cover
Contact Name
Rojai Zhofir
Contact Email
rojaizho@gmail.com
Phone
+6285709037738
Journal Mail Official
j.pustakaindonesia@gmail.com
Editorial Address
Jl. Jaya Wijaya No.64, Dusun Besar, Kec. Singaran Pati, Kota Bengkulu, Bengkulu 38224
Location
Kota bengkulu,
Bengkulu
INDONESIA
Indonesian Science Education Journal (ISEJ)
ISSN : -     EISSN : 27163350     DOI : https://doi.org/10.62159/isej.XXXX
ISEJ : Indonesian Science Education Journal ISEJ : Indonesian Science Education Journal is a medium of communication used by researchers, lecturers, teachers, practitioners, and University student for submitting result of studies and prioritized result of the study and review of the literature in the sphere of natural science education in primary education, secondary education and higher education. Additionally, this journal also covers the issues of environmental education & environmental science.
Arjuna Subject : Umum - Umum
Articles 184 Documents
Generative AI in University Mathematics Education Using the ‎Theory–Context–Methodology Framework: A Systematic ‎Literature Review ‎ Ida Nuraida; Nurjanah; Rahayu Kariadinata
ISEJ : Indonesian Science Education Journal Vol. 7 No. 3 (2026): June-September
Publisher : Yayasan Darussalam Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62159/isej.v7i3.2140

Abstract

Advances in generative AI are opening new opportunities for mathematics learning in higher education, but systematic mapping of the Theory–Context–Methodology (TCM) framework to integrate generative AI into mathematics learning remains limited. This study aims to identify the theories or conceptual frameworks, application contexts, and research methods used in studies of generative AI in higher education mathematics education. The method used is a PRISMA-based systematic literature review. The search was conducted on Scopus, Lens, and Dimensions, yielding 322 initial articles; after deduplication, screening, and eligibility assessment, 31 articles were analyzed. Article quality was assessed using the MMAT, while data were analyzed through TCM-based coding. The results indicate that generative AI is conceptualized as a pedagogical tool, a technology adoption, pedagogical competence, mathematical activity, and a cognitive partner. Its application is dominant in the classroom, general mathematics, ChatGPT, and for conceptual understanding. Methodologically, the studies were dominated by qualitative and quantitative approaches, while mixed methods and R&D/development research remained limited. These findings underscore the need for AI-based instructional designs that are critical, ethical, and supported by more integrative empirical research.
Adaptive Leadership in Academic Supervision for Deep Learning ‎at MAN 2 Tanah Datar Silvia Susrizal; Nur Azizah; Merri Yelliza; Susi Ratna Sari; Mega Adhyna Movitaria; Syahrur Ramli; Yoga Habibullah; Ramza Husmen
ISEJ : Indonesian Science Education Journal Vol. 7 No. 3 (2026): June-September
Publisher : Yayasan Darussalam Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62159/isej.v7i3.2462

Abstract

This study examines how adaptive leadership is interpreted in academic supervision for deep learning-oriented Islamic Religious Education (PAI) at MAN 2 Tanah Datar. It addresses the tension between supervision as an administrative requirement and supervision as professional support for pedagogical change. A descriptive qualitative approach within a constructivist paradigm was employed, involving semi-structured interviews with 27 PAI teachers, one madrasah principal, and one vice principal for curriculum affairs (29 participants). Interview accounts were examined through thematic analysis and source triangulation across participant roles. The analysis identified three interrelated patterns: open communication and participatory decision-making, mentoring and professional support, and the use of local Minangkabau cultural references to interpret supervisory relationships. These patterns indicate a collaborative and developmental orientation to supervision, while the available interview evidence does not demonstrate measured changes in classroom practice or student learning. The findings are interpreted through adaptive leadership and culturally responsive school leadership, with deep learning treated as the pedagogical orientation of supervision rather than an empirically measured outcome. The study contributes a context-sensitive account of how formal supervision can be mediated by locally meaningful communication and deliberation. Practically, it suggests that madrasah leaders should connect dialogic supervision and mentoring to specific instructional decisions and follow-up, while future research should verify these relationships through classroom observation, documentary evidence, and longitudinal designs.
Needs Analysis of Interactive Learning Media Based on Local Biodiversity for Senior High School Science Education in the Kei Islands, Southeast Maluku Anggelina Jeujanan; Diana Rochintaniawati; Riandi; Taufik Rahman
ISEJ : Indonesian Science Education Journal Vol. 7 No. 3 (2026): June-September
Publisher : Yayasan Darussalam Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62159/isej.v7i3.2488

