1. Introduction

This chapter discusses the centrality of student engagement in SLA, highlighting its behavioral, cognitive, emotional, and social dimensions as drivers of academic success and psychological well-being for EFL learners. It identifies a gap in brief, domain-specific instruments that can reliably capture these dimensions without overburdening respondents, especially among secondary students in Confucian heritage cultures whose collectivist values and hierarchical classroom norms shape learning experiences. Existing EFL engagement scales are lengthy and neglect the social aspect, while the promising online student engagement scale lacks validation beyond virtual settings. The study therefore adapts this scale into a concise four-dimensional tool that explicitly includes social engagement, targeting Chinese secondary EFL learners. It further addresses underexplored criterion-related validity by employing academic emotions—established predictors of engagement—as external criteria, while investigating gender as a potential moderator of these relationships.

2. Literature Review

This chapter reviewed the theoretical foundations and empirical landscape that frame student engagement in English as a Foreign Language (EFL) contexts, with particular attention to secondary learners in Confucian heritage cultures. Engagement is portrayed as the outward expression of motivation, manifesting through students’ interactions with learning resources embedded in physical, social, and educational environments. It is conceptualized as a multidimensional construct whose traditional three pillars—behavioral, emotional, and cognitive—have been augmented by two additional dimensions: agentic engagement, capturing learners’ proactive efforts to optimize their own learning, and social engagement, reflecting the quality of interpersonal interactions and the degree to which learners invest in relationships within the learning community. The discussion emphasized that these dimensions are not universal constants; rather, their salience varies across disciplines and cultural settings. Language learning is presented as an inherently interactive domain in which social engagement is indispensable, aligning with communicative and constructivist paradigms that treat social interaction as the primary vehicle for linguistic development. Empirical evidence shows that interactive tasks and cohesive social relations predict higher L2 outcomes, whereas social disconnection precipitates cognitive and behavioral disengagement. In East Asian contexts shaped by collectivist values and Confucian traditions, learners attach heightened importance to social harmony, yet the same traditions suppress agentic engagement by privileging teacher authority and hierarchical classroom dynamics. Against this backdrop, the chapter advances a re-conceptualized engagement framework that positions social engagement as a core dimension alongside behavioral, emotional, and cognitive engagement, arguing that such a model is essential for understanding EFL learners in Confucian heritage settings.

Research on engagement in EFL learning positions it as a pivotal predictor of academic achievement. Multiple antecedents have been identified, including classroom social climate and teacher-related factors such as enthusiasm, scaffolding, and respect. A burgeoning body of work has also examined the reciprocal relationship between engagement and academic emotions. Drawing on the broaden-and-build theory, positive emotions are shown to broaden learners’ cognitive and behavioral repertoires, thereby enhancing engagement, whereas negative emotions such as boredom narrow these repertoires and erode engagement. Empirical studies conducted in China and Iran corroborate these claims, revealing that enjoyment positively predicts engagement, while boredom exerts a negative influence. Gender differences further complicate this relationship. Female learners generally report higher enjoyment and lower boredom, and they also exhibit higher engagement levels than males, plausibly due to superior language aptitude, stronger motivation, and better academic performance. However, cross-cultural inconsistencies exist, suggesting that cultural norms, classroom dynamics, and instructional practices moderate these effects. Despite theoretical plausibility, empirical evidence on gender as a moderator of the emotion–engagement link remains scarce, highlighting a notable gap in the literature.

The measurement of English engagement has predominantly relied on self-report questionnaires, which, while convenient, often neglect the domain-specific nature of engagement. Existing instruments have been tailored to particular foci such as informal L2 use, task engagement, or interpreting learning, yet few scales simultaneously address the unique characteristics of EFL learning and the cultural realities of East Asian secondary classrooms. Most validated tools target Chinese learners from primary school to university, but few concentrate on the high-stakes examination pressures faced by secondary students immersed in Confucian heritage cultures. Furthermore, existing scales tend to be lengthy, increasing respondent burden and threatening data quality. A recently introduced nine-item classroom engagement scale mitigates length concerns but adheres to the traditional tripartite model and omits the social dimension. An online student engagement scale explicitly incorporates social engagement and demonstrates sound psychometric properties, yet its applicability is restricted to virtual learning environments and its criterion-related validity remains underexplored.

