JOURNAL OF SCIENCE EDUCATION AND PRACTICE

Content Validity Analysis of Multitier Diagnostic Test Instruments for Measuring Junior High School Disaster Literacy

2026-07-24T12:08:07+00:00

Background: Disaster literacy is an essential competency for students in disaster-prone areas such as Bengkulu Province. However, conventional school assessments cannot adequately measure students’ conceptual understanding, misconceptions, and disaster literacy. Moreover, no validated four-tier diagnostic instrument is available for vibration and wave topics.

Methods: This study used the ADDIE R&D model, with research limited to the development phase. Five experts validated the 32-item instrument. Two university lecturers and three junior high school teachers served as experts. The Content Validity Ratio (CVR) and Content Validity Index (CVI) were used to assess content validity.

Results: The CVR values for all 32 test items ranged from 0.81 to 1.00. Based on Lawshe's criteria, 18 items met the minimum CVR requirement, while the remaining 14 were revised per the validators' recommendations. The Content Validity Index (CVI) was 1.00 for the knowledge and attitude aspects and 0.96 for the skills aspect, yielding an overall CVI of 0.972, indicating excellent content validity. The revised instrument was then deemed ready for implementation with junior high school students to collect empirical response data for subsequent validity and reliability analyses.

Conclusion: The developed disaster-literacy-based four-tier diagnostic test demonstrated excellent content validity following expert validation and revision. This study contributes a validated four-tier instrument for assessing disaster literacy and students’ conceptual understanding in vibration and wave topics. Despite the challenges of developing coherent four-tier items, the instrument was considered suitable for empirical testing with junior high school students.

Analysis And Comparison of Computational Thinking Skills of Elementary School Students in Science Learning on Ecosystem and Food Chain Materials in Public And Private Schools

2026-06-12T03:15:47+00:00

Background: Computational thinking (CT) has become an essential competency in twenty-first-century education and is increasingly integrated into elementary science learning. Ecosystem and food chain topics provide opportunities for students to analyze relationships, identify patterns, and apply systematic reasoning. However, comparative studies examining computational thinking among public and private elementary school students within science learning contexts remain limited.

Methods: This study employed a quantitative comparative design involving 29 fifth-grade students from one public elementary school and one private elementary school in Purwakarta, Indonesia. Data were collected using Data were collected using an eight-item open-ended test selected from an initial pool of 16 items through expert validation and empirical testing designed to assess four dimensions of computational thinking: decomposition, pattern recognition, abstraction, and algorithmic thinking. Descriptive statistics and the Mann–Whitney U test were used to analyze the data.

Results: The findings revealed that private school students achieved higher computational thinking scores than public school students. Among the four dimensions, abstraction showed the highest level of achievement, whereas algorithmic thinking demonstrated the lowest performance. Statistical analysis indicated a significant difference in computational thinking skills between the two groups (p = 0.008), suggesting that differences in learning experiences may be associated with differences in computational thinking development.

Conclusion: The study highlights the potential of ecosystem and food chain learning as a meaningful context for fostering computational thinking in elementary science education. The findings also emphasize the importance of providing structured learning experiences, including problem-solving and coding-related activities, to support the development of computational thinking skills among elementary school students.

Analysis of the Readiness of Prospective Elementary School Teachers in Integrating Artificial Intelligence in Science Learning

2026-06-12T01:43:12+00:00

Background: The determination of Artificial Intelligence (AI) as an optional subject in Permendikdasmen Number 13 of 2025 places prospective elementary school teachers as a central actor in policy implementation, but the empirical profile of the readiness of prospective elementary school teachers in integrating AI in science learning is not yet available in the Indonesian literature.

Methods: A quantitative approach with a descriptive-correlational survey design was applied to 163 PGSD students of Universitas Pendidikan Indonesia from the main campus and five regional campuses, who were selected through purposive sampling. The Readiness for Artificial Intelligence Applications Scale (RAIS) instrument was used after limited modification (16 valid items, Cronbach's α = 0.921). Data were analyzed with descriptive statistics and Spearman's Rho correlation using IBM SPSS Statistics version 21.

Results: The overall readiness level of prospective elementary school teachers was in the High category (M = 64.21; SD = 7.49; 80.26%), with 96.9% of respondents in the High to Very High category. The profile per dimension showed ethical awareness as the strongest dimension (82.88%), followed by technology self-efficacy (81.23%), and student interaction as the most moderate dimension (78.07%). Correlation analysis revealed a significant positive relationship across all dimensional pairs (p < 0.01), with the strongest correlation between student interaction and ethical awareness (ρ = 0.509).

