Testing Small Area Estimation (SAE) Method for Generating Nutrition Maps in Indonesia


ABSTRACT

This study aimed to test the small area estimation (SAE) method derived from Elbers, Lanjouw, and Lanjouw (ELL) in estimating the nutritional status of children under five years of age at the village and subdistrict levels in five districts: Lampung Tengah, Tasikmalaya, Pemalang, Jember, and Timor Tengah Selatan (TTS). The estimation results were verified quantitatively using an anthropometric census and household-level interviews, and qualitatively using in-depth interviews and village observations. The stunting, underweight, and wasting estimation results at the district level show no significant difference from the benchmark based on the 2013 Riskesdas. The average stunting prevalence obtained from the 2019 verification in the villages is relatively 20%–70% lower than the 2013 stunting prevalence. After an estimation model test was carried out, the decrease of the stunting prevalence proved that the SAE model is strong enough to estimate the 2013 prevalence. Such decrease may have been influenced by changes in demographic and socioeconomic characteristics between 2013–2019. Factors affecting the nutritional status include clean water and sanitation, parental educational attainment, household welfare, and villages’ commitment to stunting prevention. On the other hand, factors known to contribute to stagnancy or low stunting prevalence declines are, among other things, unhealthy lifestyles and food vulnerability. The maps generated from this study serve as prototypes in expanding nutritional status estimation to all villages in Indonesia as part of the National Strategy in Accelerating Stunting Prevention. 
  
Keywords: small area estimation (SAE), Elbers, Lanjouw, and Lanjouw (ELL), stunting, wasting, underweight, nutritional status maps


Share it