As digital technologies increasingly shape learning environments in higher education, understanding how institutional support is associated with students’ digital citizenship has become an important research issue. Drawing on Self-Determination Theory (SDT), this study examined the relationships between perceived school support (PSS) and digital citizenship literacy (DCL) among 600 university students in China. The measurement model demonstrated satisfactory reliability and validity. Results showed that PSS was positively associated with DCL, with competence need satisfaction serving as the only significant mediator. The non-significant mediating effects of autonomy and relatedness should be interpreted cautiously, as they may reflect the domain-specific nature of DCL, the operationalization of basic psychological needs, or contextual characteristics of Chinese higher education rather than their theoretical irrelevance. Theoretically, this study extends the application of SDT to digital citizenship by highlighting the differentiated roles of the three basic psychological needs. Practically, the findings suggest that supportive digital learning environments may strengthen students’ digital competence while encouraging reflective engagement and responsible digital participation. These implications are consistent with SDG 4’s emphasis on educational quality and inclusive learning environments. 1. Perceived school support is positively associated with digital citizenship literacy. 2. Competence need satisfaction mediates this association. 3. Autonomy and relatedness show no significant mediating effects. 4. Informal school support may complement formal digital citizenship education. 5. Evidence is drawn from 600 Chinese university students.
While identity construction is recognized as an integral part of higher education, particularly for interdisciplinary students who need to negotiate their identities between two markedly different disciplines, relevant research remains limited. Moreover, existing research on interdisciplinary learners’ identity tends to regard participants as a unified collective and gives insufficient attention to variations across grade levels. Using Q methodology and follow-up interviews, this study compares identity profiles of first- to fourth-year students in a Chinese “English + Engineering” dual degree program, revealing between-group differences and key influences on their identity construction. Findings show that first- and second-year students exhibit dual English-Engineering and engineering-oriented identities. Third-year students hold three profiles, including engineering-centric career preparers, unsettled identity navigators, and a small group of English-centered learners. Fourth-year students are predominantly engineering-dominant, with a smaller subset of unsettled navigators. Personal interest, curriculum fragmentation, career planning, and family background emerge as primary drivers. These findings also enable this study to propose some practical implications for curriculum design and student support within interdisciplinary programs.
With Generative Artificial Intelligence (GAI) deeply integrated into education, teachers serve as the core agents bridging technology and pedagogy. Teachers’ AI literacy (TAIL) is pivotal in facilitating teaching innovation, cultivating innovative talent, and driving educational transformation. Yet, existing research remains inconclusive regarding which technology-related psychological factors underlie the systematic association between TAIL and Teachers’ Teaching Innovation Behaviors (TTIB). Based on the 3P model, this study constructed a multiple mediation model and employed structural equation modeling (SEM) to analyze questionnaire data from 528 primary and secondary school teachers in China. The results revealed that TAIL was significantly and positively associated with TTIB, GAI-related Technology Self-Efficacy (GAI-TSE), GAI-induced Technology Stress (GAI-TS), and GAI-driven Cognitive Reconstruction (GAI-CR). Additionally, the relationship between TAIL and TTIB was mediated not only by GAI-TSE and GAI-CR, but also by two serial mediation pathways: GAI-TSE→GAI-CR and GAI-TS→GAI-CR. These findings clarify the multifaceted psychological pathways linking TAIL and TTIB, highlighting the prominent role of GAI-TSE within the model. This study not only deepens our understanding of the hierarchical processing logic intrinsic to the 3P model, but also provides robust empirical evidence for the design and delivery of targeted GAI training programs and teacher support systems. Reveals dual pathways of teachers' AI literacy affecting teaching innovation (China). Breaks 3P model's parallelism flaw; highlights Emotional-Cognitive pathway effect. Deepens differentiated cognition of psych mediation roles. Proposes 3D strategies for synergizing teachers' AI literacy and teaching innovation.
