This of course is a rhetorical question since geography is a critical component of all analyses which subsequently affects our interpretation and understanding of the data (or phenomena) being analyzed. Unfortunately, there is no single natural scale at which phenomena should be studied (i.e., the fundamental problem), so, a possible solution to address this issue is through characteristic scales. Depending on the purpose of the study and intended analysis, and perhaps the mandate, an appropriate (or more suitable) scale of analysis will exist. However, to truly develop a full understanding of a place in space, multi-scale analyses are often required.
Overall, the lecture served as an important reminder that scale matters; where results are dependent on scale and observed relations can change according to scale. So, prior to conducting any analyses, forming a thorough understanding of the underlying geography is critical in ensuring a meaningful research design process.
Keywords: Scale, Grain, Extent, Modifiable Areal Unit Problem, Gerrymandering, Kriging, Spatial Autocorrelation, and Simpson’s Paradox.