Abstract

Science learning resources that authentically represent local biodiversity remain limited in senior high schools, particularly in the Kei Islands of Southeast Maluku, Indonesia. This study identified students' and teachers' perceived needs for developing interactive learning media based on local biodiversity for senior high school science education. A descriptive survey was conducted with 23 students from one senior high school and 12 biology and science teachers from several senior high schools in the Kei Islands, recruited using purposive sampling. Data were collected using separate 23-item Likert-scale questionnaires with open-ended questions. Two science education experts reviewed the questionnaires, and content validity was assessed using Aiken’s V, while internal consistency was examined using Cronbach’s alpha for each dimension. The results showed a high overall level of need among students (M = 4.03) and a very high level among teachers (M = 4.23). The need for interactive learning media was rated very highly by students (M = 4.24) and teachers (M = 4.30). Open-ended responses highlighted participants' interest in locally relevant biodiversity content and visual, interactive, contextual, and experiential features. Leatherback turtles (Tabob/penyu belimbing) were mentioned by 3 of 23 students (13.0%) and 2 of 12 teachers (16.7%). These findings provide preliminary, context-specific evidence to guide the design of local-biodiversity-based interactive learning media. However, the study is limited by its small purposive sample, regional context, and reliance on self-reported data, and it does not include prototype validation, practicality testing, or effectiveness testing.
Developing a RAW-Based Science E-Comic to Enhance Junior High ‎School Students’ Learning Outcomes and Higher-Order Thinking ‎Skills in Environmental Science Anip Dwi Saputro; Adib Khusnul Rois; Siti Mukaromah
ISEJ : Indonesian Science Education Journal Vol. 7 No. 3 (2026): June-September
Publisher : Yayasan Darussalam Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62159/isej.v7i3.2534

Abstract

This study aimed to develop a RAW-based E-Comic Sains Green Ranger for teaching environmental pollution and global warming and to examine its preliminary effectiveness in improving seventh-grade students’ science learning outcomes and higher-order thinking skills (HOTS). Using a research and development approach structured by the ADDIE stages (analysis, design, development, implementation, and evaluation), the e-comic combined contextual environmental narratives with Reading, Analyzing, and Writing activities. A limited trial involved 64 Grade VII students at SMPN 2 Ponorogo, Indonesia, divided equally between an experimental class (n = 32) and a control class (n = 32). Data were obtained from pretests and posttests of science learning outcomes and HOTS observation sheets and analyzed descriptively using normalized gain (N-gain). The experimental class achieved a higher mean posttest score (88.44) than the control class (81.72). Its learning-outcome N-gain was 0.78 (high), compared with 0.64 (moderate) in the control class; corresponding HOTS N-gain values were 0.76 and 0.51, respectively. These descriptive findings suggest that integrating an environmental e-comic with structured reading, analysis, and writing activities may support conceptual learning and higher-order thinking. However, the limited two-class trial and absence of reported inferential testing constrain claims of comparative effectiveness. Further studies with larger samples and rigorous statistical evaluation are needed to establish the intervention’s impact and transferability.
Evaluating The Attenuation Effects of Wood and Glass Building Materials on 4g Cellular and Wifi Signal Transmission Power Using the Opensignal Application Kukuh Aris Santoso; Muhammad Sobirin
ISEJ : Indonesian Science Education Journal Vol. 7 No. 3 (2026): June-September
Publisher : Yayasan Darussalam Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62159/isej.v7i3.2552

Abstract

Building materials can modify indoor wireless performance, yet low-cost measurements often conflate radio-frequency attenuation with end-to-end network throughput. This study compared observed 4G cellular and WiFi performance inside 0.3 m³ enclosures made of commercial wood and clear glass. A smartphone running OpenSignal recorded download speed, upload speed, and latency for Telkomsel, Tri (3), and campus WiFi under outdoor-reference and enclosed conditions. For Telkomsel, wood reduced download throughput by 28.1%, whereas glass reduced it by 16.4%. For Tri, glass maintained download throughput near baseline, but its upload increase coincided with latency rising to 183 ms, indicating unstable network behavior. Campus WiFi showed a different pattern: wood produced only 3.6% and 1.6% reductions in download and upload throughput, while glass produced reductions of 46.9% and 55.5%, respectively. The findings therefore indicate material- and network-dependent performance rather than a universally superior barrier. Because OpenSignal measures application-level throughput rather than transmission power directly, the results should be interpreted as descriptive performance evidence. The experiment offers a contextual, low-cost inquiry activity for teaching electromagnetic-wave propagation, experimental control, and data interpretation in science education.
Multisensory-Based Realistic Mathematics Education (RME) Study of Elementary School Students' Conceptual Understanding and Mathematical Dispositions Dessy Trisnawati‎; Ririn Widiyasari
ISEJ : Indonesian Science Education Journal Vol. 7 No. 3 (2026): June-September
Publisher : Yayasan Darussalam Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62159/isej.v7i3.2624

