Creating global impact through Yale’s data-driven social sciences
This story is the third in a series about Yale’s evolution under President Peter Salovey as he prepares to return to the faculty later this year.
To understand how people moved around during the COVID-19 pandemic, Yale sociologist Emma Zang needs data — a lot of it. When asked, she reels off a list of datasets she works with to analyze patterns: voter-registration data, cell-phone location data, credit-card transaction data, X (formerly Twitter) data, and net-light data that captures population density through satellite imagery showing the geographic areas glowing brightest at night.
Traditionally, mapping migration patterns required survey data, which takes great effort and time to compile, Zang explains.
“By the time you collect the data, clean it, and use it, it’s already two years old, which limits its usefulness,” said Zang, an assistant professor of sociology in Yale’s Faculty of Arts and Sciences (FAS). “These new, innovative datasets are game-changers.”
For Zang and other social scientists, the emergence of large, computerized datasets containing information from the government, nonprofit sector, and private enterprise is altering how they approach their work — and Yale, during President Peter Salovey’s presidency, has responded by investing broadly to facilitate this work and other aspects of data-driven social science.
In the past decade, the university has launched bold initiatives and built key infrastructure to create a collaborative campus environment for data-driven social science that is being used to guide domestic policy, address climate change and other pressing global challenges, and which seeks to understand human cognition, values, and behavior.
“Social science research has the potential to shape policy and effect change on a global scale,” said Salovey, a distinguished psychologist who will return to social science research and teaching when he steps down at the end of the academic year. “That requires bringing together deep expertise on data, on theory, and on policy, facilitating collaboration, convening leaders and guiding important conversations, and pursuing insights into human relationships and society.
“At Yale, we are committed to leading the way in this field, to understand urgent global challenges and increase knowledge, without partisanship or ideology.”
Since Salovey began his tenure, in 2013, the university has strengthened the social science faculty by hiring top-flight scholars, many of whom have taken on leadership roles at the university. It has enhanced social-science education by transforming the Department of Statistics into the Department of Statistics and Data Science and establishing a major in economics and computer science to teach undergraduates the skills needed to harness data for a better understanding of the world.
Yale has also opened several innovative and cross-cutting centers for research. These include the Data-Intensive Social Science Center, which helps scholars across the university access and manage the comprehensive, complex datasets that increasingly drive boundary-expanding social science research; the Institute for Foundations of Data Science, which applies the methodology of data science across multiple disciplines; and the Tobin Center for Economic Policy, which brings Yale’s widely recognized excellence in economics and other fields to bear on public policy.
In 2021, the university launched the Wu Tsai Institute, an interdisciplinary research center that combines neuroscience, social science, and data science to accelerate breakthroughs unlocking the mysteries of human cognition.
And in 2022, Yale established the Yale Jackson School of Global Affairs, the university’s first new professional school since 1976.
“These are all remarkable accomplishments,” said Alan Gerber, Sterling Professor of Political Science in the FAS, who served as its inaugural dean of social science from 2014 to 2021 and currently directs Yale’s Institution for Social and Policy Studies. “Through them, President Salovey has created a fertile environment that encourages collaboration and enables social scientists in the FAS and professional schools to pursue ambitious, exciting research that has a real-world impact.”
New areas of excellence
Over the past 10 years, Yale has built on its historic strengths while advancing new techniques in data and analysis.
The Tobin Center, established in 2019, exemplifies the approach, harnessing recent advances in economics, data science, and analytics to conduct rigorous, evidence-based research that helps define and inform policy debates.
“Yale is traditionally strong in economics,” Gerber said. “The Tobin Center is expanding that traditional excellence into a new area of enormous public importance.”
The center unites researchers across campus in the pursuit of policy-relevant scholarship while connecting them with lawmakers and government officials at the local, state, and federal levels.
For example, faculty from the Yale School of Management, FAS, and the Yale School of Public Health recently met with administrators from Connecticut’s Medicaid program to discuss how the latest evidence-based research might strengthen the state’s health system. The center’s staff includes data analysts and investigators to support scholarship, as well as veterans of the policy process to help faculty share their work with lawmakers.
SELECTED MILESTONES IN DATA-DRIVEN SOCIAL SCIENCES
While the Tobin Center supports research primarily in the domestic sphere, the Yale Jackson School of Global Affairs provides opportunities for social scientists to tackle international issues. Its faculty includes economists, political scientists, historians, and anthropologists studying agricultural markets in sub-Saharan Africa, authoritarian regimes and transitions to democracy, mental health and child development among refugee populations, and education policy in developing countries, among other issues of global importance.
