Enhancing Elementary School Students' Speaking Skills through Self-Recorded Video-Assisted Deep Learning

Abstract

Speaking instruction in elementary schools often emphasizes performance outcomes while providing limited opportunities for students to reflect on their speaking process. Although video-assisted learning has been widely used, limited research has integrated a Deep Learning Approach with Self-Recorded Video to support reflective speaking development in elementary education. This study aimed to analyze the effect of a Deep Learning Approach assisted by Self-Recorded Video on elementary students’ Speaking Skills and explain their learning experiences. An explanatory sequential mixed-methods design employed a Nonequivalent Control Group Design involving 49 sixth-grade students, with 25 in the experimental group and 24 in the control group. Data were collected through speaking tests, Self-Recorded Video documentation, and semi-structured interviews. Quantitative data were analyzed using descriptive statistics, N-Gain, Mann–Whitney U Test, and effect size, while interviews with three students explained the quantitative findings. The experimental group improved from 9.80 to 16.68 with an N-Gain of 69.94%, while the control group improved from 9.88 to 13.29 with an N-Gain of 31.86%. The posttest difference was significant (U = 111.500; Z = -3.804; p < 0.001; r = 0.543). These findings indicate that integrating Deep Learning and Self-Recorded Video supports reflective practice and improves speaking performance