Geographical Information Systems require both spatial (e.g. maps) and non-spatial data or attribute data (e.g. statistical information such as population figures, financial data, etc.). Maps can be entered or digitised into a computer either by scanning the maps or digitising them. On the other hand, non-spatial data has to be entered into tables or databases. Unfortunately, many GIS systems are not particularly user-friendly and they often do not allow the user to enter or manipulate attribute data easily. Since attribute data is a key ingredient for creating thematic maps the use of spreadsheets for entering data has grown in popularity.
There are two possible ways of entering attribute data:
- Relational databases have to be developed using professional database software packages (e.g. dBase, Access, Oracle). The advantage of using relational databases includes the ability to develop database models that combine or link data that has been entered in endless combination possibilities. The complexity and level sophistication needed to successfully develop and maintain useful databases is a distinct disadvantage of relational databases. This often necessitates qualified professionals who are capable of creating, modifying and maintaining the complex databases.
- Most GIS users do not require such sophisticated systems based upon relational databases, entering their data into spreadsheets usually perfectly satisfactory. Programmes such as MS Excel offer the possibility of entering data in spreadsheets, which can be easily created. The programme also allows calculations to be automatically undertaken (i.e. population projections). The data can then be easily exported or embedded as part of the GIS database. The use of common spreadsheet programmes and the ability for the user to familiarize themselves quickly with the programmes capability of entering and manipulating data are a distinct advantage. Most user-friendly GIS systems use Excel as the main attribute data interface. These GIS packages are ideal for relatively small sets of data. Where large volumes of data have to be applied the GIS user will have to use relational database programmes but this is more likely to be the exception and not the rule.