625 Where is the Poverty Area? Quantifying the Neighborhood Effect in a Deprivation Index Estimation: A case Study in Quito, Ecuador Chunzhu Wei, Pablo Cabrera Barona and Thomas Blaschke Department of Geoinformatics – Z_GIS, University of Salzburg, Austria · chunzhu.wei@stud.sbg.ac.at Full paper double blind review Abstract Automated zoning procedures offer efficient, systematic and objective methodologies for identifying the neighborhood effects on socio-economic statistics. However, the automatic spatial aggregation of census data over manually defined geographic units based on land- scape heterogeneity characteristics are barely studied. In this study we utilize high-resolu- tion remote sensing data and census data and apply a multi-level zoning system in order to analyze how a deprivation index differs in the corresponding urbanization environment within the Shannon’s diversity Index. Our study area is the capital city of Ecuador, Quito. The results of the autocorrelation analysis show that, within the Shannon’s Diversity-based multi-level zoning system, areas with a low degree of deprivation in the city center of Quito tend to be larger as the size of the neighborhood increases, and the poverty areas, which are mainly located in the north-east and south-west of Quito, differ significantly between dif- ferent zoning levels. Our conclusion is that the neighborhood effect influences not only the composition of the spatial pattern and social data, but also their correlation and autocorrela- tion. Therefore, when analyzing the environment effect of urbanization and its influence on the society development, different levels of zoning systems should be taken into considera- tion. 1 Introduction In order to deal with the Modifiable Areal Unit Problem (MAUP) (OPENSHAW 1984) asso- ciated with aggregated census data, automated zoning procedures have offered several methods for zoning boundary identification and have allowed the investigation of the neighborhood effect on health statistics, human distribution and choosing a regional-build- ing area (FLOWERDEW 2008, MARTIN 2003, OPENSHAW 1977). Some landscape ecologists also assume that ecological processes affect the neighborhood context within a zone and the interaction between zones, which means that environmental heterogeneity may impact socio-economic variables (Wagner and Fortin 2005). Nevertheless, most design criteria that have been used to define multi-scale zoning systems for census data aggregation only take the social-economic factors into consideration, such as population size, deprivation, homo- geneity of ill-health dynamics policy intervention policies (COCKINGS & MARTIN 2005) GI_Forum ‒ Journal for Geographic Information Science, 1-2015. © Herbert Wichmann Verlag, VDE VERLAG GMBH, Berlin/Offenbach. ISBN 978-3-87907-558-4. © ÖAW Verlag, Wien. ISSN 2308-1708, doi:10.1553/giscience2015s625. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).