STUDY OF SMALL AREA ESTIMATION TO ESTIMATE UNEMPLOYMENT RATE
Statistics Indonesia (BPS) is nationally responsible for the availability of data required for sectoral and cross sectoral development planning in Indonesia. The main source of employment data in Indonesia is National Labour Force Survey (Sakernas). Since 2015 Sakernas has been conducted in semester period, i.e. February and August. Sakernas February to produces employment indicators up to province level, while Sakernas August to district/regency level. In 2016, Sakernas August not produced employment indicators in district/regency level because inadequate size samples. The Small Area Estimation (SAE) method is an alternative that can be used to estimate the parameters in an area when the sample size in the area is too small to obtain adequate precision compared to the direct estimation. This research applies the proposed SAE Zero Inflated Binomial (ZIB) method to unemployment data at the village unit level and estimate proportion of unemployment in each district/regency during period August 2016 in Jambi Province, Indonesia. The application to unemployment data results that the proportion estimate of SAE ZIB method is similar with the current publication (4 unemployments of 100 labor forces) in Province Jambi level and can produces employment indicators during period August 2016 in district/regency level.