Document Type


Degree Name

Master of Arts (MA)


Geography & Environmental Studies


Faculty of Arts

First Advisor

Not applicable

Advisor Role

Not applicable


The main objective of this study is to develop a computer model to predict the food stores which individual consumers choose for the major part of their grocery purchases. It is worthwhile to develop such a predictive model for two reasons. In the first place, it is useful to be able to predict consumer store choices for purposes of retail planning. Secondly, a working predictive model illuminates the main bases of the consumer spatial choice process and thus adds to our understanding of urban spatial organization.

A basic idea of model building is that a model is a simplification of reality in which a small number of variables may serve with great economy of data to capture and make clear the essence of the processes involved. As Haggett suggests, “Successful models are those which manage a considerable amount of simplification without introducing extraneous noise.” If it is possible to design a model which predicts consumer store choice on the basis of a few simple variables, the resulting model is of greater value than a model in which so many variables are introduced that the results approaches a duplication of reality rather than an enlightening simplification and abstraction.

Convocation Year