Bringing interactive visual analytics to the classroom for developing EDA skills
Jessica Zeitz Self, Nathan Self, Leanna House, Jane Robertson Evia, Scotland C. Leman, Chris L. North · Journal of computing sciences in colleges · 2018
This paper addresses the use of visual analytics in education for teaching Exploratory Data Analysis (EDA) skills. EDA is inherently a creative, knowledge discovery process that often takes place before formal technical statistical analyses. A challenge in teaching EDA is that there is often no right nor wrong way to conduct EDA, yet, given a dataset, some EDA can be more comprehensive or insightful than others, based on the kinds of insights made. Also, in the face of high-dimensional data, students are often limited by how they relate to the data and their technical skills for EDA. How can students make complex insights from high-dimensional data, if they do not have the technical skills to explore the data from multiple, highdimensional perspectives? In this paper, we use our own tool called Andromeda that enables human-computer interaction with a common, easy to interpret visualization method called Weighted Multidimensional Scaling (WMDS) to promote the idea of making complex insights. We present Andromeda and report findings from a series of classroom assignments to 18 graduate students. These assignments progress from spreadsheet manipulations to statistical software such as R and finally to the use of Andromeda. In parallel with the assignments, we saw students' cognitive dimensionality (CD) begin low and improve.