Social Sciences

Abstract: The digital transformation of higher education is central to sustainable and equitable international education, as reflected in Sustainable Development Goals 4, 5, and 10. Yet gender equity research in digital education has focused on access and technology acceptance, leaving unknown whether male and female international students differentially evaluate the educational technology services their host universities deliver. This study analyzed a cross-sectional survey of 329 international students in China (135 male, 194 female), who completed mirrored importance–satisfaction batteries for six educational technology services, online admission, diversified information access, timely website information, social media information, smart classrooms, and digital libraries, analyzed with Welch’s t-tests, effect sizes, paired tests, and HC3-robust regressions. Male students rated technology importance significantly higher than female students (4.46 vs. 4.29, p = 0.011, d = 0.278), while no statistically significant gender difference in experienced satisfaction was detected (p = 0.614, d = −0.057). Consequently, male students exhibited a larger negative expectation–perception gap (−0.48 vs. −0.27, p = 0.016), concentrated in information-intermediated services. Controlling for region of origin and family income attenuated the male coefficient to non-significance, although the gap persisted within the Asian subsample (p = 0.012). Because importance and satisfaction were measured concurrently, the importance ratings are interpreted as retrospective evaluative anchors, proxies for, rather than direct observations of, pre-arrival expectations. Within this interpretive frame, the pattern is consistent with expectation inflation without experience discrimination: gender appears to operate at the expectation-formation stage rather than at service delivery, although this mechanism requires longitudinal confirmation. The paper concludes with theoretical implications and managerial recommendations for universities and national policy makers, in the context of China’s national strategy of opening up education to the world: expectation calibration, realistic pre-arrival information and social-media promise management, is a lever for sustainable, gender-equitable digital education alongside interface quality.

Abstract: Written sentence construction is an important but underdocumented component of primary-level Mandarin literacy. This retrospective descriptive content analysis audited archived classroom responses to the conditional frame commonly taught as yaoshi–jiu (roughly, “if … then …”) in three Grade 6 classrooms in central Taiwan. The supplied source roster contained 71 records; two lacked a codable target response, yielding an analytic corpus of 69 responses. We preserved the inherited classroom coding matrix and reported its 39 teacher-identified code assignments rather than recoding the corpus as if it had undergone an independent reliability study. Grammar/structure was the largest macro-category (26 assignments, 66.7%), followed by semantic/referential issues (8, 20.5%) and logical/pragmatic issues (5, 12.8%). The most frequent subtype was omission, redundancy, repetition, or structural blending (21, 53.8%). De-identified Chinese source-image snippets and functional English translations illustrate how repeated wording, unstable perspective, and unspecified agency complicated the intended condition–consequence relation. Because the archive does not document a second independent coding cycle, the findings should be read as a transparent audit of teacher-identified revision targets in this corpus—not as population estimates, a causal evaluation of teaching, or a definitive account of Mandarin grammaticality. The study offers a cautious, traceable model for reporting classroom writing archives while protecting minors’ identities.

Abstract: This study aims to substantiate and evaluate an economic-mathematical model for de-veloping the open research competence of academic staff at Ukrainian universities align-ing academic publishing with European Standards. Data on open publications by Ukrainian institutions, the number of Ukrainian journals indexed in Scopus, and the au-thors’ proprietary metric of Open Science training activity for 2020–2025 were used. Line-ar programming and scenario analysis were applied to justify the resource distribution. The proposed model serves as a normative decision-support framework to evaluate re-source allocation priorities rather than an empirical causal-inference model. While publi-cation openness (% OA) acts as the primary outcome proxy, institutional presence (jour-nals in Scopus) and training activities represent enabling capacity-building factors. The share of open-access publications increased from 50.25% in 2020 to 63.64% in 2025. However, the number of Ukrainian journals in Scopus decreased from 202 in 2022 to 168 in 2025. The integral indicator of open research competence increased from 0.088 in 2020 to 0.816 in 2024 but decreased to 0.720 in 2025. The optimisation model showed that un-der the chosen weights and constraints, the optimal allocation is 45% to support OA pub-lications, 35% to Open Science training, and 20% to journal infrastructure. The proposed framework is a normative decision-support tool that translates a pre-specified set of Open Science indicators and policy weights into an optimisation model for resource allocation. It does not claim to establish causal links between institutional capacity and Open Science outcomes; rather, it provides a transparent, quantitative lens for prioritising limited re-sources among open-access support, journal infrastructure and researcher training.