The purpose of the study is twofold: first, to describe the characteristics of classroom discourse and, second, to identify and explore the different interactive sequences that can be captured with a sequential statistical analysis. The purpose of the study is twofold: first, to describe the characteristics of classroom discourse and, second, to identify and explore the different interactive sequences that can be captured with a sequential statistical analysis. This study investigated the characteristics of teacher–student activities in English-speaking lessons within the specific context of Vietnamese secondary schools. Data were collected through classroom observations and video recordings of 26 lessons (1157 min) across six public secondary schools to capture a rich description of current pedagogical practices. Specifically, the study employs Epistemic Network Analysis, underpinned by the Initiation-Response-Feedback framework, as a way of providing a nuanced mapping of pedagogical strategies. The results indicate that a teacher-centric instructional model remains prevalent, with over half of the total lesson time characterized by teacher dominance. While teachers primarily engaged in explaining and reinforcing content, students were observed to participate mainly through recalling prior knowledge and summarizing information, suggesting a limited scope for communicative interaction. These patterns suggest that a traditional Initiation-Response-Evaluation pedagogical framework continues to exert a considerable impact on classroom practice, despite the national orientation toward more communicative approaches. Furthermore, a notable difference was found between the epistemic networks of early-career and experienced teachers. While experienced teachers utilized more diverse feedback strategies, such as conceptual reinforcement, early-career teachers showed a tendency to rely on narrow, recall-based quizzes, a finding that highlights the pressures and constraints shaping novice teachers’ early experiences in the Vietnamese EFL context. Therefore, this study sheds more light on the complexities of classroom interaction and provides findings that are valuable for informing targeted professional development and teacher training programs designed to enhance students' speaking competence.
The rapid integration of generative artificial intelligence (GenAI) into higher education has raised critical questions regarding cross-cultural variability in adoption and ethical perception. This study investigates university students’ GenAI use across three distinct socio-technical contexts: China (state-driven technological ecosystem), Ireland (EU regulatory framework), and Kenya (emerging digital economy). Grounded in a multi-dimensional contextual framework integrating an adapted Technology Acceptance Model (TAM), we examined how national infrastructures, educational pressures, and governance discourses shape student engagement. A mixed-methods survey was administered simultaneously to 580 undergraduates and postgraduates. Multivariate analysis of variance revealed significant cross-national differences in usage frequency, perceived academic impact, and ethical concerns (p < 0.001). Chinese students reported one of the highest usage frequencies (alongside Kenyan students) but the lowest ethical concerns, utilizing predominantly local platforms (e.g., DeepSeek). Irish students demonstrated moderate usage with heightened ethical vigilance regarding academic integrity. Kenyan students exhibited highest AI familiarity despite resource constraints, leveraging AI to bridge educational gaps. Findings demonstrate that GenAI adoption is fundamentally mediated by contextual factors rather than cultural typologies alone. We propose context-sensitive policy frameworks that move beyond universalist approaches to AI integration in higher education. This study contributes empirical evidence for culturally responsive AI governance in diverse educational ecosystems. Comparative study of GenAI adoption across Asia–Pacific, Europe, and Africa. Significant cross-national differences in usage, impact, and ethics. China has the highest usage, low ethics concerns and uses mostly DeepSeek. Ireland has a moderate usage, high ethics concerns and uses mainly ChatGPT. Kenya has the highest familiarity despite constraints, leveraging AI to bridge gaps. Contextual factors mediate adoption more than cultural dimensions.
This study develops a teacher AI literacy scale comprising six dimensions: perception of AI, AI knowledge, AI skills, applications of AI, innovation of AI, and AI ethics. The research involved three sequential phases. In the first phase, a Random Forest Model (RFM) was employed to select 30 representative items from an initial pool of 60 questions. The second phase focused on psychometric validation using the Rasch model. Results demonstrated that the overall scale met psychometric standards, with all item fit indices falling within acceptable ranges. In the third phase, the validated scale was applied to a sample of 658 teachers using Latent Class Analysis (LCA). The analysis revealed five distinct latent classes of AI literacy among teachers: AI Pioneers, Practical Teachers, Developing Teachers, Emerging AI Teachers, and Traditional Teachers. In sum, the scale is promising but still requires cross-cultural validation and predictive-validity checks.
