A Qualitative Evaluation and Taxonomy of Student Annotations on Bar Charts

Abstract

When sharing visualizations, annotations provide valuable insights into the data by focusing attention on important visual elements. As a result, annotations have become an essential part of visualizations, primarily when externalizing data or engaging in collaborative analysis. Therefore, it is crucial to understand how people annotate visualizations. This two-phase study used individual and group settings to investigate how visualization students annotate bar charts when asked to answer high-level questions about the data in the charts. The resulting annotations were coded and summarized into a taxonomy with several interesting findings. For example, we notice that several annotation types were broadly applied, while others were just used in special cases. In addition, ensembles of annotations were required to sufficiently annotate specific tasks. Our findings provide an early framework for contextualizing the usage of annotations, further providing guidance for best practice uses.

Exploring Annotation Strategies in Professional Visualizations: Insights from Prominent US News Portals

“…In our previous research, where we developed a taxonomy and design framework that encompasses common annotation practices in various chart types [28], we defined annotations as "supplementary elements, such as graphical shapes, text, or color, added to a visualization to provide additional context beyond the basic data presented, and to draw attention to specific portions or elements of the data, thereby enhancing the understandability and expressiveness of the visualization". We began by conducting a focused pilot study on grouped bar charts, which allowed us to establish a taxonomy of annotations.…”

“…Other studies have explored annotations in visualization design, including contextual linking of text and interactive graphics [24], expressing chart construction errors to a lay audience [13], and optimizing bar chart layout using annotations on trends in bar heights and text values [31]. Moreover, Rahman et al conducted a study on visualization students' annotation practices with grouped bar charts, where they identified five main annotation types -enclosure, connector, text, mark, and color used in response to high-level chart-related questions [29,30].…”

“…To accomplish this, we assembled a corpus of professional charts from four prominent US news portals, encompassing diverse chart types. These charts were systematically coded using the design space framework proposed by Rahman et al [28]. By systematically analyzing the corpus of professional charts, we aim to uncover common patterns, gaining valuable insights that inform the development of effective annotation practices for creating impactful and informative visualizations.…”