Pendekatan Zero-Inflated Ordered Probit Dalam Mengidentifikasi Faktor-Faktor Yang Memengaruhi Prevalensi Stunting Baduta Di Provinsi Banten

Apsara, Alyda (2026) Pendekatan Zero-Inflated Ordered Probit Dalam Mengidentifikasi Faktor-Faktor Yang Memengaruhi Prevalensi Stunting Baduta Di Provinsi Banten. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003221006-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Stunting pada baduta merupakan salah satu permasalahan gizi yang masih menjadi tantangan dalam pembangunan sumber daya manusia di Indonesia. Meskipun prevalensi stunting nasional berdasarkan Survei Status Gizi Indonesia (SSGI) 2024 mengalami penurunan menjadi 19,8%, Provinsi Banten masih memiliki prevalensi stunting yang relatif tinggi dibandingkan provinsi lain di Pulau Jawa. Kondisi tersebut menunjukkan perlunya identifikasi faktor-faktor yang memengaruhi status stunting baduta sebagai dasar dalam penyusunan kebijakan penanggulangan stunting yang lebih efektif. Penelitian ini bertujuan untuk menganalisis faktor-faktor yang memengaruhi status stunting baduta di Provinsi Banten menggunakan pendekatan Zero-Inflated Ordered Probit (ZIOP). Model ini dipilih karena mampu mengakomodasi dominasi kategori normal pada data status stunting sekaligus memodelkan tingkat keparahan stunting yang bersifat ordinal. Data yang digunakan merupakan data sekunder SSGI Tahun 2024 dengan jumlah sampel sebanyak 1.181 baduta di Provinsi Banten. Variabel prediktor yang digunakan meliputi jenis kelamin baduta, usia ibu saat mulai kehamilan, jarak kehamilan, pemeriksaan kehamilan (antenatal care), pemberian ASI eksklusif, usia pemberian makanan pendamping ASI (MP-ASI), status imunisasi dasar lengkap, dan kejadian infeksi saluran pernapasan akut (ISPA). Hasil analisis menunjukkan bahwa pada komponen inflated, variabel usia ibu saat mulai kehamilan, jarak kehamilan, dan usia pemberian MP-ASI berpengaruh signifikan terhadap kecenderungan baduta berada pada kategori normal. Sementara itu, pada komponen ordinal, variabel jenis kelamin baduta berpengaruh signifikan terhadap tingkat keparahan status stunting. Berdasarkan hasil uji Vuong, model ZIOP terbukti lebih baik dibandingkan model probit ordinal dalam memodelkan status stunting baduta di Provinsi Banten.
========================================================================================================================================
Stunting among children under two remains one of the major nutritional challenges in Indonesia's human resource development. Although the national prevalence of stunting based on the 2024 Indonesian Nutritional Status Survey (SSGI) decreased to 19.8%, Banten Province continues to exhibit a relatively high prevalence compared to other provinces on Java Island. This condition highlights the need to identify factors associated with stunting status among children under two as a basis for developing more effective stunting reduction policies. This study aims to analyze the factors influencing stunting status among children under two in Banten Province using the Zero-Inflated Ordered Probit (ZIOP) model. The ZIOP model was selected because it can accommodate the dominance of the normal category in stunting data while simultaneously modeling the ordinal severity levels of stunting. The study utilized secondary data from the 2024 SSGI, comprising 1.181 children under two in Banten Province. The predictor variables included child sex, maternal age at the beginning of pregnancy, pregnancy interval, antenatal care (ANC), exclusive breastfeeding, age at introduction of complementary feeding, complete basic immunization status, and incidence of acute respiratory infection (ARI). The results indicate that, in the inflation component, maternal age at the beginning of pregnancy, pregnancy interval, and age at introduction of complementary feeding significantly affect the likelihood of a child being classified in the normal category. Meanwhile, in the ordinal component, child sex significantly influences the severity level of stunting status. Based on the Vuong test, the ZIOP model outperformed the conventional ordered probit model in modeling stunting status among children under two in Banten Province.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Regresi Probit, Regresi Zero-Inflated Ordered Probit, Status Gizi Baduta, Stunting, Child Nutrition Status, Probit Regression, Stunting, Zero-Inflated Ordered Probit
Subjects:	H Social Sciences > HA Statistics > HA31.3 Regression. Correlation. Logistic regression analysis.
Divisions:	Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Alyda Apsara
Date Deposited:	30 Jul 2026 01:34
Last Modified:	30 Jul 2026 01:34
URI:	http://repository.its.ac.id/id/eprint/140264

Actions (login required)

View Item