<![CDATA[This study investigated whether a pedagogical Deep Learning approach integrating meaningful, mindful, and joyful learning could strengthen eighth-grade students’ mathematical conceptual understanding. The study employed a quantitative quasi-experimental method with a randomized control-group-only design. The participants were 62 eighth-grade students at MTsN 6 Lima Puluh Kota during the 2024/2025 academic year. Class VIII.6, consisting of 31 students, was assigned as the experimental group and received instruction through the Deep Learning approach, whereas class VIII.4, also consisting of 31 students, served as the control group and received instruction through the scientific approach. Data were collected using a five-item essay test covering the abilities to restate concepts, provide examples and non-examples, represent concepts in multiple mathematical forms, classify objects according to conceptual characteristics, and apply concepts or procedures in problem-solving. The instrument was reviewed by three mathematics teachers and demonstrated high internal consistency, with a Cronbach’s alpha coefficient of 0.84. Descriptive analysis showed that the experimental group achieved a higher mean posttest score than the control group, with scores of 85.60 and 73.80, respectively. The experimental group also outperformed the control group across all conceptual-understanding indicators. The largest difference occurred in applying concepts or procedures to solve problems, with mean scores of 77 and 50, respectively. The independent-samples t-test yielded a value of 3.801, indicating stronger mathematical conceptual understanding among students taught through the Deep Learning approach. These findings suggest that learning experiences emphasizing conceptual connections, reflection, active engagement, and authentic problem-solving can support students in developing deeper and more transferable mathematical understanding. Nevertheless, the findings should be interpreted within the context of a single school, two intact classes, a relatively short intervention, and a posttest-only design.]]>
Copyrights © 2026
