Pemodelan Status NEET dengan Pendekatan Regresi Logistik Biner Multilevel

Wijayanti, Maharani (2026) Pemodelan Status NEET dengan Pendekatan Regresi Logistik Biner Multilevel. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003211004-Undergraduate_Thesis.pdf - Accepted Version Restricted to Registered users only Download (4MB) | Request a copy

Abstract

Not in Education, Employment, or Training (NEET) merupakan indikator penting untuk mengukur kerentanan dan potensi produktivitas pemuda. Sebagai salah satu provinsi dengan persentase NEET tertinggi, Provinsi Jawa Barat memiliki variasi persentase NEET cukup besar antar kabupaten/kota, berkisar antara 12,13% hingga 32,34%. Variasi geografis yang signifikan ini mengindikasikan bahwa karakteristik wilayah (level 2) kemungkinan memengaruhi status NEET di samping faktor individu (level 1). Penelitian ini bertujuan untuk menganalisis determinan status NEET pemuda di Jawa Barat tahun 2024 dengan menggunakan pendekatan regresi logistik biner multilevel. Model hierarkis dua level dibangun dengan individu sebagai level 1 dan kabupaten/kota sebagai level 2. Data yang digunakan bersumber dari Survei Angkatan Kerja Nasional (SAKERNAS) Agustus 2024. Dari hasil penelitian didapatkan bahwa model dengan random slope tingkat pendidikan merupakan model terbaik dengan enam variabel prediktor level 1 signifikan, yaitu kategori usia, status perkawinan, tingkat pendidikan, pengalaman pelatihan, penguasaan teknologi digital, serta klasifikasi wilayah tempat tinggal. Sementara variabel UMK dan IPM merupakan variabel signifikan pada level 2. Model terbaik berupa model random slope tingkat pendidikan memiliki nilai ICC sebesar 1,2%, akurasi sebesar 65,15%, sensitivitas 71,47%, dan spesifisitas 63,433%.
==================================================================================================================================
Not in Education, Employment, or Training (NEET) is a key indicator for measuring the vulnerability and productive potential of young people. As one of the provinces with the highest NEET rates, West Java Province exhibits significant variation in NEET rates across its districts and cities, ranging from 12.13% to 32.34%. This significant geographical variation indicates that regional characteristics (level 2) likely influence NEET status in addition to individual factors (level 1). This study aims to analyze the determinants of youth NEET status in West Java in 2024 using a multilevel binary logistic regression approach. A two-level hierarchical model was constructed with the individual as level 1 and the district/city as level 2. The data used were sourced from the August 2024 National Labor Force Survey (SAKERNAS). The results indicate that the model with a random slope for educational level is the best model, featuring six significant Level 1 predictor variables: age category, marital status, educational level, training experience, digital technology proficiency, and residential area classification. Meanwhile, the UMK and HDI variables are significant at Level 2. The best model which is a random-slope model for educational level has an ICC value of 1,2%, an accuracy of 65,15%, a sensitivity of 71,47%, and a specificity of 63,43%.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	NEET, Jawa Barat, regresi logistik biner multilevel, random slope,kabupaten/kota, West Java, multilevel binary logistic regression, random slope, regencies/municipalities
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:	Maharani Wijayanti
Date Deposited:	01 Aug 2026 03:40
Last Modified:	01 Aug 2026 03:40
URI:	http://repository.its.ac.id/id/eprint/140321

Actions (login required)

View Item