Aside from their research, the school’s faculty members are helping to train a new generation of leaders equipped to embrace the use of data to inform policy. The school aims to give its students deep knowledge of the world around them, fluency with data, and an agility working across disciplines that will equip them to meet global challenges.
“The Jackson School is a unique convening space for scholarship on a range of pressing global issues. We have a multidisciplinary faculty — with top scholars from the humanities, social sciences, and Yale’s other professional schools as well as senior practitioners of global affairs — in dialogue with one another,” said Jim Levinsohn, dean of the Jackson School and Charles W. Goodyear Professor in Global Affairs. “We’re training our students to take this cross-disciplinary, collaborative approach and bring fresh insight to urgent problems facing the world.”
The Wu Tsai Institute, established through a gift from Joseph Tsai ’86, ’90 J.D. and Clara Wu Tsai, bridges multiple fields of neuroscience research, from biology to psychology, and data science to engineering. Its leadership team includes three social scientists and a neuroscientist: Nicholas Turk-Browne, a professor of psychology and the institute’s director; Kia Nobre, Wu Tsai Professor of Psychology, associate director of the institute and director of the Center for Neurocognition and Behavior; John Lafferty, the John C. Malone Professor of Statistics and Data Science and director of the Center for Neurocomputation and Machine Intelligence; and Daniel Colón-Ramos, the Dorys McConnell Duberg Professor of Neuroscience and Cell Biology and director of the Center for Neurodevelopment and Plasticity.
The institute has united more than 150 faculty across disciplines, including from four social science departments, in a moonshot effort to reveal the secrets of human cognition.
“President Salovey and Joe Tsai conceived of the Wu Tsai Institute over a breakfast in La Jolla, California,” said Turk-Browne. “They recognized that, in an era obsessed with the intelligence of machines, the remarkable capabilities of humans are not yet understood, including our ability to perceive, remember, think, decide, and create. These wonders of human cognition are the basic ingredients of all knowledge, relationships, organizations, and pursuits, and in this way, their understanding is fundamental for advancing social science.”
‘One giant sandbox’
Perhaps no development has been more important to the study, teaching, and practice of social science over the past several years than the proliferation of large data sets. The vast quantity of information now available informs the study of health care, education, political polarization, immigration, economic inequality, and climate change, among other pressing issues.
“Decades ago, a lot of social science research was largely theoretical, based on small datasets, or methodologically focused on statistical techniques that were designed to compensate for the relatively poor quality of data,” said Steven Berry, the David Swensen Professor of Economics in the FAS and the inaugural faculty director of the Tobin Center for Economic Policy.
“Computerization has made many new sources of data available, which become potential datasets for the study of human behavior, social behavior, and economic behavior.”
The rise of these massive datasets presents challenges. While the Marx Science and Social Science Library provides researchers access to many important datasets, they also must negotiate user agreements with the vendors that compile them, devise way to securely store the data that often includes confidential information, and learn computing techniques, including machine learning and artificial intelligence, to analyze it.
In 2022, Yale established the Data-Intensive Social Science Center (DISSC) to help researchers across FAS and Yale’s professional schools — the faculties of Yale School of the Environment, the Yale School of Management, Yale School of Public Health, the Yale Jackson School of Global Affairs, and Yale Law School all include social scientists — looking to access, analyze, and preserve enormous amounts of data.
Faculty members have always found ways to access and analyze the data they need, but centralized support offered by the DISSC accelerates that process and ensures all researchers are supported, said Gerber, who serves as the DISSC’s faculty director along with Berry. Creating a center that accelerates the adoption and ease the use of methods like AI and machine learning, Gerber said, is a “truly visionary effort.”
“To do that as a university service open to all faculty in the social sciences distinguishes Yale from others.”
In its first year of operation, the DISSC provided programming to help faculty stay apprised of the latest innovations in analyzing and managing data, began helping faculty negotiate the cumbersome process of reaching user agreements with vendors for access to datasets, and launched a pilot project with Yale Amazon Web Services to create secure space on the cloud to safely store datasets.
Storing data in a centralized location allows multiple researchers to work without having to move the datasets around or make copies of them, explained Ron Borzekowski, the DISSC’s inaugural director. If a researcher adds value to data by cleaning it or appending new variables, then the enhanced version becomes available to anyone with login authorization, he said.
“Centralization is the key,” Borzekowski said. “Instead of working in four or five distinct silos, they’ll be working in one giant sandbox.”
The DISSC embodies Salovey’s goal of creating a more unified Yale, Berry said.
“It allows social scientists across the university to benefit from the same infrastructure,” he said. “When a problem is solved once or progress is made one project, or for one individual, that progress becomes available to other researchers.
“I don’t know of any other university in the country that is trying this at this scale.”