A place in history: a guide to using GIS in historical research

CHAPTER 1: GIS AND ITS USES IN HISTORICAL RESEARCH

1.6 Towards good practice in GIS

Throughout this book the issues, dangers and pitfalls of using GIS will be raised and references given as to where more detail can be found. As the use of GIS among historians is in its early stages, and the potential applications of GIS are so broad, it is difficult to be precise about what constitutes good practice. The basics of good practice when using GIS can be summed up in three broad points:

1. Always think carefully about the impact of location. The basic question underlying GIS research is 'what impact is the geography having on my data?'. This applies to both genuine geographical issues such as the links between industrial development and the development of the transport network, or problems with the spatial nature of the data. These can include apparently simple considerations such as the impact of combining two layers taken from different scale sources, as well as much more conceptually complicated considerations such as the impact of spatial autocorrelation on statistical techniques.

2. Always be aware of the data's limitations. This applies in particular to the limitations of the source material, and to limitations connected with spatial issues such as scale and accuracy.

3. Avoid unnecessary simplifications when exploring, visualising or analysing data. This means trying to use the data in a form as close as possible to their original with a minimum of aggregation though space, time or attribute. Where aggregation is unavoidable or is present in the source, the impact of the aggregation on the patterns formed must be considered.

These three points are similar to good practice in historical research generally. The key difference is that with GIS the historian has to consider good practice in relation to spatial data as well as to attribute data, and issues associated with space, location and geography present particular challenges to good practice that the historian is unlikely to be familiar with. This book will raise these issues in more detail and give guidelines as to good practice in dealing with them.

GIS, in combination with other branches of scholarship, has the potential to provide a more integrated understanding of history. There are, however, risks. In particular, GIS data are expensive, GIS expertise is expensive, and the learning curve for people using GIS is steep. This means that researchers wanting to get involved in GIS should do so carefully, and be aware that the initial investments are high and that the rewards may take some time in arriving. On a more positive note, once spatially referenced databases are built they can be significant works of scholarship in their own right and can enhance our understanding of a problem. This means that "what might initially appear to be simply an exercise in automated cartography proves able to supply detailed answers to questions so detailed and time-consuming to obtain by manual methods that scholars have previously tended to avoid asking them" (Healey and Stamp 2000, 584).