SUMMARY
Infant Mortality Rate (IMR) is an important indicator because it can be used to compare health status between populations. IMR is obtained from Demographic and Health Survey (DHS) where the level of estimation is designed for national and provincial level. The decentralization system makes the importance of IMR for sub-domain of province such as district/municipality level. Small area estimation (SAE) can be used for estimating IMR in district/municipality level by using a mixed model. IMR is count data with small probability, so the distribution is Poisson. Poisson model assumes that mean equal to variance, but this assumption is often violated. One of the reasons is excess zero. In this study, Zero Inflated Poisson (ZIP) mixed model is much better than Poisson mixed model and can improve the direct estimation of IMR.