Hannah Rigdon

home  |  GIS  |  musings  |  cartography  
       |  about

Is GIS a Science?

In my experience with GIS thus far, I have interacted with it as a tool or means of analysis. Especially in classes like GEOG0120 or Remote Sensing, GIS has been used as a tool with which questions can be posed or answered, but I have not spent much time reflecting on the “intellectual foundation” of GIS. Any critical thinking I’ve done about GIS has always been in the context of another topic, like land use development, or environmental change. However, I think it is important that GIS be considered and treated as a science, given that it is not an unbiased tool for conducting inquiry and analysis. GIS is used to create models or abstractions of reality, which should not be equated with unbiased objective truth. To treat GIS as a tool, or a means to an end is to overlook the complexity and abstraction of GIS analysis, which leads to a dangerous oversimplification of results.

Open Source GIS can help move GIS from a tool to a science, in that Open Source GIS is constantly pushing GIS tools to develop and evolve. The core tenants of open source, the public availability of code and data, are important to opening analysis and research up to further development and critique, which will further legitimize GIS as a research field. Increasing transparency in geospatial analyses through advocacy for reproduction and replication could prompt more introspection within the field of GIS itself and prompt further development and growth. Open source offers the potential to open up GIS to more folks and democratize knowledge production and development within the field. Additionally, further inclusion and emphasis on reproduction and reproducibility will contribute to more rigorous science and analysis being produced.

References

Wright, D. J., M. F. Goodchild, and J. D. Proctor. 1997. GIS: Tool or science? Demystifying the persistent ambiguity of GIS as “tool” versus “science.” Annals of the Association of American Geographers 87 (2):346–362. DOI: 10.1111/0004-5608.872057