Nathaly Villavicencio
Mar 2, 2026
A comprehensive analysis examines the growing use of GPS-based mobility data in applied economic research and its policy implications.
A newly published article in the Journal of Economic Surveys examines how large-scale human mobility data, combined with machine learning tools, is reshaping applied economic research across multiple fields, including labor markets, transportation, public health, and urban economics.
The article, “Harnessing Human Mobility Data for Applied Economic Research: Current Knowledge, Challenges, and Emerging Opportunities,” is authored by Dr. Cristina Connolly (University of Connecticut), Dr. Sandro Steinbach (North Dakota State University), Dr. Mike Vo (University of Connecticut), and Dr. Xibo Wan (University of Connecticut). The paper provides a comprehensive review of the rapidly expanding literature using GPS-based mobility data and outlines both methodological challenges and future research directions.
The authors highlight how mobility datasets offer real-time, high-resolution insights into travel behavior, social interactions, migration patterns, and labor market dynamics, often surpassing traditional survey and administrative data in spatial and temporal precision. At the same time, the paper emphasizes important limitations, including measurement error, sampling bias, dimensionality concerns, and privacy protections such as differential privacy adjustments.
The review also examines how machine learning methods, including double machine learning, causal forests, and counterfactual simulations, expand economists’ ability to analyze high-dimensional mobility data and generate policy-relevant insights. The authors identify emerging opportunities in disaster response, transportation systems, segregation analysis, climate-related migration, and labor market research.
“Mobility data provide unprecedented opportunities to understand how people respond to economic shocks and policy changes in real time,” said Dr. Sandro Steinbach. “At the same time, researchers must carefully address representativeness, privacy, and methodological rigor to ensure reliable and policy-relevant results.”
The full article is available at: https://doi.org/10.1111/joes.70047
Media Contacts:
Dr. Sandro Steinbach – sandro.steinbach@ndsu.edu - North Dakota State University