This study examines Chinese international students’ continuance intention to use AI systems (e.g., EAP Talk) for English speaking practice by extending the Expectation–Confirmation Model (ECM) with selected dimensions of acculturative stress. Using a sequential explanatory mixed-methods design, survey data from 245 students in UK higher education were analysed with structural equation modelling, followed by interviews with 16 participants. Results show that confirmation, perceived usefulness, and satisfaction significantly predict continuance intention, supporting traditional ECM pathways. Language insufficiency, social isolation, and academic stress also positively influence continued use. Qualitative findings indicate that AI systems serve as low-risk, compensatory resources when students experience language-related, social, and academic adaptation pressures, while sustained use depends on alignment with learners’ evolving needs. The study highlights the value of contextualising technology continuance within international students’ acculturative stress experiences.
Argumentative writing is a critical indicator of English as a Foreign Language (EFL) writing proficiency. Nevertheless, EFL learners often encounter difficulties in producing effective argumentative essays. Providing students with detailed diagnostic feedback through a structured checklist may offer a useful means of improving their argumentation skills. Accordingly, this study aims to develop and validate a diagnostic checklist specifically designed to assess argumentation in the argumentative writing of EFL undergraduates. The checklist was developed on the basis of Toulmin’s Argument Model and the existing literature on argumentative writing. It was subsequently refined through feedback from six EFL experts and three raters, together with an analysis of student essay samples, resulting in a final 5-point checklist consisting of 14 items with four dimensions. Each item was formulated as a single-sentence descriptor targeting a distinct aspect of argumentation competence. The study employed Partial Least Squares Structural Equation Modeling (PLS-SEM), Many-Facet Rasch Model (MFRM), and Pearson correlation analysis to examine the checklist’s reliability, structural validity, rater severity and consistency, and criterion-related validity. Additionally, qualitative analysis of student perceptions further support for the checklist’s effectiveness in providing detailed diagnostic feedback and promote students’ argumentation skill development. The findings underscore the pedagogical potential of the checklist for supporting both EFL learners and instructors in the learning and teaching of argumentative writing, with implications for classroom practice and further research.
Previous research has focused more on the positive effects of teachers’ professional identity on their professional growth and well-being, with limited attention paid to its antecedents. This study aimed to examine the significant role of authentic leadership in enhancing teachers’ professional identity by empirically testing the relationship between authentic leadership and teachers’ professional identity, as well as revealing the mediating roles of knowledge sharing and self-efficacy between them, based on the conservation of resources theory. A total of 4101 Chinese elementary and secondary school teachers were surveyed for this study. Data analysis was conducted using a mixed-methods approach combining covariance-based structural equation modeling (CB-SEM) and artificial neural networks (ANN). The results indicated that authentic leadership was significantly and positively correlated with teachers’ professional identity, knowledge sharing, and self-efficacy. Knowledge sharing was significantly and positively associated with both teachers’ professional identity and self-efficacy. Self-efficacy was also significantly and positively associated with teachers’ professional identity. Knowledge sharing and self-efficacy not only served as independent mediators between authentic leadership and teachers’ professional identity but also played a chain-mediating role. Furthermore, ANN analysis indicated that self-efficacy exhibited the strongest predictive power for teachers’ professional identity, followed by authentic leadership and, finally, knowledge sharing. This study not only elucidates the mechanisms through which authentic leadership influences teachers’ professional identity, thereby offering a leadership perspective for enhancing teachers’ professional identity, but also provides practical strategies for school administrators.
