<![CDATA[This study aims to identify the effect of school origin background on the Cumulative Grade Point Average (GPA) cluster pattern of students admitted through the National Selection Based on Academic Achievement (SNBP) at Politeknik Negeri Jakarta (PNJ). The main problem in this study is the lack of clarity regarding the effect of school origin on student academic achievement. This study applies the total population sampling technique with all PNJ students class of 2023 totaling 2.527 for the formation of the GPA cluster, while the subgroup with student data accepted through the SNBP in 2023 has complete school origin data totaling 888 students. Student GPA data is clustered using the K-Means algorithm into four groups, namely very high, high, sufficient, and low GPA. Evaluation of cluster quality was conducted using Silhouette Coefficient and Davies-Bouldin index, while analysis of the relationship between school origin and GPA was tested through Chi-Square and One-Way ANOVA tests. The results showed that the K-Means algorithm was able to produce good quality GPA clusters (Silhouette Coefficient 0.55). One-Way ANOVA showed that the average GPA differed significantly but with little impact, both according to school type (SMA, SMK, MA) and school management status (public and private). Chi-Square tests confirmed school type was not associated with GPA clustering (p = 0.107), while public school origin showed slightly higher GPAs than private schools. The findings recommend that school origin should be used as a supporting factor in the SNBP selection policy, but not as the main criterion.]]>
Copyrights © 2025
