<![CDATA[Students' mathematical reasoning, representation, and learning independence remain important concerns in elementary mathematics because classroom activities often emphasize procedural answers rather than exploration, communication, and reflection. This study examined the effectiveness of the scientific approach compared with the contextual approach in improving these three competencies. A quasi-experimental method with a nonequivalent control group design was used. The participants were 60 sixth-grade students at an elementary school in East Lombok Regency during the second semester of the 2025/2026 academic year, consisting of 30 students in the scientific approach class and 30 students in the contextual approach class. Data were collected through mathematical reasoning tests, mathematical representation tests, and a learning independence questionnaire. The instruments were validated by experts and tested empirically before use. Data were analyzed using descriptive statistics, normalized gain, prerequisite tests, and independent samples t-tests. The results showed that the scientific approach produced higher gains in mathematical reasoning (N-Gain = 0.61), mathematical representation (N-Gain = 0.60), and learning independence (N-Gain = 0.56) than the contextual approach. The independent samples t-tests indicated significant differences between the two groups (p < 0.05). These findings suggest that the scientific approach effectively strengthens students' cognitive and metacognitive competencies through observing, questioning, experimenting, reasoning, and communicating activities.]]>
