Random Effect Model pada Regresi Panel untuk Pemodelan Kasus Gizi Buruk Balita di Jawa Timur Tahun 2013–2016

Irma Ike Wahyuni (Departemen Biostatistika dan Kependudukan, Fakultas Kesehatan Masyarakat, Universitas Airlangga)
Mahmudah Mahmudah (Departemen Biostatistika dan Kependudukan, Fakultas Kesehatan Masyarakat Universitas Airlangga)

Jurnal Biometrika dan Kependudukan (Journal of Biometrics and Population)
Vol. 6 No. 1 (2017): JURNAL BIOMETRIKA DAN KEPENDUDUKAN

Severe Malnutrition among under five still become a problem in East Java. Regression analysis that has been done only use data in one year so it can not see the effect of time. In this research, the analysis was done by panel data that combine data from some location that was observed in several years so the factor that was caused severe malnutrition was known. The data that was used are secondary data from East Java Province Health Office which consist of under five severe malnutrition percentage data, percentage of under five health service, percentage of healthy house, and percentage of village with Universal Child Immunization (UCI) at 38 regencies/city in 2013–2016. This research used panel regression analysis with Random Effect Model. The result showed that the percentage of village with UCI status significantly influence under five severe malnutrition in East Java with signification value was 0.0008. The regression coefficient value of UCI variable was -0.190. Therefore, the number of village with UCI status need to be increased especially in regencies/city with high under five severe malnutrition cases by Integrated Service Post (Under Five Health Service) activity.