3. Methodology

This chapter described an exploratory mixed-methods design used to develop and validate a concise English engagement scale tailored for Chinese secondary EFL learners. Quantitative analyses determined the latent structure of engagement, while qualitative data from expert panels and student feedback refined the instrument and contextualized its use.

Participants were recruited through convenience sampling in Beijing. A pilot sample of 362 students tested the initial scale, and a main sample of 3,238 students provided data for final validation. All learners began English instruction in Grade 3 and encountered the language as a compulsory subject.

The questionnaire consisted of four sections: demographics, the adapted English engagement scale, a five-item enjoyment scale, and a six-item boredom scale. All measures employed a five-point Likert scale anchored from 1 (strongly disagree) to 5 (strongly agree).

The English engagement scale originated from a 16-item online engagement instrument covering behavioral, cognitive, affective, and social dimensions. After translation into Chinese by two linguistics graduates, an expert panel—comprising translators, secondary students, teachers, and bilingual professors—reviewed wording and added one social item reflecting student-teacher communication. Back-translation by two independent linguistics graduates ensured semantic equivalence. The resulting 17-item scale underwent pilot testing; CFA flagged low-loading items, and retrospective interviews with 15 students identified ambiguities. Thematic analysis of these interviews guided item rewording, leading to a clearer final version. The main study administered the refined scale alongside emotion measures, and comprehensive psychometric evaluation—including EFA, CFA, item, reliability, and validity tests—yielded a short, robust instrument. Example items are “I complete all English learning tasks on time” and “I look forward to my English lessons.”

The enjoyment scale, adapted from Li et al.’s 2018 private enjoyment subscale, contains five items such as “I have learned many interesting things in learning English.” Validation with the main sample produced excellent fit indices (CFI = 0.99, TLI = 0.99, RMSEA = 0.08, SRMR = 0.01) and high internal reliability (Cronbach’s α = 0.94).

The boredom scale, initially eight items from Li et al. (2023), was trimmed to six high-loading items after the pilot. An illustrative item is “The English class bores me.” In the main study the scale showed acceptable fit (CFI = 0.93, TLI = 0.89, RMSEA = 0.24, SRMR = 0.03) and outstanding reliability (Cronbach’s α = 0.97).

Data were collected via Wenjuanxing online platform. Informed consent was obtained from schools, students, and guardians; anonymity and voluntary participation were emphasized. The pilot questionnaire gathered demographics and the engagement scale; the main questionnaire added academic emotion scales. Data cleaning removed incomplete, inconsistent, or disengaged responses as well as duplicates, discarding 2.5% of cases to ensure quality.

Analyses were conducted with SPSS 27.0 and Mplus 8.3. EFA and CFA examined the engagement scale’s factor structure and validity; item analysis and reliability/validity tests assessed psychometric robustness. Correlation, multiple regression, and moderation analyses via PROCESS 4.2 explored relationships among academic emotions and engagement, including gender as a moderator.

4. Results

This chapter presents empirical evidence for the structural and psychometric adequacy of the adapted English engagement scale. A confirmatory factor analysis on Sample 1 (n = 362) produced acceptable fit indices (χ²(113)=530.27, CFI=0.90, TLI=0.88, RMSEA=0.10, SRMR=0.07). Three items with loadings below 0.70 were deleted, yielding a 14-item four-factor model. Sample 2 was split into two independent subsets (n = 1,607 and n = 1,631); t-tests confirmed no significant differences between them, justifying subsequent cross-validation. Exploratory factor analysis on Sample 3 using principal components analysis with Varimax rotation extracted four factors after two additional items (BE3 and SE2) were removed for misfit or low loading, resulting in a final 12-item structure with three items per dimension. Confirmatory factor analysis on Sample 4 corroborated this structure with excellent fit (χ²(48)=305.58, CFI=0.99, TLI=0.98, RMSEA=0.06, SRMR=0.02) and all loadings above acceptable thresholds.

Item-level diagnostics were conducted on the combined Samples 3 and 4 (n = 3,238). Skewness and kurtosis indicated normal distributions; high-low 27% group comparisons showed every item differentiated significantly (p < 0.01) between upper and lower engagement groups, supporting item discrimination.