Abstract: Generative artificial intelligence is becoming embedded in academic practice as students and institutions negotiate acceptable assistance, originality, authorship, and academic integrity. We analyzed Global ChatGPT Student Survey data to examine relationships among academic-integrity concerns, satisfaction, and perceived learning benefits across national academic settings. The source dataset contained 23,218 records from 108 identifiable countries; the primary multigroup analysis included 8,650 respondents from 15 countries meeting prespecified sample-size, completeness, and model-feasibility criteria. A nine-indicator model assessed the three constructs. Configural and full metric invariance were broadly supported, allowing comparison of structural associations; full scalar invariance was not supported, so latent means were not compared. Satisfaction was positively associated with perceived learning benefits in all 15 countries, whereas associations involving academic-integrity concerns were predominantly negative. Robust omnibus tests detected heterogeneity in path magnitudes, although equality constraints produced only trivial deterioration in global fit. A seven-country robustness analysis (N = 5,871) reproduced the directional pattern and heterogeneity evidence. By evaluating measurement comparability before cross-national structural comparison, the study shows that broadly shared relational patterns can coexist with contextual variation in emerging digital academic cultures.

Abstract: This article examines the historical development of the Islamic University of Rotterdam (IUR), today known as the Islamic University of Applied Sciences Rotterdam (IUASR), between 1997 and 2026, with particular attention to the relationship between its founding vision and subsequent institutional evolution. Using a qualitative historical case study and process-tracing approach, the analysis draws on founding statutes, conference discourse, curriculum documents, vision and mission statements, accreditation records, and institutional archives. The findings show that the university was conceived as a broad educational project integrating religious scholarship, scientific inquiry, citizenship formation, cultural engagement, professional education, and societal contribution. Its development did not follow a single linear trajectory. Instead, different elements of the founding vision evolved through realized, interrupted, re-activated, and projected pathways. Theology and chaplaincy became consolidated fields, several cultural and academic initiatives were interrupted, and teacher education re-emerged as an accredited pathway in 2026. The article conceptualizes this institutional orientation as a comprehensive Islamic university and proposes the Multiple Pathways of Vision Realization Model to explain how founding visions may continue to shape institutional continuity, diversification, and development across changing leadership, organizational, regulatory, and societal conditions.

Abstract: During the COVID-19 (-19) pandemic, young children were infected and became ill less frequently than adults. Nevertheless, children are regarded as one of the populations disproportionately affected by pandemic-related measures, particularly social isolation, which may have influenced their psychological well-being and overall development. However, the extent of these developmental impairments, especially in younger children, remains insufficiently understood.This study analyses and compares cohorts of young children (aged 4-6) with speech disorders before (2018/ 2019), during (2019–2021), and after (2021/2022) the pandemic. For all cohorts, extensive and repeated educational, logopaedic, and psychological assessment records covering a one-year period were available. Hypothesis: Pandemic-related measures resulted in behavioural, social-emotional, linguistic and developmental impairments in young children. Methodology: Cohorts of preschool children with speech disorders treated in a German speech therapy day-care centre (Sprachheilkindergarten) during the periods 2018/19 (n = 9) (A), 2019/20 (n = 8) (C-19) (B), 2020/21 (n = 8) (C-19) (C), and 2021/22 (n = 10) (D) were retrospectively analysed and compared using existing therapy documentation and validated assessment procedures. The following domains were examined: learning/performance, social-emotional behaviour, communication, speech fluency, articulation, and language comprehension. In total, 35 children aged 4–6 years across four cohorts were included in the analysis. Results: At admission to the speech therapy kindergarten, children in groups B and C (C-19, 2019–2021) demonstrated competencies comparable to those observed in groups A and D. All cohorts benefited from the therapeutic interventions provided in the kindergarten setting; however, children in groups B and C (C-19) exhibited smaller developmental gains compared to cohorts A and D.The strongest improvements were observed in the domains of learning/performance, communication and articulation, whereas no changes were found in social-emotional behaviour and speech fluency. Language development gains achieved through the therapeutic interventions were likewise reduced in groups B and C (C-19). Notably, all children showed markedly reduced linguistic abilities at the time of kindergarten admission. Conclusions: The educational and therapeutic measures implemented within the kindergarten setting appear to have substantially mitigated potential negative consequences of pandemic-related restrictions with respect to social-emotional behaviour. None of the children showed deterioration in the examined domains. Nevertheless, therapy-related improvements in groups B and C (C-19) were lower than those observed in the control cohorts A and D.Overall, the findings suggest a positive impact of educational support and highlight the importance of maintaining such measures during exceptional circumstances, including future pandemics.

