Available at: https://digitalcommons.calpoly.edu/theses/3066
Date of Award
6-2025
Degree Name
MS in Statistics
Department/Program
Statistics
College
College of Science and Mathematics
Advisor
Emily Robinson
Advisor Department
Statistics
Advisor College
College of Science and Mathematics
Abstract
Exploratory data analysis (EDA) is a method for uncovering the structure and key characteristics of data, often through the use of statistical graphics. These visual tools can reveal patterns and trends, and their effectiveness can be enhanced through interactivity. By enabling users to filter data, zoom, and toggle visual features, interactive plots can accelerate and enrich the EDA process. This study extends a previous graphical study by incorporating an interactive framework. Using a statistical lineup protocol with two target patterns (a linear trend and a clustering trend) participants interacted with plots by toggling various aesthetic features, including cluster coloring, ellipses around clusters, linear trendlines, and regression error bands. Data was collected via an RShiny application, capturing participant demographics, target choices, toggle usage, and reasoning for selections. To assess which graphical features enhanced the detectability of the linear or cluster targets, a generalized linear mixed model was applied. I also analyzed toggle behavior, exploring commonly used feature combinations, timing of toggling actions, user preferences, and individual workflows. Finally, I examined participants’ explanations and strategies to gain insight into how they interacted with and interpreted the visualizations. This study contributes to the understanding of how users engage with interactive graphical tools and how such tools support data interpretation in EDA.