Abstract
This paper considers the modifiable areal unit problem, a long-standing issue in geography, but also one of importance for all social researchers concerning the level of aggregation at which their analysis is best conducted. This problem has two related parts in spatial analyses. First, the scale at which the analysis is conducted, and second, the way in which areas are sub-divided to produce appropriate zones for analysis. A consideration of both factors is an important pre-condition to the generation of safe knowledge at any level of aggregation. This paper uses the example of measuring socio-economic segregation between schools to illustrate results at different levels of aggregation from national to artificial 'competition spaces'. However, like the 'ecological fallacy', the issues involved here have a more general application to most substantive areas of social research. ABSTRACT FROM AUTHOR
Users
Please
log in to take part in the discussion (add own reviews or comments).