Abstract

Indonesian elementary students’ mathematical literacy remains below the OECD average. This study examined the implementation of Multisensory-Enriched Realistic Mathematics Education (ME-RME) for cubes and cuboids and compared students’ conceptual understanding and mathematical disposition with conventional instruction. An exploratory sequential mixed-methods design was conducted at a public elementary school in Bekasi Regency, West Java. The qualitative phase involved structured observations of 35 sixth-grade students over four lessons and semi-structured interviews with 33 students; findings informed refinement of the lessons and instruments. The quantitative phase used a non-equivalent pretest–posttest quasi-experiment involving an experimental class (n = 35) and a control class (n = 35), each taught by the same teacher over four 70-minute lessons. Conceptual understanding was assessed through an essay test and mathematical disposition through a Likert-scale questionnaire. Analyses included ANCOVA, MANOVA, and Bonferroni-corrected follow-ups. Groups did not differ significantly at pretest, t(68) = 1.32, p = .192. At posttest, the experimental group outperformed the control group in conceptual understanding (M = 90.69, SD = 5.85 vs. M = 68.46, SD = 7.47; t(68) = 13.86, p < .001; Hedges’ g = 3.28) and disposition (M = 119.05, SD = 9.41 vs. M = 102.05, SD = 10.79; g = 1.66). MANOVA was significant, Wilks’ Λ = .249, F(2, 67) = 100.80, p < .001. ME-RME was associated with higher conceptual understanding and more positive mathematical disposition, although the short intervention, single-school setting, and non-random allocation limit generalizability.
Development of AI–CBL Learning Materials to Promote Pre-Service Physics Teachers’ Representational Competence in Solid-State Physics Nurdini Nurdini; Heni Rusnayati; Raden Giovanni Ariantara; Tiara Husnul Khotimah; Feri Apryandi; Asep Akmal Fadia Nurhalim; Iyon Suyana; Nor Farahwahidah Abdul Rahman
ISEJ : Indonesian Science Education Journal Vol. 7 No. 3 (2026): June-September
Publisher : Yayasan Darussalam Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62159/isej.v7i3.2651

Abstract

This study aimed to develop and validate AI–CBL learning materials for solid-state physics that were designed to support pre-service physics teachers’ representational competence. A research and development approach using the ADDIE framework was applied and limited to the Analysis, Design, and Development stages. The analysis stage identified key learning difficulties and representational demands in solid-state physics instruction for pre-service physics teachers. Based on these findings, an AI-integrated Case-Based Learning package was designed and developed, consisting of contextual case studies, structured learning tasks aligned with case-based learning phases, and guidelines for responsible interaction with open-access generative artificial intelligence. The materials were organized through a learning management system to support consistent access, structured sequencing, and documentation of learning artifacts. The developed materials were reviewed by three expert validators using the Learning Object Review Instrument (LORI) 2.0 across eight aspects on a five-point scale. The validation data were analyzed descriptively and supported by Many-Facet Rasch Measurement using Minifac to examine rater consistency and measurement quality. The expert validation results showed an average feasibility percentage of 76.50% across the eight LORI aspects, indicating that the developed materials were feasible for further implementation. Rasch-based analysis also indicated good unidimensionality, with 47.84% of variance explained by Rasch measures, and acceptable rater consistency, with reliability of 0.87. Revisions were made based on validators’ comments to improve clarity, learning-goal alignment, usability, and responsible artificial intelligence integration. These findings indicate that the developed AI–CBL learning materials are feasible as a validated instructional product and provide a foundation for subsequent classroom implementation studies in solid-state physics.
Teacher Engagement and Adversity Quotient as Predictors of ‎Academic Supervision in Vocational Schools in Pesisir Selatan ‎Regency Fetmawati; Hanif Al Kadri; Nellitawati; Merika Setiawati
ISEJ : Indonesian Science Education Journal Vol. 7 No. 3 (2026): June-September
Publisher : Yayasan Darussalam Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62159/isej.v7i3.2728