Reliability analyses revealed strong internal consistency: Cronbach’s alpha for the full scale was 0.96; subscale alphas ranged from 0.88 to 0.96. Split-half reliability yielded a correlation of 0.83 between halves and a Spearman-Brown coefficient of 0.90. Composite reliability for the total scale reached 0.98, with subscale coefficients from 0.88 to 0.96.

Validity evidence was multifaceted. Convergent validity was supported by average variance extracted values exceeding 0.50 for all factors (0.71–0.88). Discriminant validity was demonstrated because the square roots of AVE for each factor surpassed their correlations with other factors. Criterion-related validity was evidenced by strong associations between engagement and academic emotions: enjoyment correlated positively (r = 0.89 overall, 0.67–0.93 across dimensions), whereas boredom correlated negatively (r = –0.38 overall, –0.29 to –0.41 across dimensions). Multiple regression showed enjoyment positively predicted engagement (β = 0.86, p < 0.001) and boredom negatively predicted it (β = –0.07, p < 0.001), together explaining 79 % of variance. Moderation analysis indicated gender significantly moderated these effects: enjoyment exerted a stronger positive influence on male students (β = 0.77) than on female students (β = 0.70), while boredom had a more detrimental impact on female students (β = –0.30) than on male students (β = –0.24).

5. Discussion

This chapter discussed the four-factor structure of the adapted English engagement scale, validated through EFA and CFA, confirming behavioral, cognitive, emotional, and social dimensions as reliable and interrelated facets of language learning engagement. The inclusion of social engagement extends prior three-dimensional models, recognizing peer cooperation and teacher-student interaction as essential drivers of language gains. High engagement across all four dimensions among Chinese secondary EFL learners was attributed to the interplay of high-stakes examinations and Confucian values emphasizing diligence, persistence, and hierarchical respect, producing a culturally distinctive engagement profile.

The chapter also examined how academic emotions shape engagement. Enjoyment positively predicted engagement by broadening cognitive resources and sustaining motivation, whereas boredom narrowed focus and triggered avoidance strategies, eroding engagement. Gender moderated these effects: the benefits of enjoyment were stronger for males, while the detrimental impact of boredom was more severe for females. Biological differences in neural activation, gender-specific coping strategies—rumination among females and expressive suppression among males—and sociocultural expectations that frame language learning as feminine all contributed to this divergence. These findings highlight the need for emotion-focused and gender-sensitive pedagogical interventions to enhance English engagement in Confucian-heritage classrooms.

6. Conclusion

This chapter described the development and validation of a four-dimensional English engagement scale tailored for Chinese secondary school learners, building upon the earlier framework of Hoi & Le Hang (2021). The resulting instrument is brief yet covers behavioral, cognitive, emotional, and—most prominently—social engagement dimensions. Psychometric analyses confirmed strong reliability and validity, and criterion-related evidence was provided through the associations among emotions, engagement, and gender. Enjoyment predicted higher engagement more strongly for boys, while boredom predicted lower engagement more strongly for girls, indicating gender-specific emotional pathways.

Theoretical contributions include the expansion of the engagement construct to encompass often-overlooked social interaction components, situating social engagement as a key driver of language learning success. By situating the findings within Confucian Heritage Culture (CHC), the study clarified how collectivist values and classroom norms shape the configuration and salience of engagement dimensions, offering a culturally nuanced view absent in prior Western-centric research.

Practical implications center on equipping language teachers with an efficient diagnostic tool that detects overt and subtle disengagement, including social withdrawal. Teachers can leverage the scale to tailor tasks that balance cognitive demand, relevance, and appeal, thereby sustaining attention and participation. Gender-sensitive pedagogy is encouraged: problem-solving activities that heighten perceived competence for boys, and collaborative, feedback-rich environments that buffer negative affect for girls. Such targeted strategies are expected to enhance both emotional well-being and overall engagement.

Limitations were acknowledged: self-report measures risk common method and social desirability bias; a single time-point design precludes causal or developmental claims; convenience sampling restricts generalizability beyond the studied Chinese EFL context; and the exclusion of agentic engagement, though culturally justified, warrants empirical scrutiny. Future studies should triangulate self-reports with classroom observations and teacher ratings, adopt longitudinal tracking to capture trajectories, employ stratified sampling across diverse CHC regions, and test the viability of agentic engagement within these cultural settings.