The Modifiable Areal Unit Problem in the Analysis of Brazilian Internal Migrations

Abstract

As a source of potential biases and errors, the Modifiable Areal Unit Problem (MAUP) poses a major challenge for all those working with spatial data in general, and more specifically, for migration scholars. The goal of this article is to discuss and measure the scale and zonation effects in the analysis of migration indicators, considering the dimensions of intensity, impact, connectivity and distance at different scales and spatial configurations. In addition to the effects related to the use of indicators at different levels of spatial aggregation (official political-administrative boundaries), aggregation and spatial modeling functions of the IMAGE Studio software were used. The results revealed strong scale and zonation effects, including significant variations in indicators regarding corresponding aggregation levels.