Faktor yang Berpengaruh terhadap Kepesertaan Metode Kontrasepsi Jangka Panjang (MKJP) di Jawa Timur : Analisis Regresi Logistik Biner SMOTE-NC

Febrianti, Wanda Dwi (2025) Faktor yang Berpengaruh terhadap Kepesertaan Metode Kontrasepsi Jangka Panjang (MKJP) di Jawa Timur : Analisis Regresi Logistik Biner SMOTE-NC. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 5003211041-Undergraduate_Thesis.pdf] Text
5003211041-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only

Download (4MB) | Request a copy

Abstract

Indonesia merupakan negara dengan populasi terbesar keempat di dunia, disertai tingkat kelahiran yang relatif tinggi. Salah satu strategi pemerintah dalam mengendalikan laju pertumbuhan penduduk adalah melalui program Keluarga Berencana (KB), yang mendorong penggunaan metode kontrasepsi, khususnya Metode Kontrasepsi Jangka Panjang (MKJP) seperti IUD, implan, dan sterilisasi. Meskipun lebih efektif dibandingkan metode non-MKJP, angka kepesertaan MKJP di Provinsi Jawa Timur masih rendah, yaitu sebesar 21,3% dari total 3.818.151 wanita kawin pengguna kontrasepsi pada tahun 2021. Mayoritas peserta MKJP menggunakan jenis kontrasepsi IUD. Penelitian ini bertujuan untuk menganalisis faktor-faktor yang memengaruhi kepesertaan MKJP menggunakan regresi logistik biner. Pendekatan Synthetic Minority Oversampling Technique Nominal Continuous (SMOTE-NC) digunakan untuk menangani ketidakseimbangan data. Data bersumber dari Pendataan Keluarga 2021 oleh BKKBN, dengan variabel prediktor mencakup usia wanita dan suami, jumlah anak lahir hidup, jumlah anak ideal, usia anak terakhir, tingkat pendidikan, status bekerja, perolehan pelayanan KB, serta kepesertaan asuransi kesehatan. Hasil analisis menunjukkan bahwa seluruh variabel tersebut berhubungan dan berpengaruh signifikan terhadap kepesertaan MKJP. Wanita kawin peserta MKJP umumnya berusia 35–40 tahun, memiliki dua anak, berpendidikan tamat SLTA/sederajat, bekerja, memiliki suami bekerja, dan memperoleh pelayanan KB dari puskesmas/klinik TNI/Polri serta tercakup dalam asuransi BPJS PBI. Model regresi logistik biner yang diperoleh menunjukkan performa klasifikasi yang baik dengan sensitivitas sebesar 72,1%, spesifisitas 91,67%, G-mean 81,3%, dan AUC 81,88%, sehingga dapat dikategorikan sebagai model yang andal dalam memprediksi penggunaan MKJP di Provinsi Jawa Timur.
======================================================================================================================================
Indonesia has the fourth largest population in the world, with a relatively high birth rate. One of the government's strategies in controlling the population growth rate is through the Family Planning (KB) program, which encourages the use of contraceptive methods, especially Long-Term Contraceptive Methods (MKJP) such as IUDs, implants, and sterilization. Although more effective than non-MKJP methods, the rate of MKJP participation in East Java Province is still low, at 21.3% of the total 3,818,151 married women using contraception in 2021. The majority of LTCM participants use IUD contraception. This study aims to analyze the factors that influence LTCM participation using binary logistic regression. The Synthetic Minority Oversampling Technique Nominal Continuous (SMOTE-NC) approach was used to handle data imbalance. Data was sourced from the 2021 Family Data Collection by BKKBN, with predictor variables including age of the woman and husband, number of children born alive, ideal number of children, age of last child, education level, employment status, acquisition of family planning services, and health insurance membership. The results of the analysis show that all of these variables are related and have a significant effect on the participation of MKJP. Married women participating in MKJP are generally 35-40 years old, have two children, have a high school education or equivalent, work, have a working husband, and obtain family planning services from puskesmas/clinics of TNI/Polri and are covered by BPJS PBI insurance. The binary logistic regression model obtained shows good classification performance with a sensitivity of 72.1%, specificity of 91.67%, G-mean 81.3%, and AUC 81.88%, so it can be categorized as a reliable model in predicting the use of MKJP in East Java Province.

Item Type: Thesis (Other)
Uncontrolled Keywords: Metode Kontrasepsi Jangka Panjang, Regresi Logistik Biner, SMOTE-NC, Binary Logistic Regression, Long-Term Contraceptive Methods, SMOTE-NC
Subjects: 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: Wanda Dwi Febrianti
Date Deposited: 01 Aug 2025 06:23
Last Modified: 01 Aug 2025 06:23
URI: http://repository.its.ac.id/id/eprint/125496

Actions (login required)

View Item View Item