By Megan Heyman (Rose-Hulman Institute of Technology); Annie Liner (Indiana State University)
Spatio-temporal data analysis is an approach for modeling dependencies associated with location and time simultaneously. These types of dependencies naturally appear in a plethora of real applications, but implementing models that properly account for the dependencies necessitates strong mathematical, statistical, and programming knowledge. The prerequisite knowledge associated with these spatio-temporal techniques creates a barrier to many undergraduates. With the abundance of spatio-temporal data, it is still important for undergraduate students to interact with such data and cultivate conceptual understanding of spatio-temporal dependencies. Conceptual understanding then provides a basis to approach spatio-temporal techniques later.
In this presentation, we describe how to naturally weave ideas of identifying and describing spatio-temporal dependencies into the first course in statistics. These recommendations map to existing discussions of the investigative process, multivariable thinking, and assessment of model conditions. Participants will leave with resources to also begin incorporating such examples to their existing introductory statistics course.