Although the integration of generative artificial intelligence (AI) into second language (L2) writing has drawn increasing attention to its multifaceted influence on the writing process and behaviors, few studies have examined the antecedents of writing procrastination in AI-mediated environments, particularly in non-English learning contexts. To address this gap, this study, grounded in self-efficacy theory, employed an explanatory sequential design to investigate how AI-assisted writing use interacts with writing self-efficacy (linguistic, performance, and self-regulatory efficacy) to influence writing procrastination among learners of Chinese as a second language (CSL). Quantitative findings revealed that AI-assisted writing use positively predicted CSL writing task-initiation procrastination and the three dimensions of writing self-efficacy. Self-regulatory efficacy significantly predicted writing procrastination and mediated the relationship between AI-assisted writing use and CSL writing procrastination, whereas linguistic and performance efficacy did not play similar roles. Qualitative findings further revealed execution-phase inefficiencies, extending beyond initiation-related delay, and deepened the interpretation of these results by suggesting that while AI use fosters confidence and perceived control, it can also induce illusory efficacy, a false sense of mastery that increases delay, and trigger emergent self-regulatory tensions marked by uncertainty, hesitation, and inefficient task progression. This study extends self-efficacy theory to the CSL writing in AI-mediated environments, providing important pedagogical implications for L2 writing.
Drawing on self-determination theory, this study investigates how basic psychological needs (BPNs)—autonomy, competence, and relatedness—mediate the relationship between AI-mediated informal digital learning of English (AI-IDLE) and learner engagement. A total of 672 university students from wide-ranging regional areas in central, northern, and southwestern China provided data for evaluation through structural equation modeling. The results revealed that: (1) AI-IDLE positively predicted engagement and all three BPNs; (2) competence and relatedness, but not autonomy, emerged as significant predictors of engagement; and (3) competence and relatedness partially mediated the relationship between AI-IDLE and engagement, whereas autonomy did not function as a mediating conduit. Suggestions for sustaining learner engagement during informal learning activities have been presented.
Many schools choose to use ability grouping, despite longstanding evidence that it is inequitable and does not improve overall academic outcomes. One reason that the practice of grouping students by ‘ability’ continues is because of pressure to perform on standardised tests. Educators hold beliefs that grouping students for English classes, for example, will improve their literacy scores. This paper presents findings from a study about class ability grouping for English in Australia and how different class grouping practices relate with students’ achievement on Australia’s standardised literacy test. The findings draw on data from our Class Ability Grouping Survey, where participant principals or their delegates characterised the different English class grouping practices being used in their schools from flexible to rigid. Multivariate linear regression modelling was used to explore if any of the types of grouping for English for Year 7–9 were predictive of student scores on standardised literacy tests. The results demonstrate an absence of consistent evidence that class ability grouping in English, regardless of form, reliably predicts student performance on test scores. These findings are significant for challenging mistaken beliefs that class ability grouping improves student literacy scores in Australia, providing new knowledge about varying class grouping practices through the example of Australia.
Sustainable university-school partnerships are increasingly recognized as critical for teacher development and pedagogical innovation; however, existing STEM partnership models often privilege technical expertise over ethical, cultural, and human-centered dimensions of learning. Addressing this gap, this study conceptualizes humanizing STEM education through an expanded Scientist-Teacher-Student Partnership (STSP) framework that integrates STEM and humanities perspectives within secondary science learning. Grounded in humanizing pedagogy and partnership theory, the study draws on in-depth semi-structured interviews with 22 interdisciplinary experts comprising secondary science and humanities teachers, scientists, and humanities scientists, alongside document analysis of curriculum standards and policy documents. Using reflexive thematic analysis, three interrelated constructs were identified: integrative collaboration, contextualized learning, and pedagogical strategies for humanizing STEM practice. Findings reveal that sustainable interdisciplinary collaboration enables teachers to embed ethical reasoning, cultural meaning, and socio-scientific relevance within STEM instruction, thereby strengthening professional capacity and pedagogical innovation. The expanded STSP model reconfigures university-school relationships from serialized engagement toward a sustained partnership that supports teachers’ long-term professional learning. This study contributes to learning sciences scholarship by demonstrating how humanizing STEM can be operationalized through sustainable university-school partnerships that position humanities disciplines as partners in STEM education, with implications for teacher development, curriculum reform, and values-oriented pedagogy.