Abstract: The objective of this study is to determine the relationship between costly prosociality toward known and unknown school peers and well-being, school aggression, and school climate in secondary school students, including the predictive capacity of costly prosociality regarding these variables. A correlational and predictive quantitative study was conducted using a sample of 529 Chilean secondary school students recruited through convenience sampling. Likert-type self-report scales were administered to measure costly prosociality, school aggression, well-being, and school climate. Correlation among the variables was measured and two predictive models of school climate were tested using structural equation modeling. Differences were found in the relationship among variables and in the predictive power of school climate, depending on the type of beneficiary who received help. In the discussion section, results are analyzed in light of the literature.

Abstract: STEM education increasingly asks students to work on real-world problems, yet ethics is still taught mostly through retrospective studies of past failures that position students as analysts after a decision. This case study examines what undergraduate engineering students express when they instead work on an unresolved, real-world public-safety problem: the illicit manufacture of machinegun conversion devices (MCDs) on consumer 3D printers, for which no design safety standard currently exists. We ask how five stu-dents across three roles in a multi-year design research program express their com-mitment to public welfare, and in what language. Five undergraduate contributors completed the recruitment survey (n = 5), semi-structured interviews (n = 4), and a pilot survey of public-welfare obligation adapted from the NSPE Code of Ethics (n = 5); re-sponses were analyzed with an a priori dictionary-based text analysis and integrated through a joint display. Students voiced their obligation through the language of care and purpose rather than the vocabulary of the professional code. We advance these results as hypotheses for a larger prospective study: that real-world problem solving elicits pub-lic-welfare commitment and may bear on the documented decline across engineering education.

Abstract: Generative artificial intelligence (GenAI) is now part of everyday study practices in higher education, yet routine use may become problematic when students delegate reasoning, evaluation, and task completion to AI tools. This study examines student AI dependency as a socio-technical systems outcome shaped by peer norms, AI-use behavior, psychological strain, trust, AI literacy, and educational context. Using a public cross-sectional survey of mainland Chinese university students (N = 860), we combined predictive modeling, SHAP explanation, and structural equation modeling. XGBoost outperformed random forest, extra trees, and K-nearest neighbors in predicting dependency. SHAP identified peer norms, daily AI-use time, academic stress, and trust in AI as the most influential predictors. SEM results supported a partial pathway from peer norms and trust in AI to dependency through usage intensity, together with a strong direct association between academic stress and dependency. AI literacy did not significantly moderate the usage–dependency path in the full sample, although institutional and disciplinary differences were observed. The findings identify system-level leverage points for course design, assessment, student support, and responsible GenAI guidance in higher education.

Abstract: Despite the significant role of hormonal health in adolescent well-being, substantial cycle education gaps remain in how menstrual cycle education is standardized and delivered. This literature review evaluates the current landscape of comprehensive, evidence-based menstrual cycle curriculums for adolescents to determine what instructional frameworks are available and have been effective. Findings highlight a shortage of comprehensive, evidence-based curriculums that address overall cycle health and well-being for adolescents and stigma around discussing menstruation. Comprehensive cycle education integrates biology with cycle-tracking and the psychosocial impacts of the menstrual cycle. This review attempts to map out the strengths and deficiencies identified in current programming and resources in K-12 public schools in the United States and community-based educational curriculums and discusses digital interventions that have emerged independently to provide foundational resources for educators and curriculum developers aiming to implement holistic cycle health education programs for adolescent populations.

