Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Qualitative Analysis of Common Practices in Annotations: A Taxonomy and Design Space

Md Dilshadur Rahman, Ghulam Jilani Quadri, Bhavana Doppalapudi, Danielle Albers Szafir, Paul Rosen

TL;DR

This work addresses the lack of a unified design space for chart annotations by conducting a qualitative study of over 1,800 annotated charts across 14 chart types. It develops an actionable design space organized around Why? How? What?, identifying seven annotation types and three data-source categories, and validating the space through three case studies. The study provides a taxonomy, usage patterns, and ensembles to guide practitioners in constructing clear, contextual visual annotations and to enable systematic critique of annotation practices. The work also offers a publicly available dataset of annotated charts to support future research and tool development, with implications for improving externalization, storytelling, and collaborative sensemaking in data visualization.

Abstract

Annotations play a vital role in highlighting critical aspects of visualizations, aiding in data externalization and exploration, collaborative sensemaking, and visual storytelling. However, despite their widespread use, we identified a lack of a design space for common practices for annotations. In this paper, we evaluated over 1,800 static annotated charts to understand how people annotate visualizations in practice. Through qualitative coding of these diverse real-world annotated charts, we explored three primary aspects of annotation usage patterns: analytic purposes for chart annotations (e.g., present, identify, summarize, or compare data features), mechanisms for chart annotations (e.g., types and combinations of annotations used, frequency of different annotation types across chart types, etc.), and the data source used to generate the annotations. We then synthesized our findings into a design space of annotations, highlighting key design choices for chart annotations. We presented three case studies illustrating our design space as a practical framework for chart annotations to enhance the communication of visualization insights. All supplemental materials are available at {https://shorturl.at/bAGM1}.

A Qualitative Analysis of Common Practices in Annotations: A Taxonomy and Design Space

TL;DR

This work addresses the lack of a unified design space for chart annotations by conducting a qualitative study of over 1,800 annotated charts across 14 chart types. It develops an actionable design space organized around Why? How? What?, identifying seven annotation types and three data-source categories, and validating the space through three case studies. The study provides a taxonomy, usage patterns, and ensembles to guide practitioners in constructing clear, contextual visual annotations and to enable systematic critique of annotation practices. The work also offers a publicly available dataset of annotated charts to support future research and tool development, with implications for improving externalization, storytelling, and collaborative sensemaking in data visualization.

Abstract

Annotations play a vital role in highlighting critical aspects of visualizations, aiding in data externalization and exploration, collaborative sensemaking, and visual storytelling. However, despite their widespread use, we identified a lack of a design space for common practices for annotations. In this paper, we evaluated over 1,800 static annotated charts to understand how people annotate visualizations in practice. Through qualitative coding of these diverse real-world annotated charts, we explored three primary aspects of annotation usage patterns: analytic purposes for chart annotations (e.g., present, identify, summarize, or compare data features), mechanisms for chart annotations (e.g., types and combinations of annotations used, frequency of different annotation types across chart types, etc.), and the data source used to generate the annotations. We then synthesized our findings into a design space of annotations, highlighting key design choices for chart annotations. We presented three case studies illustrating our design space as a practical framework for chart annotations to enhance the communication of visualization insights. All supplemental materials are available at {https://shorturl.at/bAGM1}.
Paper Structure (52 sections, 26 figures, 1 table)

This paper contains 52 sections, 26 figures, 1 table.

Table of Contents

  1. Introduction
  2. What is an Annotation?
  3. Design Spaces of Annotations
  4. The Roles of Annotations in Visualizations
  5. Comprehension, Memorability, and Recall
  6. Narrative Storytelling
  7. Collaborative Sensemaking
  8. Other Applications
  9. Real-World Annotation Practices
  10. Data Collected
  11. Data Coding
  12. Taxonomy of Annotation Types
  13. Enclosure
  14. Connector
  15. Text
  16. ...and 37 more sections

Figures (26)

  • Figure 1: A line chart from The Washington PostTan2022 illustrates COVID-19 peak comparisons, plotting time on the horizontal axis and percentage growth relative to the January 2021 peak vertically: (a) shows the baseline chart with basic visualization elements (i.e., axes, labels, lines, legends, and gridlines) but with annotations removed; (b) uses color+enclosure+text ensembles of annotations to help identify the peaks of different COVID-19 waves; (c) uses text+connector ensembles to present additional context from the associated article; and (d) displays the completely annotated chart.
  • Figure 2: We scrapped annotated visualization images from Google Images using the search pattern "annotated {chart type}", where the chart type included 14 commonly used charts (e.g., scatterplot). Two coders qualitatively coded the images into the seven annotation types shown.
  • Figure 3: An illustrative example of annotations in a scatterplot showing (a) enclosure of a group of data, (b) a connector between text and an indicator, (c) text describing the result, (d) glyph and color that highlight certain data points, two indicator annotations show (e) correlation and (f) average, and (g) a geometric annotation in the form of a zoom box.
  • Figure 4: Our design space of annotations is divided into three key sections. The design space is used by starting in the Why? section, which identifies a task and potential annotation types. Then, How? elaborates common usages of annotation in two parts: a color-coding system indicating the usage frequency of annotations (6-25%, 26-50%, and 51+%), and the types of annotation ensembles. Finally, What? is used to categorize the annotation data source.
  • Figure 5: A line chart from The Economisteconomist2023religion depicting factors influencing deaths of despair from 1979 to 2019 on the horizontal axis and the number of deaths on the vertical. (a) shows the base chart; (b) uses an indicator+text ensemble to help identify OxyContin's introduction to present its impact on the deaths of despair; (c) employs indicator+text ensembles to aid in summarizing data trends, and color+enclosure and connector+text ensembles to compare data trends; (d) presents the fully annotated visualization.
  • ...and 21 more figures