Reflections on Data Interpretation and Resolution

Blog Post created by jkerski-esristaff Employee on Aug 17, 2012
One of the recurring themes in GIS education blogs, forums, listservs, conferences, and in our recent book The GIS Guide to Public Domain Data is the importance of paying attention to the characteristics of data so that correct interpretations of that data can be made when mapping that data. During some recent work with one of our favorite resources mentioned in the book, that of the Center for International Earth Science Information Network, (CIESIN), I examined a map showing the population density for central Europe (shown in image).ciesin_population_density_map-285x300.jpg
As I examined the map, population density for Germany seemed unnaturally higher than that of France. True, according to a variety of outside sources, Germany does have roughly twice the population density as does France (235 people per square kilometer for Germany versus 108 for France). But the map seemed to indicate that broad areas of Germany are even higher in density, approaching that of The Netherlands. Is this accurate?

Upon further investigation into the metadata, indeed, the resolution of the data set was to blame. CIESIN uses the highest resolution data available for generating maps like this, and the data they were able to obtain for France had a much higher resolution. The resolution is calculated as the square root of the land area divided by the number of administrative units. For Germany, it was 28, and for France, 4. The resulting population per administrative unit in 2000 was 184,000 for Germany versus only 2,000 for France. A-ha!

Close examination of the map prompted my initial concern: The areas mapped in Germany are much larger than in France. But the differences in data collection could not be confirmed until the metadata was examined. Fortunately, CIESIN does an extremely good job documenting their sources and methods. The user still needs to make it a point to read that documentation. But what should we do when working with sites that do not document their data well? In today’s world of a myriad of data, maps, and tools, it is more important than ever to have a good grounding in map interpretation and spatial analysis, but also to ask questions of the data you are using. How can we as the GIS education community foster this kind of questioning of data by our students?

- Joseph Kerski, Esri Education Manager