Graduate students from a variety of agricultural and natural resource fields are incorporating geographic information systems (GIS) analysis into their graduate research, creating a need for teaching methodologies that help students understand advanced GIS topics for use in their own research. Graduate-level GIS exercises help students understand advanced concepts that they can later apply to their research. We have developed a laboratory exercise based on a published journal article, "Predicting Rainfall Erosivity in Honduras," that introduces the concept of applying regression analysis to raster GIS. The objective of this laboratory exercise was to have students learn how standard statistical methods, in this case regression analysis, can be applied to spatial datasets to produce a GIS-based map of predicted results. By using a journal article as the framework for this exercise, students were able to both understand the analysis method and also compare their results to the published map. Students stated that this lab helped them to understand how to apply regression analysis techniques using raster analysis tools in a GIS environment. (Contains 3 figures and 3 tables.)