Pengaruh Pendekatan Deep Learning terhadap Hasil Belajar Bahasa Indonesia Siswa Kelas II di SDN 15 Cakranegara

Abstract

The low ability of students in using uppercase and lowercase letters in Indonesian language learning remains a problem in elementary schools. This condition is influenced by conventional learning methods, causing students to be less active and not fully understand spelling concepts meaningfully. This study aimed to determine the effect of the Deep Learning approach on the Indonesian language learning outcomes of second-grade students at SDN 15 Cakranegara on the topic of uppercase and lowercase letter usage. This study employed a quantitative approach using a quasi-experimental research method with a Nonequivalent Control Group Design. The population consisted of all second-grade students of SDN 15 Cakranegara in the 2025/2026 academic year, with a sample of 67 students, including class II A as the experimental group and class II B as the control group, selected using saturated sampling technique. The research instrument was a multiple-choice test that had been tested for validity and reliability. Data collection techniques included pretests, posttests, and observations. Data analysis was conducted using the Kolmogorov-Smirnov normality test, Levene’s Test for homogeneity, Independent Sample t-Test for hypothesis testing, and N-Gain test. The results showed a significance value of 0.000 < 0.05, indicating that H₀ was rejected and H₁ was accepted. The average posttest score of the experimental class was 88.12, which was higher than the control class score of 66.85. The N-Gain score of the experimental class was 73.96 (moderately effective), while the control class obtained 25.90 (ineffective). Therefore, the Deep Learning approach had a positive effect on students’ Indonesian language learning outcomes.