Abstract: Professional internships constitute a privileged space for articulation between universities and educational establishments; however, their evaluation has traditionally focused on the learning of the future teacher, paying less attention to the learning and transformation processes that interaction can generate in collaborating teachers, universities and schools. The purpose of this study was to analyze how the literature has conceptualized and evaluated bidirectionality in professional practices of initial teacher training, identifying its conceptual evolution, the mechanisms by which it is produced, and the evidence used to recognize its contributions and impacts. A systematic review of 71 studies published between 1990 and 2026 was conducted. The analysis shows an evolution from models focused on supervision and knowledge transfer to perspectives based on functional reciprocity, mutual learning, knowledge co-construction and institutional transformation. Based on these results, a taxonomy of six levels of maturity is proposed—presence, coordination, functional reciprocity, mutual learning, epistemic co-construction, and institutional transformation—and mechanisms that favor bidirectionality are identified, including structured professional dialogue, joint action, hybrid spaces, shared objects of work, and the agency of the future teacher as a mediator of knowledge. Likewise, eight guiding dimensions are proposed for its evaluation: enabling conditions, relational symmetry, reciprocity of the flow of knowledge, mutual learning, co-construction, contributions, institutionalization and impact. The results show that the available evidence focuses on learning and contributions, while the demonstration of sustained and institutional impacts remains limited. The review thus provides a conceptual framework to understand bidirectionality as a process that can be classified and evaluated empirically, rather than as an inherent property of any university-school relationship.

Abstract: This article introduces Innovation-Based Coaching (IBC) as a strategic framework to bridge the growing divide between AI-driven work and the predominantly content-focused nature of education. Drawing on insights from AI in education, coaching, innovation management, and employability studies, IBC uses a 3C mindset, a 3D process, and a 3M toolkit to cultivate an innovative mindset, build innovative skills, and produce innovative outputs. The 3C mindset emphasizes being Curious, Critical, and Creative; the 3D process involves Detecting, Dissecting, and Discovering to translate complex problems into targeted coaching actions; and the 3M toolkit incorporates Mapping, Measuring, and Monitoring for ongoing assessment and scaling. This approach positions AI as a collaborative partner that supports feedback, simulation, and experimentation while preserving human-centric skills such as ethical judgment, trust, and critical reflection. The article illustrates IBC in action through a case in a course, where students use AI to explore real-world issues and create tangible innovation artifacts. It also addresses major challenges, including conceptual ambiguity, cultural resistance, workload and incentive misalignments, equity concerns, and assessment difficulties, and offers practical recommendations for institutions, educators, and policymakers. The conclusion proposes a research agenda focused on IBC’s long-term impact on innovation, employability, and AI literacy, emphasizing coaching as a systemic tool to transform education in the AI era.

Abstract: Artificial Intelligence (AI) is increasingly recognised as a transformative enabler of Sustainable Development Goal 4 (SDG 4): Quality Education. This study uses systematic review to synthesis and critically evaluates the growing body of literature on how AI particularly large language models and content-generation systems in advancing equitable access to education, enhancing the quality of learning outcomes, and promoting inclusive, lifelong learning within diverse and resource-constrained contexts. The findings indicate that AI supports SDG 4 through personalised and adaptive learning, intelligent tutoring, automated assessment and feedback, multilingual content generation, and teacher support, with notable benefits for underserved and resource-constrained contexts. The evidence reveals an uneven distribution of research and implementation, with a strong emphasis on higher education and limited empirical work in early childhood, primary education, and low-income regions. This review contributes a structured mapping of AI applications to SDG 4 and highlights critical research gaps. It concludes that while AI holds substantial promises for advancing quality education, its impact depends on inclusive design, robust policy frameworks, ethical governance, and sustained investment in digital capacity building.

