Background: Daily energy requirements must be fulfilled for a healthy and active life. Prevalence of Undernourishment (PoU) is an indicator which are developed for goal. A number of factors were expected affecting PoU were tested in this study. We also tested the possibility of these factors could have a spatial correlationObjectives: The study produced a map of the spread of PoU and the factors which were influenced them  (independent variables). The study need to yield the best estimation model.Method : This study produce spread map of all variables, for visible purposes, and the classical regression were  made. All OLS assumption will be test. Then, SAR and SEM models will be made. Finally, the best model for this study will be chosen. GeoDa software helps all steps.Results : This study concludes PoU has positive auto correlation and growth has a negative autocorrelation. The best model which produced is the Spatial Error Model (SEM). The slow trend of PoU in West Sumatera Barat is strongly suspected due to the habit of West Sumatera residents in consuming high-calorie foods such as rendang, curry and coconut milk.

The Human Development Index (HDI) is calculated from four variables, however, the HDI value cannot directly show which variables are superior in an area and which are not. For this reason, this study aims to analyze the strength of the HDI variables in each region in Sumatera Barat Province, using Biplot Analysis. Important findings resulted from biplot analysis indicate that the Sumatera Barat Province is categorized into five regional groups that are affected by the proximity value of its constituent variables. Out of the five groups, two groups are in an area not adjacent to the variable. The other two groups are in the variable area. While the last group is an extreme region compared to other regional groups. Areas that are not in the area of variables are weak against variables, while the area falling in the variable area is superior in the relevant variable. For extreme groups, Kota Padang is very superior in the adjusted per capita expenditure variable and superior in other variables too. Meanwhile, the Mentawai Islands Regency is very weak in all variables. Kota Padang Panjang is very superior in all variables of education but weak in the economic variable. The Biplot Analysis also revealed that Sijunjung Regency had a fairly good in expenditure, but it was not good enough in other dimensions of the HDI. Things like this are not given enough attention to ordinary HDI analysis, this is what makes the Biplot analysis important to do.