Seeing What Matters: UChicago’s Alex Kale Receives NSF Early CAREER Award for Rethinking Data Visualization Ethics
When you glance at a chart in the news, an interactive dashboard in a healthcare app, or a scientific report, you trust that what you’re seeing tells the truth–or at least most of it. But the reality is more complex: every visualization is a product of choices, revealing some data while obscuring or distorting other parts. Assistant Professor Alex Kale from the University of Chicago’s Department of Computer Science and Data Science Institute has received a prestigious National Science Foundation Early CAREER Award to explore this very tension.
The CAREER Award is the NSF’s highest honor for early-career faculty, supporting those who show potential to lead transformative research and education. For the department, it is a mark of distinction, underscoring a commitment to advance foundational progress and societal benefit.
The project, “Seeing What Matters: Reframing Visualization as Data Disclosure,” reimagines how we approach visualization–not just as a tool, but as an object of computation. By framing visualizations as mechanisms for responsible data disclosure, the research creates new pathways to recommend designs and guide audiences in interpreting charts more effectively. Kale’s research asks: What does a visualization truly show, and just as importantly, what does it leave out?
Why This Matters: From the Classroom to Public Policy
Visualizations play a pivotal role in shaping public understanding, from medical outcomes to climate science and city budgets. Beyond privacy, this work recognizes the importance of disclosure in a wide array of scenarios: progressive data sharing, varying security clearance levels, and crafting simple summaries that suppress superfluous information. Honest, effective design requires carefully balancing what to disclose with what must be hidden or simplified.
In today’s world of information overload, a single chart can hide signals critical for audiences. Scientific charts often summarize complex results through statistical intervals, but if key variables are omitted or misunderstood, audiences may come away with an inaccurate picture. This project aims to help both visualization creators and users better recognize these pitfalls.
“Many common pitfalls in data interpretation—such as base rate neglect, miscomprehension of confidence intervals, and the modifiable areal unit problem—stem from not recognizing what a visualization cannot show by design”, said Kale. “The impetus for the project is to identify a systematic way to think about these failures of expressiveness in visualization, both in terms of the technical details of how they arise and in terms of how to address the cognitive and ethical challenges they pose for communication.”
Kale’s project tackles both the challenges faced by both visualization creators and audiences: how to design responsibly, and how to interpret visualizations that provide only a partial view on the information they care about. For those who make visualizations, he’s developing new tools grounded in mathematical formalism and software to help designers weigh the ethical trade-offs of data disclosure. These tools will offer practical recommendations for honest design, and interfaces that explain to viewers what a chart can’t show and what it may not be able to depict accurately.
For students and general audiences, the project will deliver educational materials and games that build skills in defensive interpretation, equipping people to ask, ‘What am I not seeing here?’ These resources will feature in college classrooms, outreach events, and workshops for city colleges and local non-profits.
To make these ideas tangible, Kale and his team created a hands-on software platform for data communication games deployed in UChicago classrooms to immerse students in the ethical challenges of visualization design. Unlike traditional lessons that focus only on avoiding deception, these interactive games place students in realistic, morally complex scenarios, encouraging them to practice responsible design and develop their own judgement. The game is a fresh, hands-on approach that Kale hopes to share more widely in the future.
Building Trust Through Design: Ethics, Privacy, and Communication
Kale’s framework reframes visualization as a mechanism for “responsible data disclosure,” rather than pure data presentation. That means developing software and a domain-specific language (DSL) that lets visualization authors specify disclosure goals, generate visual alternatives, and score designs on how much information is lost or distorted.
Consider the privacy challenges faced by healthcare researchers, or the need for city officials to communicate policy impacts without exposing sensitive information. The tools and theories from this project offer new support, optimizing visual designs for both transparency and protection.
Through controlled experiments and interviews, Kale’s team will study how non-specialists interpret charts with hidden signals and assess whether new explanation tools enhance their understanding. Students will participate in data communication games that simulate real-world scenarios and negotiate which signals to disclose in their visualizations, including the ethics of those choices.
”Visualizing data can present ethical dilemmas where a designer’s responsibilities to others are incompatible with full transparency: the imperative to protect the rights of data subjects, to keep intellectual property secure, or to provide a simplified data summary,” explained Kale. “The project contributes tools, frameworks, and educational materials to help visualization designers and students become better informed moral actors, and to guide data interpretation and the formation of ethical standards for data communication in such sensitive settings.”
This award not only recognizes Kale’s deep commitment to ethical standards in data science, but also reaffirms the University of Chicago’s Department of Computer Science’s growing influence in the field. By advancing fundamental theory, building usable tools, and driving education reforms, the department is helping shape how society understands data.