Abstract: Institutional evaluation in higher education is increasingly shaped by multiple, coexisting systems, including research assessment, university rankings, accreditation and quality assurance, and responsible research assessment. Although these systems are widely examined individually, their dynamic interactions and cumulative institutional effects remain insufficiently understood. This study develops a conceptual System Dynamics framework to explain how multiple evaluation logics interact through reinforcing and balancing feedback processes. Drawing on institutional logics, organizational response, research evaluation, quality assurance, and System Dynamics perspectives, the framework integrates four evaluation subsystems: research prestige, competitive ranking, accreditation and quality assurance, and responsible assessment. Their interaction generates cross-system feedback mechanisms involving evaluation internalization, reputation and resource attraction, capability development, internationalization, strategic alignment, diversification, and organizational sustainability. The study further introduces Sustainable Institutional Capacity (SIC), defined as the level of evaluation-oriented activity that an institution can sustain without excessive organizational strain, quality deterioration, mission displacement, or capability erosion. By linking evaluation pressures to endogenous feedback and capacity constraints, the framework explains how initially beneficial evaluation-driven strategies may produce growth, adaptation, stabilization, or overextension as feedback dominance changes over time. The framework provides a theoretical basis for empirical validation and subsequent stock-and-flow simulation of institutional evaluation dynamics.

Abstract: Generative AI challenges the evidence that universities use to certify general education learning. A polished final product can no longer show, by itself, that a student can reason, write, evaluate information, use data, or make responsible judgments. This paper argues that general education should move from course completion alone to a competency-and-practice model. In this model, students first demonstrate core abilities without AI assistance. They then use AI in structured assignments where they must disclose use, verify outputs, revise results, and explain their decisions. The paper applies this model to thinking, communication, quantitative and data reasoning, AI and information literacy, ethical and civic judgment, and integrative learning. It also proposes a dual-condition assessment design that separates independent competence from documented AI-assisted performance.

Abstract: University students’ adoption of generative artificial intelligence has become nearly universal, yet institutional guidance has not kept pace, leaving responsible use dependent on students’ own self-regulation. Drawing on social cognitive theory, this study tested whether self-regulated learning mediates the effect of AI literacy on the responsible use of generative AI, and how technostress influences that process. A cross-sectional survey of 820 Peruvian undergraduates from public and private universities was analyzed using covariance-based structural equation modeling; a two-stage moderated mediation model with latent interactions was estimated using latent moderated structural equations. Measurement invariance held across gender and university type, and common method variance was negligible. AI literacy predicted self-regulated learning (β = .626) and responsible use (β = .243), and self-regulated learning predicted responsible use (β = .445). The indirect effect was significant, β = .278, 95% CI [.226, .337], accounting for 53.4% of the total effect. Technostress did not moderate either the first stage or the direct path, but it significantly attenuated the second stage (b₂ = −.211, p < .001): the conditional indirect effect fell from .344 to .111 across moderator levels and ceased to differ from zero above 1.40 SD. Technostress does not prevent students from acquiring self-regulation; it prevents them from putting it into practice.

Abstract: Distance education is increasingly important for mass education because it expands access, supports disadvantaged groups, and creates more equal learning opportunities across society. It is also critical for lifelong learning and broader social transformation, especially in a period when education is closely intertwined with technological change. Yet despite the strategic importance of this education-technology nexus, transformative research in distance education has not developed at a level commensurate with its importance, and the field still lacks a comprehensive synthesis of its evolution. This systematic literature review examines 119 Scopus-indexed studies published between 1993 and 2026. Combining bibliometric analysis with thematic analysis, the review maps the field’s growth, diversity, and major lines of inquiry. The findings show that transformative research in distance education is increasingly influential, while also revealing persistent conceptual and methodological gaps. The study therefore provides a timely foundation for clarifying concepts, identifying trends, and strengthening future research in a field with major educational and societal implications.

