A Framework for Better Visual Communication Practices
Problem
Researchers communicate findings through data visualizations, but have little tested guidance on what the public understands.
Researchers routinely present data and findings in graphics, charts, and tables, yet little is known about how well a broad audience reads and interprets these formats. How often people encounter data graphics, how comprehension differs across groups, and how design choices shape what readers take away all remain largely untested with a large, representative sample
Without that evidence, researchers design visualizations by convention and instinct, with no shared standard for what works. Closing that gap is a foundational step toward communicating research in ways the public can use, and toward a common set of tested practices researchers can rely on.
Solution
We are building a framework of tested design practices based on a national survey of how U.S. adults interpret charts and graphs.
In collaboration with researchers at the University of Nebraska-Lincoln, NORC is developing and fielding a nationally representative survey to test viewer understanding of eight common formats of visual data communication. Using NORC’s AmeriSpeak® panel, the team is surveying 2,000 respondents on their familiarity and understanding of the content in sample charts and tables. The visuals shown to survey participants will use several design structures, such as different scales, labeling choices, and supporting language, to gauge how these variables impact understanding.
The study also oversamples groups of U.S. adults who are typically understudied in this area, including adults with lower levels of educational attainment and those living in rural areas. This range lets the survey carefully document how comprehension varies across audiences rather than for a general population alone.
We will then build the framework itself, translating what the survey reveals about how people read data into an evidence-based set of design recommendations.
Result
Researchers will have tested guidance for presenting findings in ways their audiences can more accurately understand.
Researchers, journalists, government analysts, and others who present data to the public will be able to ground their design choices in evidence about how people consume data visualizations, rather than on assumption and precedent. Because the survey data measure how comprehension varies from one audience to the next, the guidance will speak to which designs hold up for which readers. Researchers will also gain access to the survey data through a public use file, allowing for the work to be replicated and built on, and the team will share findings through publications and presentations to the research community.
The wider benefit is a public better equipped to engage with research on its own terms. Visualizations that are read as intended give people a sounder basis for decisions about critical areas such as health, spending, and policy, and leave less room for misinterpretation or misuse.
Related Tags
Project Leads
-
Kiegan Rice
Senior StatisticianPrincipal Investigator
