Alfariz, Fitria Nurul (2019) Pemodelan Jumlah Anak Putus Sekolah Usia Wajib Belajar Dan Jumlah Wanita Menikah Dini Di Jawa Timur Dengan Pendekatan Geographically Weighted Biva-riate Negative Binomial Regression. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06211540000051-Undergraduate_Theses.pdf Download (2MB) | Preview |
Abstract
Dalam dunia pendidikan telah berlaku kebijakan wajib belajar 12 tahun sejak tahun 2015. Meski peningkatan pendidikan sudah dilakukan oleh pemerintah namun masih banyak terjadi masalah putus sekolah di Indonesia khususnya di Provinsi Jawa Timur. Kondisi ini dipicu oleh beberapa hal diantaranya ekonomi, sosial dan lingkungan tinggal seo-rang anak. Salah satu hal yang berhubungan dengan masalah ini adalah fenomena menikah di usia dini yang juga banyak terjadi di masyarakat. Kedua masalah tersebut dimodelkan menggunakan Geographically Weighted Bivariate Negative Binomial Regression. Sebelum dilakukan pemodelan dengan metode GWBNBR, dilakukan pula pemodelan menggunakan metode BNBR. Dengan metode BNBR didapat 2 variabel prediktor yang berpengaruh terhadap jumlah anak putus sekolah yaitu besaran upah minimum rata-rata dan rata-rata jumlah anggota keluarga. Begitu pula untuk kasus wanita menikah dini terdapat 2 faktor yang berpengaruh yaitu persentase penduduk miskin dan rata-rata jumlah anggota keluarga. Sedangkan pemodelan menggunakan GWBNBR menghasilkan 7 kelompok kabupaten/ kota dengan variabel signifikan yang sama untuk anak putus sekolah dan 5 kelompok kabupaten/kota dengan variabel signifikan yang sama untuk wanita menikah dini. Model terbaik dengan nilai AICc terkecil yang diperoleh adalah model dengan metode BNBR. Namun metode GWBNBR memiliki keunggulan yaitu informasi yang didapat lebih banyak karena pemodelan dilakukan untuk setiap wilayah atau unit penelitian. =================================================================================================================================
9-year compulsory education policy has even been applied and has now become a 12-year compulsory education since 2015. Even though the increase in education has been carried out by the government, there are still many dropouts in Indonesia, especially in East Java Province. This condition is triggered by several things including the economic, social and environmental conditions of a child. One of the things that are related to this problem is the phenomenon of getting married at an early age which also happens a lot in society. Both problems are modeled using Geographically Weighted Bivariate Negative Binomial Regression. Be-fore modeling with the GWBNBR method, modeling was also done using the BNBR method. With the BNBR method, there are 2 predictor variables that influence the number of school dropouts, namely the average mi-nimum wage and the average number of family members. Similarly, there are 2 factors that influence early married women case, they are the percentage of the poor and the average number of family members. While modeling using GWBNBR produces 7 groups of districts/ cities with the same significant variables for school dropouts and 5 districts/cities with the same significant variables for early married women. The best model with the smallest AICc is a model using BNBR method. But the GWBNBR method has the advantage that more information is obtained because modeling is carried out for each region or research unit.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSSt 519.536 Alf p-1 2019 |
| Uncontrolled Keywords: | Heterogenitas, Overdispersi, Putus Sekolah, Spasial, Wanita Menikah Dini |
| Subjects: | Q Science Q Science > QA Mathematics 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: | Fitria Nurul Alfariz |
| Date Deposited: | 21 Dec 2023 03:31 |
| Last Modified: | 21 Dec 2023 03:31 |
| URI: | http://repository.its.ac.id/id/eprint/64089 |
Actions (login required)
 |
View Item |