Peningkatan Kemampuan Pemecahan Masalah Matematis Mahasiswa melalui Blended Learning Berbasis Etnomatematika

Abstract

This study aims to analyze the improvement of students' mathematical problem-solving ability through inferential statistics learning based on ethnomathematics designed within a blended learning framework. The relatively low problem-solving ability of students in inferential statistics courses, along with instruction that remains abstract and disconnected from real contexts, motivated this study. A pre-experimental one-group pretest-posttest design was employed, involving 16 students enrolled in the Inferential Statistics course. The test instrument was developed based on Polya's four problem-solving indicators-understanding the problem, devising a plan, carrying out the plan, and looking back-contextualized within South Sulawesi's local wisdom, such as seaweed cultivation, fishermen's catch, small and medium enterprises, and cocoa farming. Data were analyzed using the Shapiro-Wilk normality test, Wilcoxon Signed-Rank test, and N-Gain analysis. The results showed that the average pretest score of 66.69 increased to 85.13 on the posttest, with an average N-Gain of 0.60 (moderate category). The Wilcoxon test indicated a significant improvement (p < 0.05). A total of 87.5% of students fell into the moderate improvement category and 12.5% into the high category, with no students in the low category. These findings suggest that statistics learning that integrates local wisdom context within a blended learning scheme effectively and consistently improves students' mathematical problem-solving ability across varying levels of prior ability.