Amount Awarded: $25,000
Despite the ideal of equal opportunity regardless of birthplace, neighborhoods decisively shape education, health, income, and overall well-being. This project addresses a critical limitation at the frontier in neighborhood effects research by moving beyond average effects – masking substantial individual variation – to identify which groups are most affected by neighborhood conditions. We use a unique combination of novel causal machine learning methods and rich population-wide registry data spanning multiple decades to discover previously unrecognized vulnerable subgroups that can inform targeted policy interventions.
Our collaborative project creates a strong synergy unavailable to either institution alone: Professor Harding (UC Berkeley) contributes world-leading substantive and methodological expertise in neighborhood effects and causal inference, while Dr. Widding-Havneraas (University of Oslo) brings access to exceptional Norwegian registry and genotyped data. This combination enables us to address two key questions: (1) Which groups are most vulnerable to neighborhood effects on health, education, and labor market outcomes? (2) Can data-driven causal machine learning methods reveal previously unidentified at-risk subgroups?
A research stay at UC Berkeley for Dr. Widding-Havneraas (11 months) is essential to this collaboration as it provides direct access to Berkeley's exceptional methodological resources and meeting arenas, including the Center for Targeted Machine Learning, Berkeley Institute for Data Science, and weekly causal inference seminars. This research stay will also foster a strong and consistent collaboration, build lasting networks, and position Widding-Havneraas, an early career researcher, to develop US-Norway research projects through future planned grant applications.
Our project will produce two high-impact studies examining how neighborhoods impact health outcomes and education/labor market outcomes, advancing interdisciplinary knowledge across sociology, economics, and public health. Our findings will additionally provide evidence for targeted policies to reduce persistent inequalities and promote equal opportunities, thereby addressing central pressing societal challenges.