AGGREGATION AND ECOLOGICAL EFFECTS IN GEOGRAPHICALLY BASED DATA

Statistics calculated using the means of geographic areas can differ s ubstantially from the corresponding statistics based on data from indi viduals. Analysts who base their conclusions about individual-level re lationships on area-level analyses run the risk of committing the ecol ogical fallacy. Statistical models are proposed that capture the essen tial features of the structure of a population composed of geographica lly defined groups and can encompass grouping processes and contextual effects. These models are used to show how small effects an the analy sis of individual-level data can be magnified substantially when the c orresponding analysis based on aggregated data is carried out. Thus th e source of aggregation effects is exposed. While aggregation effects have been studied by many authors, no general approach has been offere d to the problem of adjusting an area-level analysis so as to correct for aggregation effects and hence remove, or at least reduce, the bias that leads to the ecological fallacy. The statistical models proposed are used to provide an approach to this problem. Data from the 1991 U .K. Census of Housing and Population are used to illustrate the size o f the aggregation effects and the extent to which the proposed adjustm ents succeed tin their objective.