How Data Narratives Go Wrong: A Taxonomy of Issues Across the Data Communication Process

Jul 16, 2026

Upload by Levi

Data narratives increasingly shape public understanding, yet their failures often stem from more than isolated factual mistakes or deceptive charts. Failures arise within a broader meaning-making process where quantitative evidence is transformed into claims, representations, and arguments. The work introduces TIC, a taxonomy synthesized from prior research and refined through qualitative annotation of 700 real-world data narratives. It organizes recurrent breakdowns across six dimensions and situates them within end-to-end communication from analysis and narrative construction to audience reception, enabling diagnosis and guidance for trustworthy communication.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages20

Reading Time50 min

Introduction

Data-driven narratives as meaning-making artifacts

Complexity of narrative authoring and validation

Generative AI and additional trust challenges

How Data Narratives Go Wrong: A Taxonomy of Issues Across the Data Communication Process Yu Fu , Jiawei Zhou , Sichen Jin , Munmun De Choudhury , Cindy Xiong Bearfield , and John Stasko arXiv :2607 . 10523v1 [ cs .HC] 12 Jul 2026 Abstract—Data narratives increasingly shape public understanding, but their failures are rarely just isolated factual errors or deceptive charts. Instead, they emerge through a broader meaning-making process in which quantitative evidence is transformed into claims, representations, and arguments. While prior work has examined these failures across disparate fields (e.g., statistics, visualization, and fact-checking), the community lacksa holistic lens to explain how these issues arise, propagate, and compound. To address this gap, we introduce TIC, a Taxonomy of Issues in Data Communication, synthesized from prior literature and refined through the qualitative annotation of 700 real-world data narratives from fact-checking sites, research datasets, and controversial media. TIC organizes recurring breakdowns across six dimensions—data, analysis, visual encoding, text, reasoning, and interpretation—and situates them within a framework spanning analysis, narrative construction, and audience reception. Alongside the taxonomy and process framework, we contribute a qualitatively annotated case corpus with coding justificationsand an interactive browsing interface. Collectively, these contributions provide a structured lens for diagnosing problematic data narratives and informing future sociotechnical support for trustworthy data communication. Index Terms—Data-driven communication, narrative visualization, and misleading visualization. I. INTRODUCTION DATA communication has become one of the primary lenses through which the public understands complex social, scientific, and political phenomena [1] . From tracking pandemic trends to interpreting climate risks and election results, quantitative evidence is routinely leveraged not only to inform but also to persuade [2] . However, the aura of objectivity surrounding “the numbers” [3] can obscure a fragile construction process in which narratives can break down not only through overt fabrication, but also through mechanisms such as misused statistics, cherry-picking, and systematic sampling biases. This critique is not new: classic statistical skepticism—often summarized by the aphorism “lies, damned lies, and statistics”—has long illustrated how valid numbers can be framed to support flawed conclusions [4] . More recently, influential critiques of data absence have highlighted how “data silences” can quietly shape public beliefs about risk and safety while maintaining an appearance of mathematical objectivity [5], [6], [7] . Collectively, these examples underscore a critical reality: choices in how data are collected, analyzed, represented, and framed propagate through narratives to materially shape public understanding at scale. Yu Fu is with the University of Central Florida, Orlando, FL, USA. E-mail: [yu.fu@ucf.edu](yu.fu@ucf.edu) Jiawei Zhou, Sichen Jin, Munmun De Choudhury, Cindy Xiong Bearfield, and John Stasko are with Georgia Tech, Atlanta, GA, USA. At the core of data communication lies the data-driven narrative, a communicative artifact that integrates quantitative evidence with interpretation through text and visual forms [8] . Such narratives do more than present numbers; they organize evidence into interpretive accounts by situating data within context, foregrounding particular patterns, and guiding how audiences understand what is plausible, credible, or actionable. In practice, data-driven narratives encompass many forms, from textual claims that summarize key values and metrics [9],[10], to visual artifacts such as tables and charts [11], to richly layered stories that integrate interactive visualizations with explanatory prose [8], [12] . Across this spectrum, text and visualization work in tandem [13]: visualizations sur

FAQs

数据叙事（data narratives）会如何影响公众理解？

数据证据在文本与可视化中被组织成解释性叙述，从而引导受众理解何者可信、可行以及风险或结论的合理性。

TIC（Taxonomy of Issues in Data Communication）主要解决什么问题？

先前研究虽关注不同领域的失败案例，但缺少统一视角解释问题如何产生、传播并相互叠加。TIC提供跨过程的结构化诊断框架。

TIC将数据叙事的常见问题划分为哪些维度？

TIC将反复出现的失效或断裂组织为六个维度：data、analysis、visual encoding、text、reasoning与interpretation。

Jul 16, 2026

Upload by Levi