SASI Home| Publications | Data | Maps | SASI Staff
Dorling, D. (1991) The Visualization of Spatial Structure, PhD Thesis, Department of Geography, University of Newcastle upon Tyne


This dissertation is about new ways of presenting and understanding some of the vast amounts of information which have been collected about our society. It is based on the premise that huge numbers of figures, incomprehensible in themselves, could contain a lot of information that is distorted, even lost, in conventional statistical analysis and cartographic display. This work has used the hopes and ideas of numerous sources as to how to visualize information about society more effectively. Recent dramatic increases in computer capability have allowed me to create images which would not have been conceivable two or three years ago. The approach has been experimental. I often had little idea what an image would look like on screen or paper until it was actually produced. The illustrations presented here are only a small selection of those created.

Many ideas and several themes are contained in this thesis. The text describes the rationale for, and development of, a new way of visualizing information in geographical research. Through the pictures the methods are illustrated; mistakes, techniques and discoveries shown. From the footnotes, which are largely quotations from a disparate literature, the origins of many of the ideas can be found. Time and again the suggestions of others to move in these directions are cited. Through technical asides some of the practical realities of the work are described. Through the illustrations and their legends, a picture of what has been happening to Britain in recent years unfolds. Many of the pictures justify an extended discussion, but I have aimed to keep the commentary brief. I have not included much detail about the computer software I have written and used because much of that is dependant on a novel (but inexpensive) hardware configuration and progress is so rapid that such knowledge is of only transient value. Periodically I have commented on the changing political and social geography of Britain that the mapping has revealed.

Numerous case studies are included. Questions concerning the implications of the spread of people in time and space are addressed at many points. The patterns through ten British general elections are depicted. The distribution of voting in ten thousand local ward contests is shown in a unique illustration. Some aspects of what the census can tell us about many people from their ages, sexes, occupations, activities and qualifications is revealed over a large set of very small areas. How people get to work, and the structure of the towns and cities in which they live, is examined. Migration is studied in several ways. The changing patterns of migration from birthplace are shown, and the streams of movement that cut across the country are drawn in unprecedented detail. House price change is visualized across several years and thousands of places. New techniques are developed to show the structure of local housing markets. Through other methods, the changes in this country’s industrial structure are seen as they have affected people in actual communities. The spatial and social manoeuvering of political allegiances is viewed from several angles, over the same period, and the relationships discussed. Finally, a smaller scale of analysis is considered, looking at what many images can tell us about the distribution of a disease, viewed from many different directions in space and time.

These social and political subjects are not arranged in their own, individual chapters, but run through the dissertation, which itself looks at methods of visualization, rather than the visualization of subjects. The work begins with a plea for a more human cartography to depict the events of our lives. The long history, but recent explosion, of envisioning information is then introduced. The rationale for this method of studying people, places and spaces is discussed. The form of artificial reality we require — area cartograms — are produced. The central part of the work looks at the honeycomb structure formed by the spatial patterns of society at single points in time; and how that alters through transforming the mosaic. The cobweb of flows which is responsible for most of the changes and stability is then drawn. This part of the dissertation is illustrated by many case studies. The last part attempts to show more complex aspects on the surface of social landscapes. Sculptured symbols allow us to see the relationships between the wood and the trees of social structure. Finally, a three-dimensional volume visualization of geographical and historical social structures is attempted. The thesis concludes by describing how all these methods and insights can together create another geography.

Human geography demands that we consider what is happening in many places at the same time. We do not need to study aspects of the world out of context. Here, an attempt is made to cover much ground and show numerous relationships. To do this it is necessary to be brief in detail, to be broad in scope, so the pictures often have to speak for themselves. Much of what has been shown here has never been successfully analysed by conventional means. However, this research does not come out of the blue, but accompanies thousands of papers written in the last three years describing the academic hope for a revolution in visualization, the history of which goes back to the first drawings. The message of visualization is that we should literally look at what is happening, drawing pictures in preference to writing words, listing numbers or designing theoretical models. It has only recently become possible to do this in such quantity and quality of detail. The prints resulting here are the tools of enquiry, not simple pictures for embellishment, but the foundation of the thesis. This is a story of invention and discovery.