Abstract: Artificial intelligence (AI) has emerged as one of the most transformative technologies of the twenty-first century, reshaping scientific discovery, academic research, knowledge dissemination, teaching, learning, and institutional operations. In higher education, however, AI presents a unique institutional challenge. Universities have played a central role in creating modern AI technologies through their research, scholarship, and educational activities, yet they are also among the institutions most responsible for regulating AI use, evaluating its outputs, and safeguarding the academic values most affected by it. This paper conceptualizes this tension as the AI paradox in higher education. Drawing on recent empirical research, systematic reviews, policy frameworks, and governance initiatives, the paper examines both the opportunities and challenges associated with AI adoption across teaching, research, and learning. It argues that the central challenge facing higher education is not whether AI should be embraced or restricted, but how universities can reconcile their dual roles as innovators and guardians of academic integrity, intellectual originality, scientific rigor, and educational quality. To address this challenge, the paper proposes three integrated governance frameworks for educators, researchers, and students. The educator framework positions AI as a source of content and a partner in learning design while maintaining full human accountability for educational outcomes. The research framework permits AI-assisted ideation and dissemination support while requiring transparent disclosure, rigorous validation, and auditable evidence of human intellectual contribution. The student framework recognizes AI as a legitimate learning resource while maintaining that assessment should remain aligned with intended learning outcomes and authentic student achievement. The paper concludes by presenting an institutional roadmap for responsible AI governance grounded in human accountability, AI literacy, transparency, and values-based stewardship. Together, these contributions provide a principled approach for navigating the AI paradox while enabling responsible innovation across higher education.

Abstract: Definition The term high-readability, dyslexia-friendly font refers to a typeface specifically designed or configured to reduce the cognitive effort associated with reading, thereby improving text accessibility for individuals with dyslexia without compromising overall readability. Such typefaces aim to facilitate the accurate mapping between graphemes and phonemes during the reading process. In addition, for individuals with dysorthographia, they may support the inverse process by facilitating the encoding of phonemes into their corresponding graphemes during writing. Abstract This entry examines the pedagogical relevance of typographic design as a compensatory strategy for improving accessibility in reading and writing for individuals with dyslexia and related Specific Learning Disorders (SLDs). A high-readability, dyslexia-friendly font is defined as a typeface designed to reduce the cognitive effort associated with reading while facilitating the mapping between graphemes and phonemes, without compromising readability for the general population. Grounded in the principal cognitive and neurophysiological models of dyslexia, this work analyzes how specific typographic features can mitigate cognitive load and support more efficient text processing. Cross-linguistic considerations further emphasize the need for culturally and linguistically responsive design solutions. Based on the evidence reviewed, a user-centered workflow for the development of high-readability, dyslexia-friendly fonts is proposed. Although specialized typefaces such as OpenDyslexic have introduced innovative design principles, current empirical evidence does not conclusively demonstrate their superiority over carefully designed and appropriately configured conventional fonts. The findings highlight the importance of typography as a key component of inclusive design and educational accessibility, reinforcing its role in reducing learning barriers, promoting equitable access to written information, and supporting more inclusive digital learning environments. Drawing on cognitive, neuropsychological, anthropological, and technological perspectives, the work examines how reading, as a culturally acquired technology, interacts with human cognitive processes. It analyzes the potential of high-readability and dyslexia-friendly fonts and it discusses the integration of typography with Artificial Intelligence and adaptive learning technologies to develop personalized and inclusive educational environments.

Abstract: In order to reduce teachers' correction burden and improve teaching efficiency, this paper first establishes a database of CLIL teaching translation assignments and analyzes the characteristics of their translation assignments in CLIL teaching environment. LDA category feature extraction based on TF-IDF is used to improve the quality of category features. Then, N-gram subword filtering method based on lexical information entropy is adopted in order to filter out the subwords with low contribution to category differentiation in N-gram subwords. Finally, the model of automatic correction of instructional translation using named entity information extracted from text and N-gram to represent documents is constructed. Through empirical analysis, the category 1 dataset varies between 85.59-92.14 and the category 2 dataset varies between 84.15-3.98, which can effectively capture the collocation relationship between words. And the students' satisfaction with the automatic correction model was generally high for the four translation datasets in the three classes. The satisfaction scores all reached 4-5, and the students believed that the auto-critique model was able to give quick feedback, which helped to identify and correct errors in time.