Abstract

Academic supervision is an important strategy for improving teacher professionalism and learning quality in Vocational High Schools (SMK). This cross-sectional explanatory study examined whether Teacher Engagement and Adversity Quotient predict teacher-reported Academic Supervision among vocational high school teachers in Pesisir Selatan Regency. The population comprised 586 teachers, and 234 respondents were selected through proportionate simple random sampling. Data were collected using a 48-item five-point Likert questionnaire and analyzed with multiple linear regression in IBM SPSS. Teacher Engagement was a positive predictor of Academic Supervision (B = 1.211, SE = .060, β = .866, 95% CI [1.093, 1.329], p < .001), as was Adversity Quotient (B = .858, SE = .057, β = .462, 95% CI [.746, .970], p < .001). The overall model was significant, F(2, 231) = 356.425, p < .001, with R² = .755 and adjusted R² = .753. Thus, the two predictors jointly explained approximately 75.5% of the variance in teacher-reported Academic Supervision. These findings indicate that Teacher Engagement and Adversity Quotient are strongly associated with how teachers perceive and participate in academic supervision; however, the cross-sectional design does not establish causality. School leaders should therefore combine developmental supervision with strategies that strengthen teacher engagement and adaptive coping.
The Influence of Transformational Leadership and Teacher ‎Professional Competence on Organizational Citizenship Behavior ‎Through Job Satisfaction of Public Middle School Teachers in ‎Padang Pariaman Regency Srina Desmira; Nellitawati; Anisah; Novriyanti Achyar
ISEJ : Indonesian Science Education Journal Vol. 7 No. 3 (2026): June-September
Publisher : Yayasan Darussalam Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62159/isej.v7i3.2740

Abstract

Organizational citizenship behavior (OCB) is essential to school effectiveness because it reflects teachers’ voluntary contributions beyond formal role requirements. This study examines a conceptual model linking principals’ transformational leadership and teachers’ professional competence with OCB through job satisfaction among public junior high school teachers in Padang Pariaman Regency, Indonesia. A quantitative correlational design was employed, with a target sample of 314 teachers selected through proportional and cluster-based random sampling. Data were collected using five-point Likert instruments representing transformational leadership, professional competence, job satisfaction, and OCB. Before structural testing, the instruments were piloted with 30 teachers and evaluated using Pearson item–total correlations and Cronbach’s alpha. The reported results show strong measurement readiness: 35 of 36 OCB items, all 39 transformational-leadership items, 18 of 26 professional-competence items, and 34 of 35 job-satisfaction items met the validity criterion (r > 0.361). Internal-consistency coefficients were very high for OCB (α = 0.955), transformational leadership (α = 0.989), professional competence (α = 0.907), and job satisfaction (α = 0.968). These findings establish a sufficiently valid and internally consistent measurement foundation for subsequent path analysis of the proposed direct and mediated relationships. Practically, the validated instruments can support evidence-based diagnosis of leadership, competence, satisfaction, and extra-role behavior in secondary-school management, although structural-effect claims require the full path-analysis results to be reported.
Effectiveness of STEM Project Using Caesalpinia sappan L. Natural ‎Dye to Enhance High School Students’ Chemical Literacy and Food ‎Safety Awareness Nurmala; Dini Hadiarti; Fitriani
ISEJ : Indonesian Science Education Journal Vol. 7 No. 3 (2026): June-September
Publisher : Yayasan Darussalam Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62159/isej.v7i3.2768

Abstract

This study examined changes following STEM-integrated Project-Based Learning (STEM-PjBL) using natural dye extracted from secang wood (Caesalpinia sappan L.) to examine chemical literacy and describe post-intervention food safety awareness in acid–base learning. A pre-experimental one-group pretest–posttest design was employed with 35 eleventh-grade students selected through purposive sampling. Data were collected using a validated chemical literacy test and a food safety awareness questionnaire. Validation scores ranged from 93% to 100%, and chemical literacy test reliability was acceptable (Cronbach’s alpha = 0.80). The Shapiro–Wilk test indicated that posttest scores were not normally distributed (p = .001); therefore, the Wilcoxon signed-rank test was used to examine differences between pretest and posttest scores. The results showed a significant increase in chemical literacy, with the mean score increasing from 46.88 to 88.29. The Wilcoxon test indicated a statistically significant difference (Z = −5.193, p < .001), with a large effect size (r = 0.88). The N-gain was 0.78, indicating a high level of improvement. The post-intervention food safety awareness questionnaire score was 85.29%; no baseline awareness score was available. These findings indicate an improvement in chemical literacy scores following STEM-PjBL, while food safety awareness was described only after the intervention.