Paedagogia: Jurnal Pendidikan

AI-TPACK, Pedagogical Ethics Awareness, and Teaching Readiness among Accounting Teacher Professional Education Students: Evidence from an Universitas Negeri Medan

2026-09-29T14:51:18+07:00

Teacher professional education is increasingly expected to prepare graduates who can integrate generative artificial intelligence into classroom practice both competently and responsibly. Conventional technological pedagogical content knowledge frameworks were theorised around passive digital tools rather than generative systems, and empirical evidence on artificial intelligence specific teacher competencies within Indonesian teacher professional education remains limited. This study specifies and tests a model in which artificial intelligence literacy, prompt engineering skill, self-efficacy, and constructivist teaching beliefs are associated with teaching readiness through two sequential mediators: artificial intelligence integrated pedagogical content knowledge and pedagogical ethics awareness. A cross-sectional survey was administered to 128 students in the Accounting Education teacher professional education programme at Universitas Negeri Medan, selected purposively on the criterion of prior generative artificial intelligence use. Data were analysed with partial least squares structural equation modelling using five thousand bootstrap subsamples; effect sizes and ninety-five per cent bias-corrected and accelerated confidence intervals are reported for every path, and the reflective specification of each construct was examined through confirmatory tetrad analysis. All eight hypothesised relationships were statistically supported. Each of the four antecedents was positively associated with integrated pedagogical content knowledge, which was in turn positively associated with both teaching readiness and pedagogical ethics awareness, and ethics awareness complementarily and partially mediated the competence–readiness association. The study offers initial empirical support for prompt engineering skill as a construct that is empirically distinguishable from artificial intelligence literacy, and for pedagogical ethics awareness as a mechanism that helps account for the association between competence and readiness. Because the design is cross-sectional and the sample is drawn from one subject-specific cohort at a single institution, the findings are interpreted as associations within this setting rather than as causal effects or as representative of Indonesian teacher professional education more broadly.

Cultural Resilience Pedagogy in History Education: Integrating Sapat Taun into Participatory Historical Learning

2026-10-15T15:08:53+07:00

Globalization and digital culture have intensified concerns about declining historical awareness, cultural identity, and cultural continuity among younger generations. This study examines how the Sapat Taun tradition contributes to cultural resilience and historical consciousness through participatory history learning using a Participatory Rural Appraisal–Participatory Learning and Action (PRA–PLA) approach. A qualitative participatory design was conducted at SMAN 1 Marau and Riam Batu Gading Village, Ketapang Regency, West Kalimantan, Indonesia, involving two history teachers, 32 eleventh-grade students, five traditional leaders, three community elders, and village representatives. Data were generated through participant observation, in-depth interviews, focus group discussions, cultural mapping, oral history projects, reflective journals, and document analysis, and analyzed using reflexive thematic analysis. Findings identify four interconnected processes: reconstruction of collective memory, development of historical consciousness, strengthening of cultural resilience through active engagement and documentation, and cultivation of inclusive national identity and civic responsibility. Based on these findings, the study develops Cultural Resilience Pedagogy (CRP) as an empirically grounded conceptual framework linking collective memory, historical consciousness, and cultural resilience in history education.

Proposed Pedagogical Framework for Positioning Generative AI as a Thinking Companion in Recount Writing

2026-10-15T17:06:52+07:00

Generative artificial intelligence (GenAI) can provide feedback, questions for reflection, and language support in writing instruction. However, without clear guidance, students may delegate idea generation, drafting, and revision to the technology. This study identifies local needs in Grade IX recount-writing instruction and proposes a preliminary framework for positioning GenAI as a thinking companion. It draws on content analysis of the Grade IX Indonesian Language Student’s Book and Teacher’s Book, a semi-structured interview with one Indonesian language teacher, and descriptive analysis of six selected questionnaire items answered by 56 students. The materials examined provide model texts and initial writing activities but lack systematic revision guidance and recount-specific criteria in the final assessment rubric. Student responses indicate reported use of AI as a thinking companion, reported submission of assignments containing content partly or entirely copied from GenAI, and an expectation for appropriately integrated teaching materials. Considered alongside relevant literature, these findings inform six proposed principles: student ownership of ideas, independent work before AI use, limited AI assistance, guided prompts, critical evaluation of AI output, and transparency about AI use. The framework has not undergone expert validation or classroom testing. It should therefore be regarded as a context-specific design proposal, not an established guideline or evidence of instructional effectiveness.

