The Urban Data Research Lab uses novel big data and spatial data science to understand urban inequality at the scale of the neighborhood in order to inform housing policy and opportunities for more equitable cities.
Our primary areas of research are:
Measuring Neighborhood Dynamics with Big Data
Much of how researchers, advocates, and policy-makers understand the geography of opportunity is largely based on the residential neighborhood context. To study neighborhood dynamics, especially at a large scale, we have traditionally relied on administrative data such as those produced by the U.S. Census Bureau. This area of research aims to think, measure, and generate data products to describe the wide range of activities and socio-spatial dynamics that make up how people experience their social and environmental context and how these can change across the day or seasonally. We investigate the possibilities of using novel, unstructured data sources such as newer iterations of cell phone location data, in a ground-truthed and statistically rigorous manner, to shift how we think about context in a more holistic, representative way.

Legacies of Housing and Real Estate in the 20th Century
Public policy in the U.S. is at a critical turning point where the need to address historical and ongoing housing discrimination calls for more nuanced understandings of longstanding spatial inequalities and their impacts on residential segregation and stratification. We study large-scale historical housing discrimination through such as practices as federal-level redlining and urban renewal in the mid-20th century United States and its impacts on housing and socioeconomic outcomes. In this area of research, we aim to add nuance to historical narratives through the discovery and analysis of historical big data at the scale of the neighborhood and the individual, with experiences that may not be measurable by traditional data sources.

COURSES


PUBLICATIONS
A National Zoning Atlas to Inform Housing Research, Policy, and Public Participation
Xu, Wenfei, Scott Markley, Sara C. Bronin, and Diana Drogaris.
Cityscape, 25(3), 55-72.
2024
The contingency of neighbourhood diversity: Variation of social context using mobile phone application data
Xu, Wenfei
Urban Studies, 59(4), 851-869.
2022
Ghost cities of China: Identifying urban vacancy through social media data
Williams, Sarah, Wenfei Xu, Shin Bin Tan, Michael J. Foster, Changping Chen
Cities, 94, 275-285.
2019
Housing Markets, Residential Sorting, and Spatial Segregation
Tan, Shin Bin, Wenfei Xu, Sarah Williams
China Urbanizing, University of Pennsylvania Press
2022
Where did Redlining Matter?: Regional Heterogeneity and the Uneven Distribution of Advantage
Xu, Wenfei
Annals of the American Association of Geographers, 113(8), 1939-1959.
2023
Legacies of institutionalized redlining: a comparison between speculative and implemented mortgage risk maps in Chicago, Illinois
Xu, Wenfei
Housing Policy Debate, 32(2), 249-274
2022
A roundtable discussion: Defining urban data science
Organizers, Kang, Wei, Oshan, Taylor, Wolf, Levi J., Discussants, Boeing, Geoff, Frias-Martinez, Vanessa, Gao, Song, Poorthuis, Ate, and Xu, Wenfei
Environment and Planning B: Urban Analytics and City Science, 46(9), 1756-1768
2019
Is "Regulation from Below" Possible?
Xu, Wenfei
Public Books
2022
NEWS
Announcing the Urban Data Hub at Cornell AAP
April 3, 2023
The Cornell AAP Urban Data Hub is a cross-disciplinary initiative aimed to catalyze longer-term urban big data infrastructure within AAP to accelerate our research, teaching, and engagement with contemporary and critical discourse in design and planning practice.
PEOPLE
Wenfei Xu is an assistant professor in GIScience and Urban Data Science in the Department of Geography at UCSB and the Director of the Urban Data Research Lab. Her research examines how housing policies, practices, institutions, and technologies have influenced urban inequality, with a focus on methods in spatial data science. She works on topics in social-spatial stratification, segregation, race and ethnicity, data science, mapping, and neighborhood change in the United States.
Wenfei's work, which has been funded by the National Science Foundation, the Russell Sage Foundation, and the Washington Center for Equitable Growth, ranges from an interest in the historical legacies of structural housing discrimination and its contemporary spatial-temporal manifestations to exploring the uses of big data in characterizing human activity for urban social science research. Wenfei has published in several journals, including Proceedings of the National Academies of Science, Annals of the Association of American Geographers, Urban Studies, Nature Scientific Data, and Housing Policy Debate, among others. She currently serves on the editorial board of Nature Scientific Data and the Annals of the AAG.
Wenfei was previously an assistant professor at Cornell University and a postdoctoral scholar at the University of Chicago, Mansueto Institute for Urban Innovation. She holds a Ph.D. in Urban Planning from Columbia University, dual M.Arch and MCP degrees from MIT, and a B.A. in economics from the University of Chicago. Before her doctorate, Wenfei was a data scientist at the Department of Homeless Services and at CARTO.

Wenfei Xu
Director, Urban Data Research Lab
Current Members

I am Houpu Li, currently pursuing a Ph.D in Geography at Cornell University. I hold a Master's Degree in Regional Science from Cornell University, with a focus on quantitative analysis of urban spatial data.
Houpu Li
Ph.D. Student, Geography
Past Members
Moheng Ma
Ari Rousakis
Leah Chen
Charlotte Verity
Youssef Attia
Ben Zaccara
Nada Attia
Nirbhay Narang
Allie Chu
Tony Zong
Su Jeong Jo
Stella Frank
Yucheng Zhang
Michael Cao
Zoe Wang
Tung Chen
Rifqi Maluana
Jessie Fujii
Dhruv Parekh
Xueting Jin
Kanjii Fateema
Emilia Lam
CONTACT US
Please reach out to wenfeixu@ucsb.edu if you're interested in learning more or collaborating with us. Our ongoing lab projects touch on issues of housing and neighborhood change through urban big data (ranging from archival to cell phone data) and urban data science methods.
For prospective PhD students who are interested in working with us, we are particularly interested in students with a social science background and strong coding skills.