Pemodelan Georaphically Weighted Generalized Poisson Regression pada Kasus Stunting di Provinsi Jawa Timur

Ibnatia, Syafa Aulia (2024) Pemodelan Georaphically Weighted Generalized Poisson Regression pada Kasus Stunting di Provinsi Jawa Timur. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003201067-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (2MB) | Request a copy

Abstract

Masalah stunting masih menjadi sebuah tantangan yang cukup serius dalam pembangunan kesehatan dan pertumbuhan anak di Indonesia. Fenomena ini menyebabkan anak-anak mengalami pertumbuhan fisik yang tidak optimal sehingga mereka cenderung lebih pendek. Tingginya kasus stunting pada beberapa wilayah di Jawa Timur merupakan salah satu bukti nyata bahwa perlu adanya penelitian mendalam untuk mengidentifikasi faktor-faktor penyebab stunting pada setiap wilayah di Jawa Timur. Setiap wilayah tentu memiliki kondisi geografis yang berbeda-beda. Oleh karena itu, pada penelitian ini digunakan metode Geographically Weighted Generalized Poisson Regression (GWGPR) yang dapat menangani kasus overdispersi sekaligus mengatasi keragaman wilayah agar mampu menghasilkan model yang sesuai dengan karakteristik setiap wilayah. Selain itu, dipertimbangkan pula penggunaan variabel exposure untuk mengakomodasi perbedaan jumlah balita pada setiap wilayah. Jumlah balita stunting tertinggi adalah sebanyak 9.666 balita di Kabupaten Probolinggo, sedangkan jumlah balita stunting terendah adalah sebanyak 172 balita di Kota Mojokerto. Hasil pemodelan terbaik dari kriteria AICc paling minimum sebesar 648,314 adalah menggunakan metode GWGPR dengan exposure yang menghasilkan tiga kelompok kabupaten/kota berdasarkan variabel yang berpengaruh signifikan terhadap jumlah kasus stunting di Provinsi Jawa Timur. Variabel yang berpengaruh signifikan tersebut adalah persentase KK stop buang air besar sembarangan dan persentase KK akses rumah sehat.

============================================================
============================================================

The problem of stunting is still a serious challenge in the development of children's health and growth in Indonesia. This phenomenon causes children to experience physical growth that is not optimal so they tend to be shorter. The high number of stunting cases in several regions in East Java is one clear evidence that in-depth research is needed to identify the factors that cause stunting in each region in East Java. Each region certainly has different geographical conditions. Therefore, in this study, the Geographically Weighted Generalized Poisson Regression (GWGPR) method is used which can handle overdispersion cases while overcoming regional diversity in order to be able to produce models that suit the characteristics of each region. In addition, the use of variable exposure is also considered to accommodate differences in the number of toddlers in each region. The highest number of stunting toddlers is 9.666 toddlers in Probolinggo Regency, while the lowest number of stunting toddlers is 172 toddlers in Mojokerto City. The best modeling result from the minimum AICc criteria of 648,314 is using the GWGPR method with exposure which produces three district/city groups based on variables that have a significant effect on the number of stunting cases in East Java Province. The variables that have a significant effect are the percentage of KK stopping open defecation and the percentage of KK accessing healthy homes.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Balita, GWGPR, Jawa Timur, Spasial, Stunting, East Java, Spatial, Stunting, Toddler
Subjects:	H Social Sciences > HA Statistics > HA30.6 Spatial analysis
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Syafa Aulia Ibnatia
Date Deposited:	08 Aug 2024 07:27
Last Modified:	08 Aug 2024 07:27
URI:	http://repository.its.ac.id/id/eprint/114608

Actions (login required)

View Item