AI-Assisted Mathematics Learning: Jeneponto Salt Production Ethnomathematics And Junior High Students' Problem-Solving Ability

2026-10-15T19:44:49+07:00

Mathematical problem-solving ability among junior high school students remains a challenge in mathematics learning, while the integration of artificial intelligence with local cultural contexts has received limited attention when examined simultaneously. This study aimed to analyze the improvement in the mathematical problem-solving ability of ninth-grade students through Problem-Based Learning (PBL) assisted by an AI chatbot in the context of salt-production ethnomathematics in Jeneponto, both overall and at each problem-solving stage. The study employed a quantitative approach with a one-group pretest–posttest design involving 30 ninth-grade students from a junior high school in Jeneponto Regency who were selected purposively. Mathematical problem-solving ability was measured using an ethnomathematics-based contextual essay test and analyzed using the Shapiro–Wilk test, paired-samples t-test, Cohen's dz, and N-Gain. The mean score increased from 33.13 on the pretest to 80.80 on the posttest. This increase was statistically significant, t(29) = 17.419, p < .001, with a very large effect size (Cohen's dz = 3.18). The N-Gain of 0.707 indicated a high level of improvement. Analysis by problem-solving stage showed improvement across all indicators, with the largest absolute increase occurring at the carrying out the solution stage (D3), followed by the checking the result stage (D4). The findings indicate that the integration of PBL, AI chatbot support, and salt-production ethnomathematics has the potential to support contextual mathematics learning and the development of students' problem-solving ability.

Pedagogical Design of Digital Storytelling in Language Learning: Narrative Construction, Scaffolding, And Multimodal Composition

2026-10-15T20:10:35+07:00

Digital Storytelling (DST) brings together language use, story development, and meaning making through a range of digital resources. The diversity of DST practices suggests a need to explain more clearly how narrative construction, scaffolding, and multimodal composition are connected within language learning. This study used a Systematic Literature Review, with search and selection procedures reported according to PRISMA 2020 and PRISMA-S. Searches were conducted in Scopus and ERIC for English-language empirical journal articles published between 2016 and 2025. Thirty-four full-text articles met the inclusion criteria and were examined through deductive–inductive coding and methodological appraisal using the 2018 Mixed Methods Appraisal Tool (MMAT). The findings show that narrative construction shapes story content and communicative purpose, scaffolding supports story development and revision, and multimodal composition enables narrative ideas to be represented and, in some cases, reconsidered. These three dimensions interact iteratively through storyboarding, collaboration, feedback, transmediation, revision, publication, and reflection. Their configuration varies across educational levels, language contexts, task structures, collaborative arrangements, and technological environments. Synthesis produces a relational conceptual framework and seven propositions derived from recurring patterns across literature. Together, they explain recurring pedagogical processes in DST without prescribing a fixed instructional sequence or claiming an empirically validated learning model.

