Big Data and Ethical Geospatial Research
As big data research methods are incredibly efficient and helpful in analyzing prevalent issues, and continue to grow and gain prominence. However, it is important to consider the ethics of using such data. As I wrote about previously in a post about Volunteered Geographic Information (VGI), privacy and transparency is often called into question when using these types of data. In a recent session of the Geospatial Fellows Webinar Series, Xun Shi presented on a bottom-up approach to epidemic modeling, which characterizes local situations and individual properties and behaviors, as well as their interactions. This method allows for a better and more detailed understanding of disease spread, as typical top-down models tend to generalize and oversimplify the situation.
Top-down approaches seek to establish general models then apply them to the entire problem, and assumes each local space or individual is a substantiation of the general models. This model is effective in conducting large studies, however, it assumes a controlled experiment, meaning it assumes we know the environment and conditions. This leads to ecological fallacy, as this is never the case in geographic modeling.
The bottom-up approach, as recommended by Professor Shi, does not seek to build a general model beforehand, rather, centers the model on the individual and their interactions. The expectation here is that general patterns or phenomena will emerge from this process. There are several reasons this process should be more widely adopted. This model runs lower risk of oversimplifying, thus lessens the uncertainty that may arise from a top-down model. Further, it is almost impossible to conduct controlled experiments when looking at epidemics, as demographics and interactions cannot be assumed. Thus, this model provides for more accurate, and therefore informative results. The bottom-up model represents real-world complexities than a top-down model cannot.
Increasing access to big data allows for more bottom-up approaches as argued by Xun Shi. There are certainly privacy concerns that arise with the use of this data, and the ethics of it must be strongly taken into consideration. Strong documentation, clear terms of data usage, and transparent methods can help to alleviate the stress that people may have when using big data. Zook et al. (2017) outline important steps to consider when using big data, as does DiBiase (2017). Xun Shi’s study pointed to the power of big data and his bottom-up approach when tracking an epidemic, which during the current COVID-19 crisis has been proven to be more important than ever. When done correctly and transparently, these methods can be extremely helpful in informing policies and can make large contributions to public health.
Sources:
DiBiase, D. (2017). Professional and Practical Ethics of GIS&T. The Geographic Information Science & Technology Body of Knowledge (2nd Quarter 2017 Edition), John P. Wilson (ed.). doi: 10.22224/gistbok/2017.2.2
Zook M, Barocas S, boyd d, Crawford K, Keller E, Gangadharan SP, et al. (2017) Ten simple rules for responsible big data research. PLoS Comput Biol 13(3): e1005399. https://doi.org/10.1371/journal.pcbi.1005399