Uncertainty in GIScience

My senior year of high school, I was interested in looking at indie and chain bookstores in Washington DC over a period of 20 years, comparing where they were opening and closing. At that point, I had very minimal experience with GIS; the project in some ways was a way for me to familiarize myself with ArcGIS and learn some of its capabilities. One challenge I faced was with data collection. It was quite difficult to create a dataset with all of the bookstores in DC. I used Yellow Pages to make a list of all the bookstores in DC open at certain periods of time. Of course, I had to make decisions regarding which stores I would choose to include. For example, I was mainly interested in a “typical bookstore,” selling fiction, nonfiction, etc. but several of the listings were religious bookstores, gift stores that also sold books, or museum bookstores. Ultimately, I decided to include these, but even still, I’m not sure if they were what I was really interested in for the sake of my study.

I think that geographers are responsible for reporting the degrees to which their research is subject to uncertainty. Looking at Longley’s diagram for conceptualizing uncertainty (6.1), I think that geographers should be open about the ways that they are conceptualizing their data and their variables (Longley et al 2008). I imagine that it could be fairly easy for geographers to present their research as much more certain or robust than it really is, which would be misleading. Looking at Malcomb’s study, I was skeptical of his conceptual framework, which was “built on expert-derived and observed linkages of vulnerability’’ (Malcomb et al, 2014). I felt that Malcomb was vague about the way that experts where coming up with “metathemes” and other indicators/contributors to vulnerability. With this in mind, I was curious about the implications of using interviews to drive research decisions, especially in a study claiming to be reproducible. I think it would have been beneficial for readers to see interview transcripts so that they could better understand decisions that Malcomb’s experts were making in order to come up with indicators of vulnerability. After all, the decisions of the experts would be subjective, and I think it would be worthwhile for reproducers of the study to be able to see what went into their decisions, since these indicators shaped the methodology later in the study.

There are a couple strategies that geographers may use to fulfill their responsibilities regarding uncertainty. Geographers can use statistics to show measures of uncertainty in their research, as laid out by Longley. Simulation is a way in which uncertainty may be mitigated or accounted for in a study as well. Additionally, geographers should report where they are getting their data, and they should consider reporting important or useful metadata as well (Longley et al, 2014).

References Longley, P. A., M. F. Goodchild, D. J. Maguire, and D. W. Rhind. 2008. Geographical information systems and science 2nd ed. Chichester: Wiley. Chapter 6: Uncertainty, (pages 127-153)

Malcomb, D. W., E. A. Weaver, and A. R. Krakowka. 2014. Vulnerability modeling for sub-Saharan Africa: An operationalized approach in Malawi. Applied Geography 48:17–30. DOI:10.1016/j.apgeog.2014.01.004

Tate, E. 2013. Uncertainty Analysis for a Social Vulnerability Index. Annals of the Association of American Geographers 103 (3):526–543. DOI:10.1080/00045608.2012.700616.

Tullis, J. A., and B. Kar. 2021. Where Is the Provenance? Ethical Replicability and Reproducibility in GIScience and Its Critical Applications. Annals of the American Association of Geographers 111 (5):1318–1328. DOI:10.1080/24694452.2020.1806029