How Principal Self-Concept Predicts Decision-Making Quality: Empirical Evidence from School Leadership

2026-10-15T22:29:11+07:00

This study aims to analyze the predictive role of self-concept in the decision-making of elementary school principals. The study employs a survey design with a quantitative approach. The population consists of 62 public elementary school principals in the Jagakarsa District of South Jakarta, with a sample of 54 principals selected using random sampling. Data were collected via a Likert-scale questionnaire. Data analysis involved both descriptive and inferential methods, specifically simple linear regression. The results indicate a positive and significant predictive relationship between self-concept and principals' decision-making (β = 0.624; p < 0.001), with the regression equation Y = 25.261 + 0.722X. The coefficient of determination (R²) of 0.389 reveals that self-concept accounts for 38.9% of the variation in principals' decision-making, while the remaining 61.1% of the variation is not explained by this model. These findings suggest that self-concept is a personal aspect associated with principals' decision-making, alongside other factors not examined in this study. Consequently, initiatives to develop principals' leadership capacity could consider strengthening personal aspects—including the development of self-concept—as part of efforts to enhance the quality of decision-making.

Fiqh Education in the Context of Artificial intelligence: Shifts in Pedagogy, Authority, and Religious Literacy

2026-10-16T00:11:11+07:00

The rapid development of artificial intelligence (AI) has expanded the use of information and learning resources in Islamic Religious Education, including Fiqh learning, while raising concerns regarding pedagogy, the authority and validity of religious knowledge, and students’ ability to evaluate AI-generated information. This study examines the use of AI in Islamic education and synthesizes its implications for Fiqh learning, focusing on pedagogy, the authority and validity of religious knowledge, and religious literacy and AI literacy. A Systematic Literature Review (SLR) was conducted on 25 scholarly articles published between 2020 and 2026 and retrieved from Scopus, Google Scholar, and the Indonesian scholarly databases GARUDA/SINTA. Searches used combinations of keywords related to artificial intelligence, generative AI, ChatGPT, Islamic education, Islamic religious education, and fiqh education. Articles were selected based on their relevance to Islamic education or Fiqh learning and to the analytical dimensions of the study. Data were analyzed through coding, thematic categorization, cross-source comparison, and thematic synthesis. The findings indicate that AI is increasingly used to support personalized learning, information exploration, learning-material development, and instructional interaction. However, its use also intensifies the need for source verification, information validation, and teacher guidance. The synthesis further highlights the importance of integrating religious literacy and AI literacy to enable students to evaluate AI-generated information critically and responsibly. Based on these findings, the study proposes an AI-assisted but teacher-guided Fiqh learning framework that positions AI as a learning support tool, teachers as pedagogical and knowledge mediators, and students as active evaluators of information.

Unity 3D-Assisted Differentiated Problem-Based Learning versus Canva-Assisted Direct Instruction: Comparing Students' Learning Outcomes and Critical Thinking Skills

2026-10-16T00:25:34+07:00

This study aims to analyze the differences in learning outcomes and critical thinking skills between students receiving Unity3D-assisted, PBL-based differentiated instruction and those receiving Canva-assisted Direct Instruction (DI) in the fourth-grade Natural and Social Sciences (IPAS) subject, specifically regarding the topic of plant anatomy. The study employs a quantitative comparative approach and a quasi-experimental method using a posttest-only control group design. It was conducted at SD Negeri Blang Benara, Bener Meriah Regency, with a sample of 30 fourth-grade students comprising 15 in the experimental group and 15 in the comparison group selected via a saturated sampling technique. Data collection instruments consisted of a multiple-choice test for learning outcomes and an essay test for critical thinking skills. Data analysis utilized descriptive and inferential statistics via Multivariate Analysis of Variance (MANOVA) following the fulfillment of normality and homogeneity prerequisite tests. The results indicate a significant simultaneous difference between the two groups. Partial analysis revealed significant differences in both learning outcomes (F = 43.737; Sig. = 0.000) and critical thinking skills (F = 46.840; Sig. = 0.000). The group receiving Unity3D-assisted, PBL-based differentiated instruction demonstrated higher posttest scores compared to the group receiving Canva-assisted DI. These findings indicate that, within the context of this study, Unity3D-assisted, PBL-based differentiated instruction yields superior posttest results in both learning outcomes and critical thinking skills compared to Canva-assisted Direct Instruction.