This study aims to explore patterns of student academic performance specifically about questions on exponents and roots using K-Means clustering and Principal Component Analysis (PCA). An exploratory quantitative method was employed, using data from a final mathematics examination involving 100 ninth-grade students and 10 questions. Twenty students who gave no answers were excluded from the analysis, leaving 80 students for the clustering process. The function of K- means in this research were defined students on the same group with same categorize with similarities responses. PCA was used for dimensionality and visualize for reduce data distribution groups that found. The analysis revealed two distinct groups where 75 students in the primary group scored a mean of 7.36 while 23 students in the secondary group scored a mean of 3.12. The highest silhouette score from these two different groups is 0,3128 that shows different significant patterns that seen. PCA results showed that primary and secondary principal components accounted for 42.70% and 14.89% of the variance with a cumulative explained variance of 57.58%. K-Means grouping effectively shows variations on these findings based on students’ responses with exponents and roots question. However, the diverse for the result between grouping that was written down, some overlap shows that the patterns are not entirely separate. These clustering results can serve as a basis for more detailed diagnostic investigations into how students respond to mathematical problem-solving.
Copyrights © 2026