Faculty Spotlight

Quincy Thomas Stewart

Dreaming of data and demographic models to study inequities

Quincy Thomas Stewart

While demographic data might not be the stuff of most people’s dreams, for IPR fellow Quincy Thomas Stewart, it is—and applying advanced mathematics to social science issues is what led him to becoming a social demographer.

“I was always interested in understanding racial inequality–how it emerged, how it was maintained, and how it was sustained,” he said. So it seemed natural for him to pursue a degree that would allow him to “sit around and solve cool math problems all the time about things that […] really mattered.”

“I saw these methods [when looking at graduate programs], and I thought, ‘This is the best thing ever, and I need to do this for the rest of my life,’ ” Stewart said, explaining his decision to pursue his PhD in demography and sociology at the University of Pennsylvania, which he completed in 2001.

Agent-Based Modeling and Racial Disparities

To examine the dynamic processes underlying discrimination and inequities, Stewart has had to come up with new ways to quantify old problems. He has multiple research streams that use agent-based models—simulations of individual or group interactions that assess their behavioral impact on an entire system—to study elements of racial inequality.

For example, one project examines a quandary he and his colleagues observed among African American undergraduate women: They feel compelled to signal to their African American peers that they are “black enough,” while signaling to the larger society that they are “white enough” to be American.

The research, with Rachelle Winkle-Wagner of the University of Wisconsin–Madison and the University of Maryland’s Rashawn Ray, expands upon and interrogates a popular theory suggesting that minority students are socially marginalized, which contributes to the racial disparities in educational achievement.

Another project that will be in his forthcoming book, “How Many Racists? How Everyday People Contribute to a System of Social Inequality,” explores why there seems to be a significant decline in the number of people who hold racist beliefs, yet racial inequality remains high in many arenas.

Using a simple two-person Nash Bargaining model, Stewart’s investigation demonstrates that a system built on biased social institutions, even though administered or used by nonbiased individuals, can maintain racial inequality with a few, or even no, racists.

Racial Disparities in Causes of Mortality

Stewart set out to examine racial differences in mortality and how they have changed since the Jim Crow era, but quickly realized there was a “flaw” in previous research methods on disparities in mortality rates.

“So that’s when I set everything else aside,” he explained, “to write a paper to say, ‘Let’s come up with a better measure to estimate the role of underlying causes in mortality disparities.’ ”

While the usual method of causal decomposition quantifies disparities as differences in mortality rates, it does not account for the fact that many underprivileged groups are more likely to die from nearly all causes. Stewart’s “cause-deleted index” calculates underlying causes in mortality disparities as the change in the relative risk of dying that is related to deleting a specific cause. He found that using both methods together provides researchers and policymakers with a more complete picture of the disparities.

After developing the index, Stewart returned to examining the data from 1940–2000.

“The big takeaway right now is that [racially-based] mortality disparities have been very robust over the past 60 years, and they look very similar over this time period,” he said.

Understanding Acknowledgement Networks

A new research project Stewart is particularly excited about—and the one that keeps turning up in his dreams—is his study of academic networks with Saheli Nath, a student in a joint Kellogg and sociology doctoral program. With the help of IPR undergraduate research assistants, Stewart has cataloged the acknowledgements from articles in five prominent sociology journals—Demography, Social Forces, Social Problems, the American Journal of Sociology, and the American Sociological Review—between 1980 and 2012.

“We’re looking at how this network is structured, and particularly how it’s structured by field, and we’re using the network models to make inferences about who’s connected and which groups of individuals are tightly networked,” he explained, noting that data analysis began this summer.

“One of the preliminary findings is that you don’t see many women in the top 20, 25, or 50 central actors until the late 80s or early 90s,” he pointed out. “Then you see women become more prominent in it.”

“I keep dreaming of ways—every night almost—of how to look at this data differently and what different things we can do with it,” Stewart said.

 Quincy Thomas Stewart is associate professor of sociology and an IPR fellow. 

Photo: J. Ziv