Willingness to communicate (WTC) with generative artificial intelligence (GenAI) voice chatbots is pivotal for enhancing second language (L2) speaking skills. Guided by MacIntyre et al.’s pyramid model, we investigated how dark personality traits (Machiavellianism, Narcissism, Psychopathy) and regulatory focus (promotion vs. prevention) shape L2 learners’ WTC with GenAI. Data were collected from 2671 college students in China and analyzed using structural equation modeling (SEM) in AMOS 24 and fuzzy-set qualitative comparative analysis (fsQCA) in fsQCA 4.1. SEM revealed that Machiavellianism and Narcissism positively predicted WTC both directly and indirectly via promotion focus, with direct effects stronger than indirect ones, while Psychopathy exerted a direct negative effect. Prevention focus showed no significant independent influence. fsQCA identified five configurations linked to high WTC: Pragmatic Opportunist, Strategic Self-presenter, Cautious Achiever, Equipped Achievement Seeker, and Positive Performer. Across these, Machiavellianism and promotion focus consistently emerged as core drivers, with prevention focus interacting with promotion focus in certain contexts to strengthen WTC. Our findings elucidate the complex interplay between personality and motivation in GenAI-empowered settings. They underscore the need for personalized, motivation-sensitive instructional designs that leverage learners’ psychological profiles to foster effective L2 communication.
With the integration of generative artificial intelligence into EFL speaking instruction, AI-generated feedback (AIGF) has diversified in terms of interactivity and the extent of embedded teacher expertise. Although the pedagogical value of AIGF is well established, limited research has examined how these two dimensions jointly shape learners’ speaking development and willingness to communicate (WTC). Grounded in the Interaction Hypothesis, this quasi-experimental mixed-methods study investigated the effects of three AIGF modes differing in interactivity and embedded teacher expertise over a four-week intervention. Sixty-two Chinese university students were assigned to one of three feedback modes: (1) EAP Talk, a low-interactivity, assessment-oriented system integrating teacher-informed feedback through structured scoring and diagnostic guidance; (2) Doubao, a high-interactivity, open-ended dialogic system emphasising learner-initiated expression and meaning negotiation; and (3) Doubao Agent, a moderate-interactivity, pedagogically guided dialogic system balancing instructional scaffolding with conversational flexibility. Pre- and post-tests, together with measures of WTC and technology acceptance, showed that EAP Talk was particularly effective in improving grammatical accuracy and in-class WTC. Weekly structured reflective responses further suggested EAP Talk supported a feedback–revision–repractice pattern, characterised by problem noticing, output revision, and repeated practice. In contrast, Doubao supported idea generation and fluency in a low-pressure dialogic environment, while Doubao Agent demonstrated a partial balance between form-focused guidance and interactional flexibility. Overall, the findings suggest that interactivity alone does not determine effectiveness; rather, learning outcomes depend more critically on the alignment between feedback design, pedagogical structuring, and embedded teacher expertise.
