Analisis Ketimpangan Pendidikan di Indonesia Berdasarkan Rata-rata Lama Sekolah Menggunakan Metode Geographically Weighted Panel Regression

Ningtiyas, Irma (2026) Analisis Ketimpangan Pendidikan di Indonesia Berdasarkan Rata-rata Lama Sekolah Menggunakan Metode Geographically Weighted Panel Regression. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2043221013-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Pendidikan merupakan salah satu pilar utama pembangunan nasional dan kunci peningkatan kualitas sumber daya manusia. Namun ketimpangan pendidikan masih menjadi permasalahan di Indonesia yang tercermin dari perbedaan capaian rata-rata lama sekolah antar wilayah. Data Badan Pusat Statistik menunjukkan bahwa capaian rata-rata lama sekolah (RLS) di Indonesia masih berada pada tingkat wajib belajar sembilan tahun, dengan disparitas yang cukup besar antarprovinsi. Sebagian wilayah telah mencapai jenjang menengah atas khususnya daerah perkotaan, sementara wilayah lainnya masih tertinggal pada pendidikan dasar. Oleh karena itu, diperlukan analisis yang memperhatikan heterogenitas spasial dan dinamika temporal agar hasil analisis ketimpangan pendidikan dapat lebih spesifik untuk perumusan kebijakan pemerataan pendidikan di setiap wilayah. Penelitian ini menggunakan model Geographically Weighted Panel Regression (GWPR) untuk menganalisis ketimpangan pendidikan berdasarkan rata-rata lama sekolah di 34 provinsi Indonesia pada periode 2020–2024. Hasil pemetaan menunjukkan bahwa ketimpangan pendidikan di Indonesia berdasarkan rata-rata lama sekolah berbeda antarprovinsi, di mana wilayah barat Indonesia dan daerah perkotaan seperti DKI Jakarta dan Kepulauan Riau memiliki RLS tinggi dibandingkan wilayah timur dan provinsi dengan keterbatasan infrastruktur seperti Nusa Tenggara, Papua, dan Kalimantan Barat. Berdasarkan hasil analisis GWPR diperoleh bahwa faktor-faktor yang berpengaruh signifikan terhadap RLS berbeda-beda tiap provinsi meliputi rasio murid–guru, angka partisipasi sekolah, realisasi dana BOS, dan tingkat pengangguran terbuka. Hasil pemodelan GWPR menunjukkan kinerja yang lebih baik dibandingkan model regresi data panel fixed effect dengan nilai R² sebesar 90,82% yang artinya variabel prediktor mampu menjelaskan rata-rata lama sekolah sebesar 90,82%, sisanya sebesar 9,18% dijelaskan variabel lain di luar model.
===================================================================================================================================
Education is one of the main pillars of national development and the key to improving the quality of human resources. However, educational inequality remains a problem in Indonesia, as reflected in the differences in average length of schooling between regions. Data from the Central Statistics Agency shows that the average length of schooling in Indonesia is still at the level of nine years of compulsory education, with considerable disparities between provinces. Some regions have reached upper secondary level, especially urban areas, while others are still lagging behind in basic education. Therefore, an analysis that takes into account spatial heterogeneity and temporal dynamics is needed so that the results of the analysis of educational inequality can be more specific for the formulation of policies on educational equity in each region. This study uses the Geographically Weighted Panel Regression (GWPR) model to analyze educational inequality based on the average length of schooling in 34 provinces in Indonesia for the period 2020–2024. The mapping results show that educational inequality in Indonesia based on average length of schooling varies between provinces, with western Indonesia and urban areas such as DKI Jakarta and Riau Islands having high RLS compared to eastern regions and provinces with limited infrastructure such as Nusa Tenggara, Papua, and West Kalimantan. Based on the GWPR analysis results, it was found that the factors that significantly influence RLS vary from province to province, including the student-teacher ratio, school participation rate, realization of BOS funds, and open unemployment rate. The GWPR modeling results show better performance than the fixed effect panel data regression model with an R² value of 90.82%, which means that the predictor variables are able to explain the average length of schooling by 90.82%, with the remaining 9.18% explained by other variables outside the model.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Geographically Weighted Panel Regression, Ketimpangan Pendidikan, Rata-Rata Lama Sekolah, Geographically Weighted Panel Regression, Educational Inequality, Average Length of Schooling
Subjects:	H Social Sciences > HA Statistics > HA30.6 Spatial analysis H Social Sciences > HA Statistics > HA31.3 Regression. Correlation. Logistic regression analysis.
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Irma Ningtiyas
Date Deposited:	31 Jan 2026 06:39
Last Modified:	31 Jan 2026 06:39
URI:	http://repository.its.ac.id/id/eprint/131398

Actions (login required)

View Item