Article Per Year (5 Year)
p-Index From 2021 - 2026
0.562
P-Index
This Author published in this journals
All Journal Pendas : Jurnah Ilmiah Pendidikan Dasar Attractive : Innovative Education Journal
Stefanus Christian Relmasira
Universitas Kristen Satya Wacana, Indonesia
Other

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
INTEGRATING DEEP LEARNING APPROACHES INTO MATHEMATICS INSTRUCTION TO IMPROVE FORMULA MASTERY IN GRADE 5 Christina Yulinda Prihasti; Stefanus Christian Relmasira
Pendas : Jurnal Ilmiah Pendidikan Dasar Vol. 11 No. 02 (2026): Volume 11 No. 02 Juni 2026 Press
Publisher : Program Studi Pendidikan Guru Sekolah Dasar FKIP Universitas Pasundan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23969/jp.v11i02.48809

Abstract

This study was motivated by the low level of mathematical formula mastery among fifth-grade elementary school students due to surface learning, causing students to memorize formulas without understanding the concepts. This study aimed to analyze the effect of the Deep Learning approach through the Discovery Learning model on students’ mastery of mathematical formulas compared with the Cooperative Learning model. This study employed a quasi-experimental method using a non-equivalent control group design. The sample consisted of 38 fifth-grade students at SDN Kandangan 02 and SDN Kandangan 03. Data were collected using a hybrid test and analyzed through normality, homogeneity, and Independent Samples T-Test using IBM SPSS. The results showed that the experimental group I achieved a higher posttest mean score (73.684) than the experimental group II (58.947). The t-test result indicated a significant difference between the two groups with t = 2.657 and significance value 0.012 < 0.05. Therefore, the Deep Learning approach through the Discovery Learning model was more effective in improving students’ mastery of mathematical formulas.
AN EVALUATION OF MATHEMATICS INSTRUCTION IN FOSTERING CREATIVE PROBLEM-SOLVING SKILLS AMONG GRADE 3 STUDENTS Caroline Cindy Pratiwi; Stefanus Christian Relmasira
Pendas : Jurnal Ilmiah Pendidikan Dasar Vol. 11 No. 02 (2026): Volume 11 No. 02 Juni 2026 Press
Publisher : Program Studi Pendidikan Guru Sekolah Dasar FKIP Universitas Pasundan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23969/jp.v11i02.48819

Abstract

This research is motivated by field conditions where third-grade elementary school students often associate mathematics learning with feelings of fear, difficulty, and exhaustion, which ultimately hinders the emergence of Creative Problem Solving (CPS) skills. This study aims to describe students' problem-solving processes and difficulties, analyze the teacher's role as a facilitator, and evaluate the implementation of creative problem-solving instruction. A quantitative approach utilizing a Quasi-Experimental Design (Two-Treatment Design) was employed. The research sample consisted of a student group at SDN Jogoyasan receiving the Problem Based Learning (PBL) model and a student group at SDN Keditan receiving the Discovery Learning (DL) model. Data were collected through test techniques (pretest-posttest) and non-test methods (Y-Chart reflection instrument), which were analyzed statistically using SPSS and qualitatively using MAXQDA 2020 software. Hypothesis testing results revealed that both learning models had a highly significant effect on improving students' CPS abilities (p < 0.001). The PBL class experienced a Gain Score increase of 20.71 points, while the DL class increased by 18.46 points. Qualitative findings indicated that the use of concrete learning media and the creation of a Joyful Learning environment effectively transformed negative emotions into metacognitive awareness (Mindful Learning), making students braver and more enthusiastic in exploring problem-solving strategies. In conclusion, the integration of PBL and DL models with a Deep Learning approach is effectively implemented to enhance the CPS skills of third-grade elementary school students.
Pengembangan GURU CERDIG (Guru Cerdas Digital) dengan Bantuan AI untuk Perencanaan Pembelajaran Deep Learning di Sekolah Dasar Negeri Kabupaten Temanggung Gati Destiyani; Bambang Ismanto; Stefanus Christian Relmasira
Attractive : Innovative Education Journal Vol. 8 No. 2 (2026): Attractive : Innovative Education Journal
Publisher : CV. Creative Tugu Pena

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51278/aj.v8i2.1298

Abstract

The low level of students’ Higher-Order Thinking Skills (HOTS), as reflected in the 2022 Programme for International Student Assessment (PISA) results, together with the substantial administrative burden faced by teachers in developing high-quality deep learning lesson plans, remains a significant challenge in primary education. This study aimed to develop an AI-powered website-based platform called Guru Cerdig (Guru Cerdas Digital) to assist teachers in preparing instructional plans more efficiently and effectively. Product feasibility was evaluated using both qualitative and quantitative approaches, including validity and reliability testing, to ensure that the developed platform met acceptable standards of quality, consistency, and practical usability in educational settings. The study employed a Research and Development (R&D) methodology by integrating the ADDIE and Borg & Gall development models. Data were analyzed using a mixed-methods approach, utilizing MAXQDA for qualitative analysis and SPSS inferential statistical tests for quantitative analysis. The findings revealed that existing instructional planning practices remain largely dominated by conventional administrative procedures with limited pedagogical depth. The GURU CERDIG platform was rated as “Highly Feasible” by subject matter experts (93.83%) and IT/media experts (78.13%). From a practical perspective, the AI-powered features significantly reduced the time required to prepare instructional plans from several days to less than five minutes, with a user satisfaction rate of 48%. Furthermore, the Wilcoxon Signed-Rank Test produced a significance value of p < 0.001, indicating that the platform significantly improved the quality of instructional planning. The study concludes that GURU CERDIG represents a transformational digital ecosystem capable of enhancing teachers’ competence, efficiency, and confidence in implementing deep learning-oriented instructional practices.
Co-Authors Bambang Ismanto Caroline Cindy Pratiwi Christina Yulinda Prihasti Gati Destiyani