The Modifiable Areal Unit Problem

The modifiable areal unit problem (MAUP) is an important phenomenon associated with the use of data aggregated to geographical areas. It affects the aggregation of events, individuals or households and therefore applies to many variables of interest in the social sciences, including, for example, unemployment and ethnicity rates. In these cases, the geographical boundaries are 'imposed' in the sense that they do not relate in any meaningful sense to the variables of interest. Typically, data are aggregated from source observations to areas such as census wards or output areas which have no special meaning in terms of the underlying geographical distributions such as unemployment or ethnic composition.

MAUP refers to the fact that the observed aggregated values will vary according to how we draw our area boundaries. MAUP comprises both scale and aggregation effects. The scale effect relates to the size of the areal units that we use and the aggregation effect relates to the exact way in which they are assembled at a given scale. Changes in either will bring about changes in the apparent geographical distribution of the variable in question. This can be illustrated with the following examples:

Figure 1: Total population in 16 areas

Figure 2: Percentage unemployment

Figure 3: Percentage unemployment aggregated to four areas

Figure 4: Percentage unemployment aggregated to four areas

Figure 1 shows the population count in 16 areas and Figure 2 shows the corresponding percentage unemployment. By moving to a different scale in Figure 3 the apparent range of unemployment values is much reduced (all percentages rounded to nearest whole number), with four irregularly-shaped areas suggesting that there is little variation across the region. However, a different aggregation at the same scale in Figure 4 demonstrates a much greater range of unemployment percentages and suggests some distinct geographical concentration. Many different impressions of the underlying pattern can be created by use of alternative scales and aggregations, particularly when such data are presented as shaded area (choropleth) maps. This is the phenomenon known as the modifiable areal unit problem and it should be borne in mind by all users of areally aggregated data. MAUP is closely related to the ecological fallacy.

Deliberate manipulation of political boundaries in this way so as to improve the electoral chances of a particular party is known as "gerrymandering" and is another instance of the MAUP in practice. It is for this reason that political boundaries in the UK are continually reviewed by an independent Boundary Commission.