Pemodelan Kasus Stunting di Kota Surabaya dengan Metode Geographically Weighted Generalized Poisson Regression

Amalia, Ahyani Tammami (2026) Pemodelan Kasus Stunting di Kota Surabaya dengan Metode Geographically Weighted Generalized Poisson Regression. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003221176-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (7MB) | Request a copy

Abstract

Stunting merupakan salah satu permasalahan kesehatan masyarakat yang masih menjadi prioritas di Indonesia karena berdampak terhadap kualitas sumber daya manusia di masa mendatang. Meskipun prevalensi stunting di Kota Surabaya tergolong rendah dibandingkan daerah lain, jumlah kasus stunting antarwilayah kerja puskesmas menunjukkan variasi yang cukup besar. Penelitian ini bertujuan untuk menganalisis karakteristik jumlah kasus stunting beserta faktor-faktor yang diduga memengaruhinya serta memperoleh model terbaik menggunakan Geographically Weighted Generalized Poisson Regression (GWGPR) di Kota Surabaya. Hasil penelitian menunjukkan bahwa data mengalami overdispersi dan heterogenitas spasial. Model GWGPR dengan pembobot Adaptive Bisquare Kernel menghasilkan nilai AICc sebesar 425,0843 sehingga dipilih sebagai model terbaik. Hasil pemodelan mengelompokkan 63 wilayah kerja puskesmas ke dalam dua kelompok berdasarkan variabel yang berpengaruh signifikan. Variabel akses sanitasi aman dan pemberian vitamin A pada balita merupakan variabel signifikan pada pemodelan ini.
=======================================================================================================================================
Stunting remains one of the major public health concerns in Indonesia due to its long-term impact on the quality of human resources. Although the prevalence of stunting in Surabaya is relatively low compared to other regions, the number of stunting cases varies considerably across public health center (Puskesmas) service areas. This study aims to analyze the characteristics of stunting cases and the factors influencing them, as well as to determine the best model using Geographically Weighted Generalized Poisson Regression (GWGPR) in Surabaya. The results indicate that the data exhibit overdispersion and spatial heterogeneity. The GWGPR model with the Adaptive Bisquare Kernel weighting function produced the lowest Akaike Information Criterion corrected (AICc) value of 425.0843, making it the best-fitting model. The modeling results classified the 63 public health center service areas into two groups based on the significant influencing variables. Safe sanitation access and vitamin A supplementation among children under five were identified as significant variables affecting the number of stunting cases.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Geographically Weighted Generalized Poisson Regression, overdispersi, stunting, Overdispersion
Subjects:	H Social Sciences > HA Statistics > HA30.6 Spatial analysis H Social Sciences > HA Statistics > HA31.3 Regression. Correlation. Logistic regression analysis. Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Ahyani Tammami Amalia
Date Deposited:	04 Aug 2026 03:02
Last Modified:	04 Aug 2026 03:02
URI:	http://repository.its.ac.id/id/eprint/139979

Actions (login required)

View Item