Report 2: Replication Study of Malcomb et al. (2014)
In this post, I report on the status of the second main assignment of GEOG 0323: a reanalysis of a reproduction of Malcomb et al. (2014). That paper is concerned with the creation of an index to assess vulnerability to climate change across Malawi. GEOG 0323 instructor Joseph Holler and colleagues have worked on a manuscript and workflow to reproduce the study, but it is as yet unpublished. The work of students in GEOG 0323 is to exact various changes which may improve the reproduction study.
Much of the work for this assignment involved building a discussion and conclusion section to assist with explaining the possible sources of error and uncertainty in the original study, particularly using some of the terms which Tate (2013) and Longley et al. (2008) use in their frameworks for uncertainty analysis. In the former paper, the terms include normalization and weighting (Tate 2013), which constitute important steps of the workflow for which the reproduction study presents some deviations. The latter paper is applicable in that the use of terminology in Malcomb et al. (2014) can be rather confusing to interpret, a concept which connects to the term “ambiguity” (Longley et al. 2008). In particular, the authors of the reproduction study include that the context for the terms “vulnerability” and “adaptive capacity” changed at points throughout the paper. An important aspect of the reproduction study is that Holler and colleagues compare Figure 4 and 5 from the original study with the figures created using their own workflow, which involved several deviations to maintain clarity and reduce uncertainty. I mainly chose to analyze these two comparison figures in the discussion section, for they suggest that the deviations had significant impacts on visualizations of adaptive capacity and vulnerability. For the comparison of Figure 5, I also changed the code to allow for better visibility of the results; I felt this important because I discussed this figure at length.
In the process of completing this reanalysis, I learned more about the structure of spatial analysis packages in R, including sf and stars. I also learned about the importance of internal consistency with respect to the definitions of terms. I hope that I have been able to contribute the start of a fruitful discussion on the comparison figures and that I have offered a measured perspective on whether the reproduction was successful in the conclusion, which was a point of discussion for the last report.
Here is a link to the full repository for the reproduction study.
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. https://doi.org/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. https://doi.org/10.1080/00045608.2012.700616.