This study examines how university foreign language teachers in Chinese higher education negotiate occupational well-being under AI-enhanced conditions. Existing research has often conceptualised teacher well-being as a relatively stable psychological outcome, but such framings are less well suited to contexts marked by rapid socio-technical change. Drawing on constructivist grounded theory and in-depth interviews with 23 teachers, the study develops a process-oriented account of how occupational well-being is repeatedly made workable, disrupted, and partially re-stabilised over time. The findings suggest that well-being was negotiated through recurrent cycles of appraisal of AI-related change, legitimacy repair, mobilisation of psychological, relational, and digital supports, and tactical adaptation under mediated institutional and discursive conditions. These negotiation cycles were shaped by two interconnected forms of socio-technical mediation: institutional mediation, including evaluation regimes, platform reliability, training provision, and workload organisation, and broader discursive climates concerning automation, replaceability, and the value of language-related work. The analysis also identifies temporal consolidation as the partial stabilisation of repeated negotiation cycles through the development of workable routines, interpretive boundaries, and restored professional control. Grounded in the experiences of university foreign language teachers in Chinese higher education, the study reconceptualises occupational well-being as a negotiated and temporally unfolding accomplishment rather than a fixed endpoint.
Educational digital transformation is a global consensus and national strategy, with principal digital leadership key to school transformation success. Existing Western-rooted assessment tools lack adaptability to Chinese K-12 principals’ unique responsibilities and local contexts, and sufficient validation. Based on grounded theory, this study extracted four core dimensions (digital strategic deployment capability, digital ecosystem building capability, digital humanistic care capability, and digital technology mastery capability) via interviews with principals from 15 Smart Education Model Schools, developed an 18-item DL-CPSPLS through EFA and CFA. With satisfactory reliability and validity, the scale integrates scientific rigor and local adaptability, filling the methodological gap in relevant quantitative research and holding significant theoretical and practical value.
According to the situated expectancy-value theory (SEVT), reading achievement can be driven by two key motivational factors (expectancy and task value) and the interaction between them. Existing studies on reading have mostly focused on the main effects at the expense of insufficient attention to interaction. Besides, interaction has merely been operationalized as the product term between expectancy and value (i.e., expectancy multiplying value, Type 1 interaction), neglecting the relative (or comparative) importance of expectancy and value (i.e., the competition between them, Type 2 interaction) in determining reading achievement. To address these gaps, the current study examined the main effects of and the two types of interaction between two expectancy-value factors, namely, reading enjoyment (indicating value) and self-concept (indicating expectancy), in predicting reading achievement. We used the Program for International Student Assessment (PISA) 2018 data generated by 532,835 students (Mean age = 15.79, SD = 0.29, 51
This study explores the effects of involvement load, L2 motivation, and their association with L2 vocabulary learning. Non-English majors from a northern China university (N = 114) were randomly assigned to three experimental groups and one control group. The three experimental groups participated in three tasks with different involvement load indices. Two types of word knowledge (word recognition and word recall) were measured to examine the treatment effects. Participants completed an L2 motivation questionnaire measuring their L2 motivation. The results of regression analysis from the three experimental groups provided partial support for the Involvement Load Hypothesis (Laufer Hulstijn, 2001). A moderation analysis of motivation variables and word knowledge across different involvement-load tasks showed that one component of L2 motivation moderated the effects of two task conditions, revealing an association between involvement-load tasks and L2 motivation.
Teachers are the direct practitioners of integrating various technologies into teaching, yet insights from the perspectives of principal leadership and teachers’ psychological needs remain insufficient. Grounded in Organizational Support Theory and Self-Determination Theory, this study utilizes data from 1286 teachers in the Shanghai region of China from the TALIS 2024 database to explore the impact of principal leadership on teachers’ ICT self-efficacy and digital literacy. The findings reveal that principal leadership not only directly and positively predicts teachers’ ICT self-efficacy and digital literacy but also exerts significant indirect effects through teacher autonomy and teacher collaboration. The results indicate that within the educational system, principals satisfy teachers’ psychological needs through the dual pathways of autonomy and belonging by empowering teachers and fostering a collaborative atmosphere, thereby stimulating their digital teaching practices. This study provides a new theoretical perspective for understanding the mechanisms of school leadership in the context of digital transformation and offers empirical evidence for education policymakers and school administrators in promoting the integration of digital technologies.
