Family Stud., 25(1), 345–353 DOI:10.1007/s10826-015-0213-8.
  105. Yıldız M. A., (2016), Serial multiple mediation of general belongingness and life satisfaction in the relationship between attachment and loneliness in adolescents, Educ. Sci.: Theor. Pract. DOI:10.12738/estp.2016.2.0380.
  106. Zhang W. X., Hsu Y., Wang C. and Ho Y. T., (2015), Exploring the impacts of cognitive and metacognitive prompting on students’ scientific inquiry practices within an E-learning environment, Int. J. Sci. Educ., 37(3), 529–553 DOI:10.1080/09500693.2014.996796.
This journal is © The Royal Society of Chemistry 2022
Click here to see how this site uses Cookies. View our privacy policy here.