Answers are grounded in this paper and cite their sources.
Scientific literacy is an essential skill for elementary school students to understand, analyze, and explain natural phenomena through scientific reasoning. However, teacher-centered learning may limit students’ participation in observation, exploration, and problem-solving activities. This study aimed to analyze the effect of implementing the Deep Learning approach on the scientific literacy skills of fourth-grade students at SDN Windusari. The study employed a quantitative approach with a pre-experimental one-group pretest-posttest design involving 22 students selected through purposive sampling. Data were collected using a multiple-choice test that had undergone validity and reliability testing, supplemented by observation and documentation. Data analysis was conducted using the Shapiro-Wilk normality test and the Paired Samples t-Test. The results indicated that the data were normally distributed, with pretest and posttest significance values of 0.104 and 0.125, respectively. Hypothesis testing yielded a significance value of less than 0.001, accompanied by a mean score increase of 30.818 points. These findings indicate an improvement in students’ scientific literacy skills following the implementation of the Deep Learning approach, although the limitations of the research design should be considered when interpreting the results. ABSTRAK Literasi sains merupakan kemampuan penting bagi peserta didik sekolah dasar untuk memahami, menganalisis, dan menjelaskan fenomena alam melalui penalaran ilmiah. Namun, pembelajaran yang masih berpusat pada guru berpotensi membatasi keterlibatan peserta didik dalam kegiatan pengamatan, eksplorasi, dan pemecahan masalah. Penelitian ini bertujuan menganalisis pengaruh penerapan pendekatan Deep Learning terhadap kemampuan literasi sains peserta didik kelas IV SDN Windusari. Penelitian menggunakan pendekatan kuantitatif dengan desain pre-experimental one-group pretest-posttest yang melibatkan 22 peserta didik melalui teknik purposive sampling. Data dikumpulkan menggunakan tes pilihan ganda yang telah melalui pengujian validitas dan reliabilitas, serta didukung oleh observasi dan dokumentasi. Analisis data dilakukan melalui uji normalitas Shapiro-Wilk dan uji hipotesis menggunakan Paired Samples t-Test. Hasil menunjukkan bahwa data berdistribusi normal dengan nilai signifikansi pretest sebesar 0,104 dan posttest sebesar 0,125. Pengujian hipotesis menghasilkan nilai signifikansi kurang dari 0,001 dengan peningkatan rata-rata skor sebesar 30,818 poin. Temuan ini menunjukkan adanya peningkatan kemampuan literasi sains setelah penerapan pendekatan Deep Learning, meskipun keterbatasan desain penelitian perlu dipertimbangkan dalam menafsirkan hasil.
No takes yet. Share an insight, caveat, or question.
Wati et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: