As policymakers, news media, and research communities increasingly rely on big data, the ability to create good visual representations has become key to conveying complicated ideas to a more general audience. But human rights organizations that regularly use empirical analyses in their research have nevertheless been slow to use data visualizations. Professor of Clinical Law Margaret Satterthwaite  ’99 and NYU Polytechnic School of Engineering’s Enrico Bertini and Oded Nov received a grant in June to further their research exploring how advocacy organizations can effectively employ information graphics to tell human rights stories.
Satterthwaite, Bertini, and Nov have already completed two initial user-based studies that investigate how readers respond to visual presentations of data. One study verifies that data visualization is a more effective tool than text in conveying statistics to the reader. In another study focusing on deceptive visualization, Satterthwaite and her collaborators show how it is possible to deceive readers by using correct data but changing the expected visualization. Inverting the axis on a line graph, for example, can lead a reader to believe that an increasing trend is, in fact, decreasing.
“It was quite disturbing how easy and how intense was the effect of deceptive visualization,” Satterthwaite says. Understanding how readers comprehend and react to information graphics is key to helping researchers avoid accidentally overstating or understating their findings. Satterthwaite also notes that it is important to recognize the dangers of deceptive data visualization, which, out of the zeal to convince, could be used to mislead the audience.
Now, with the grant from the John D. and Catherine T. MacArthur Foundation, Satterthwaite says the next phase of research will be to work with various human rights organizations to implement these findings. “The hope,” she says, “is for it not to be just an academic study but a collaboration with real-world impact on how human rights organizations employ data visualization in ongoing research and advocacy.”
Satterthwaite’s work on data visualization is part of a series of interdisciplinary collaborations in which she hopes to encourage innovation in the ways that human rights workers document and demonstrate violations. In a chapter she is contributing to The Transformation of Human Rights Fact-Finding  (co-edited by Philip Alston , 2015), Satterthwaite joins Princeton researcher Justin Simeone in looking at whether, as researchers incorporate quantitative methods into human rights research, they can or should follow the same disciplinary standards that guide social science researchers.
Video: Using quantitative data in human rights research
Traditionally, human rights advocacy has been based on testimonial evidence, a methodology that grew from practices of law and journalism. People respond emotionally to stories, and the goal of human rights research, after all, is to persuade policymakers and the public to take action to prevent or stop violations. “One of the great strengths of the human rights movement is our ability to tell the story of the victim, of the survivor, and compel people to act,” says Satterthwaite, who co-edited the 2008 book Human Rights Advocacy Stories . “That is an ethical duty that we have, and we should not abandon it.”
The incorporation of social science methods into human rights research has the potential to amplify the power of storytelling. “Somebody might be really compelled by one story of a person getting killed,” Satterthwaite says. “But if you tell them there are 10,000 of those people being killed, their empathy suddenly shuts down.” Demonstrating the right to equal access to drinking water, however, is best shown visually through statistics such as the proportion of a population that lives within a kilometer of safe water and the prevalence of diseases caused by unsafe water.
Satterthwaite hopes the trio’s work will guide researchers to make the best choices in employing the methods at their disposal. “It may simply be a question of what kind of evidence you need,” she says, “depending on your audience and your purpose.”
2015 Home