Conclusion: Prospective elementary school teacher candidates at UPI demonstrate adequate AI readiness for the implementation of the 2025 Curriculum policy, with ethical awareness as the strongest dimension and student interaction as the most moderate dimension that needs to be strengthened through strengthening pedagogical experience.

Quantitative Descriptive Analysis of Science Curiosity in Science Learning Class V Elementary School

2026-06-12T01:43:13+00:00

Background: Science curiosity is an important scientific attitude in science learning because it encourages students to seek information, understand scientific phenomena, and engage actively in the learning process. However, research mapping the science curiosity profile of elementary school students based on indicators of stretching, embracing, and science practices remains limited.

Methods: This study uses a quantitative, descriptive approach. The research subjects consisted of 114 Grade V students from one of the elementary schools in Purwakarta Regency, selected using saturated sampling. Data were collected using a science curiosity questionnaire developed based on the Weible and Zimmerman indicators. Of the 30 items tested, 18 were declared valid, with a Cronbach's Alpha reliability of 0.796. Data were analyzed using descriptive statistics, including averages, percentages, standard deviations, and frequency distributions.

Results: The results of the study show that students' science curiosity is generally in the high category. A total of 4.39% of students were in the very high category, 57.02% in the high category, 34.21% in the medium category, and 4.39% in the low category. Based on the indicators, stretching obtained the highest average score (27.01), followed by embracing (18.34) and science practices (17.29). These findings show that students' curiosity is more developed in the exploration aspect of knowledge than in involvement in scientific investigation practices.

Conclusion: The findings indicate that elementary school students' science curiosity in science learning is generally in the high category. However, the science practice aspect still needs strengthening through learning activities that involve direct observation, experimentation, discussion, and scientific investigation.

Analysis of Teachers’ and Students’ Needs in the Development of Two-Tier Multiple-Choice Questions on Environmental Issues in Ecology and Environmental Pollution

2026-08-03T06:16:00+00:00

Background: The assessment of Higher-Order Thinking Skills (HOTS) is a crucial aspect of 21st-century science education. However, current evaluation instruments in schools are still dominated by Lower-Order Thinking Skills (LOTS) questions in the form of conventional multiple-choice questions, which trigger a “guessing effect” among students and are unable to accurately map conceptual understanding

Methods: This study is a preliminary research project employing a mixed-methods approach with a sequential explanatory design. Participants consisted of 421 junior high school students and 30 science teachers selected using purposive sampling. Data collection was conducted through Likert-scale questionnaires and open-ended questions. Data analysis used descriptive percentages for quantitative data and thematic analysis aided by NVivo software for qualitative data.

Results: The findings revealed that 60.83% of teachers still predominantly use LOTS-standardized assessments, while 70.33% of students often guess answers at random when encountering difficulties. The integration of real-world environmental case studies was highly favored by 81.01% of students because it helped make abstract ecological concepts more concrete. A total of 84.17% of teachers expressed an urgent need for a valid two-tier multiple-choice instrument, hoping for stimulus designs that are concise, visual, and feature a practical scoring system.

Conclusion: There is a high urgency to develop a two-tier multiple-choice instrument focused on environmental issues as an objective, diagnostic, and practical solution for evaluating Higher-Order Thinking Skills (HOTS).

Development of an Augmented Reality-Based Booklet on Earth Structure Learning to Improve Scientific Literacy and Disaster Mitigation Literacy

2026-07-26T09:47:26+00:00

Background: Scientific literacy and disaster preparedness among students are currently low due to the limited availability of contextual teaching materials.

Methods: This study utilized a Research and Development (R&D) approach based on the ADDIE model. The research employed a One-Group Pretest- Posttest design, involving 34 eighth-grade students at SMPS Al-Fikri.

Results: The AR-based booklet achieved an Aiken's V index of 0.98 in the subject matter expert validation and 0.97 in the media expert validation, both with very high criteria. Furthermore, the implementation demonstrated a notable improvement in students' scientific literacy with an N-Gain score of 0.645, as well as an enhancement in disaster mitigation literacy with an N-Gain score of 0.490.

Conclusion: Augmented reality-based media has a dual impact by effectivelyaddressing students' academic needs while simultaneously building their disaster resilience.