Comparison of Two Linear Regression Models for Predicting the Literacy Development Index in Indonesia

Iffatu Wardani; Kunto  Jiwandono; Okta Dyah Pradanti; Yuni Wahyu Winjarwati; Stevano Aji  Aghashie

doi:10.47709/brilliance.v5i2.7083

Authors

Iffatu Wardani Institut Teknologi dan Bisnis Trenggalek, Indonesia
Kunto Jiwandono Institut Teknologi dan Bisnis Trenggalek, Indonesia
Okta Dyah Pradanti Institut Teknologi dan Bisnis Trenggalek, Indonesia
Yuni Wahyu Winjarwati Institut Teknologi dan Bisnis Trenggalek, Indonesia
Stevano Aji Aghashie Institut Teknologi dan Bisnis Trenggalek, Indonesia

DOI:

https://doi.org/10.47709/brilliance.v5i2.7083

Keywords:

Correlation, Machine Learning, Multiple Linear Regression, SImple Linear Regression

Abstract

This study examines four suspected factors that have correlation and influence Community Literacy Development Index (IPLM). The four factors data was taken from each province in Indonesia i.e. the number of accredited libraries, the level of people’s reading interest, proportion of population living below 50% of the median income, and high school completion rate. To determine whether these four factors truly affect IPLM, a regression model analysis was conducted. The machine learning models discussed in this study are simple linear regression and multiple linear regression. One multiple linear regression model was used to integrate all four factors together. Four simple linear regression models were applied to assess each factor individually in relation to IPLM. From all these regression models, the adjusted R-squared values were compared. The analysis revealed that the level of people’s reading interest factor has a higher adjusted R-squared value in the simple linear regression (0.3828) compared to the multiple linear regression (0.3235). In contrast, the other three factors show lower adjusted R-squared values in their simple linear regressions than in the multiple linear regression. The conclusion is the reading interest factor best used to predict IPLM without involving the other factors. Meanwhile, the remaining three factors should be used collectively when predicting IPLM values.

References

Alkawaz, Ali Najem et al. 2022. Day-Ahead Electricity Price Forecasting Based on Hybrid Regression Model. IEEE Access 10(September): 108021–33.

Alshanqiti, A., & Namoun, A. (2020). Predicting Student Performance and Its Influential Factors Using Hybrid Regression and Multi-Label Classification. IEEE Access 8: 203827–44.

Aprihartha, M.A., Azzahro, S.P., & Aziza, R. (2025). Pemilihan Model Regresi Linear Berganda Terbaik Untuk Menentukan Faktor-Faktor Penyebab Kasus Balita Gizi Buruk Di Jawa Tengah. Jurnal EurekaMatika 13(1): 35–46.

Kalla Institute (2024). Rendahnya Minat Literasi di Indonesia. Accesssed: Aug 30, 2025. https://kallainstitute.ac.id/rendahnya-minat-literasi-di-indonesia/.

Karch, Julian. (2020). Improving on Adjusted R-Squared. Collabra: Psychology 6(1): 1–11.

Kusuma, P. D., (2020). Machine Learning Teori, Program, Dan Studi Kasus. Yogyakarta: Deepublish.

Lesnusa, G.N., Angreni, D.S., & Ardiansyah, R. (2024). Perbandingan Akurasi Linear Regression Dan Support Vector Regression Dalam Prediksi Suhu Rata-Rata. The Indonesian Journal of Computer Science 13(4): 6112–18.

Maulana, A., Martanto, M., & Ali, I. (2024). Prediksi Hasil Produksi Panen Bawang Merah Menggunakan Metode Regresi Linier Sederhana. JATI (Jurnal Mhs. Tek. Inform., vol. 7, no. 4, pp. 2884–2888

Ministry of Communication and Digital (2020). Teknologi Masyarakat Indonesia: Malas Baca tapi Cerewet di Medsos. Accessed: Aug 30, 2025. https://www.komdigi.go.id/berita/sorotan-media/detail/teknologi-masyarakat-indonesia-malas-baca-tapi-cerewet-di-medsos

Nasrullah, R., & Asmarini, P., (2024). Meningkatkan Literasi Indonesia Melalui Optimalisasi,” Badan Pengemb. dan Pembin. Bhs. Risal. Kebijak., no. 4, pp. 1–16.

National Library of the Republic of Indonesia (2024). IPLM 2024 Catat Rekor Tinggi, Literasi Nasional semakin Meningkat. Accessed: Aug 30, 2025. https://www.perpusnas.go.id/berita/iplm-2024-catat-rekor-tinggi-literasi-nasional-semakin-meningkat

Nuraini, A.T., Setiawan, A., & Susanto, B. (2023), Perbandingan Kinerja Regresi Decision Tree dan Regresi Linear Berganda untuk Prediksi BMI pada Dataset Asthma, Jurnal Sains dan Edukasi Sains, vol. 6, no. 1, pp. 34–43.

Nuris, Nuzuliarini. (2024). Analisis Prediksi Harga Rumah Pada Machine Learning Metode Regresi Linear. Explore 14(2): 108–12.

Prakash, Kolla Bhanu. (2022). Data Science Handbook: A Practical Approach. Wiley AI.

Prasmono, A.S.P., & Ahdika, A. (2023). Analisis Regresi Berganda Pada Faktor-Faktor Yang Mempengaruhi Kinerja Fisik Preservasi Jalan Dan Jembatan Di Provinsi Sumatera Selatan. Emerging Statistics and Data Science Journal 1(1): 47–56.

Soleh, M., Nurnawati, Kumalasari, E., & Uning, L. (2023). Penerapan Data Mining Dengan Metode Regresi Linear Untuk Memprediksi Data Nilai Hasil Ujian Menggunakan RapidMiner. JISKA (Jurnal Informatika Sunan Kalijaga) Vol. 8, No(ISSN:2527–5836 (print) | 2528–0074 (online)): Pp. 10 – 21.

Tatachar, Abhishek V. (2021). Comparative Assessment of Regression Models Based On Model Evaluation Metrics. International Research Journal of Engineering and Technology 8(9): 853–60. www.irjet.net.

Thabibi, A., & Supriyanto, R. (2023). Perbandingan Model Multiple Linear Regression Dan Decision Tree Regression (Studi Kasus: Prediksi Harga Saham Telkom, Indosat, Dan Xl), Jurnal Ilmu Teknologi dan Rekayasa, vol. 28, no. 1, pp. 78–92.

Zhao, Guangcai et al. 2023. State-of-Health Estimation With Anomalous Aging Indicator Detection of Lithium-Ion Batteries Using Regression Generative Adversarial Network. IEEE Transactions on Industrial Electronics 70(3): 2685–95.