Exploring the MAUP from a spatial perspective

Abstract

This paper deals with evaluating and exploring the modifiable areal unit problem (MAUP) from a spatial analysis perspective for socio-economic data. We introduce a semivariogram model for aggregated data, and explore relationships using the model at different levels of aggregation. Using different scales and zoning on a particular set of spatial data may leads to problems in interpreting the results. In practice, organizations publish statistics and maps at a particular area level. Pawitan and Steel (2006) explored the bias associated with aggregate level semivariogram analysis compared with the corresponding individual level analysis. Minot and Baulch (2005a) discussed some consequences of using aggregated level data in poverty mapping, which may affect the validity of the output. Openshaw (1984) noted that the results of the analysis of aggregated spatial data will depend on the scale and zoning being used. These two factors potentially affect analysis of spatially aggregated data and are referred to as the modifiable areal unit problem or MAUP. This paper will examine the topic theoretically and empirically based on survey and census data, as well as simulations. Discussion focuses on the expectation of differences between analysis undertaken at two different aggregation levels. Relationships between aggregation effects for some statistics will be discussed from a spatial perspective.