Research Papers

We developed a nonparametric model to study intergenerational income mobility, allowing for nonlinearities and interactions among factors like race, education, and parental age. Using PSID data, we analyzed how these dynamics affect children’s economic outcomes.

Standard models often overlook complex interactions in mobility. Our approach captures how race and parental education shape opportunities, offering a more nuanced understanding of barriers to economic mobility.

Our findings reveal how interactions between parental income, race, and education create mobility bottlenecks, providing insights for policies aimed at reducing economic inequality and promoting upward mobility.

We developed new measures of intergenerational persistence using “memory curves” to track how socioeconomic origins influence mobility across generations. Using U.S. data, we analyzed how initial conditions shape transitions between occupational classes and aggregated insights to describe mobility trends in the population.

Understanding intergenerational mobility requires tools that capture the long-term influence of socioeconomic origins. By modeling these dynamics as “memory,” we aim to better explain how class persistence varies and uncover patterns missed by traditional metrics.

Our findings show that class origins often fade within three generations, but mobility varies significantly by occupational background. These insights help inform policies aimed at addressing inequality and promoting equal opportunities across diverse socioeconomic contexts.

We analyzed a generalized mobility model using administrative data from Denmark to examine the roles of family selection and locational characteristics in generating spatial heterogeneity in intergenerational mobility across nearly 300 larger and 2,000 smaller neighborhoods.

We aimed to understand why intergenerational mobility varies across neighborhoods, an essential issue for shaping economic and social policies.

Understanding the mechanisms behind spatial differences in mobility helps policymakers target interventions to improve equality of opportunity across different neighborhoods.

We combined a special class of deep neural networks with directed acyclic graphs to improve our ability to draw inferences about entire systems of causal relationships.

Social science theories often postulate causal relationships among a set of variables or events, but when it comes to empirically evaluating them, researchers typically resort to methods that impose unrealistic assumptions on the data, leading to incorrect inferences. Our approach can model these relationships without invoking such restrictive assumptions.

Our approach allows researchers to more accurately learn causal relationships from data. It therefore has a very wide scope for application, ranging from studies of inequality and social mobility to health, education, and psychology.

We offered a comparison between the distributional Recentered Influence Function (RIF) and the conventional Oaxaca-Blinder decomposition methods. Additionally, we compared the multi-year cross-sectional Survey of Consumer Finances with the longitudinal Panel Study of Income Dynamics.

We tested the hypothesis that inheritance receipt explains a larger portion of the black-white gaps in net worth compared to human capital, as measured by educational attainment.

The black-white wealth gap exhibits a surge in inequality across the percentiles of the net worth distribution. It is important to determine whether the decomposition at the mean yields different results compared to the decomposition at the median or over the entire distribution.

In the context of Health and Human Resource outcomes, we developed and exemplified comparison instruments for ordered categorical data that don’t require artificial Cantril or Likert scaling of outcomes to facilitate level and variation comparisons.

Artificial cardinal scaling of ordered categorical data has been heavily criticized recently because of the inherent ambiguity it engenders (different, equally valid scales can result in substantially different conclusions). These new instruments resolve the issue by providing scale independent comparators that have an unambiguous probabilistic basis.

The generation and use of ordered categorical data has proliferated in the modern era in all areas of the empirical sciences and there is a need for analysis tools which are not subject to ambiguity critiques.

We evaluated the hypothesis that neighborhood effects on academic achievement are mediated or interact with school quality using data from the Early Childhood Longitudinal Study.

We aimed to explore the mechanisms through which neighborhood poverty affects academic achievement, focusing on the role of school quality. By examining diverse measures of school effectiveness, resources, and climate, we sought to provide insights into whether and how these factors mediate or interact with neighborhood effects.

Our findings indicate that living in a disadvantaged neighborhood reduces academic achievement, but find little evidence that this effect is mediated by measures of school quality. These results challenge prevailing hypotheses and suggest that neighborhood effects operate independently of school quality factors. Addressing concentrated poverty directly may therefore be crucial in reducing educational inequalities.

We measured the impact of the Alaska Permanent Fund Dividend on price levels using a synthetic control method to compare Alaska with similar states before the dividend’s introduction.

We aimed to determine whether unconditional cash transfers, like the Alaska Permanent Fund Dividend, stimulate consumer spending and increase price levels.

Our findings help us understand the inflationary effects of unconditional cash transfers, informing policymakers about potential economic impacts and guiding future decisions on similar programs.

We developed a new method for estimating randomized intervention analogues to the natural direct effect (rNDE) and the natural indirect effect (rNIE) in the presence of treatment-induced confounders.

We aimed to address the challenge of estimating mediation effects when treatment-induced confounders complicate the mediator-outcome relationship. Existing methods require cumbersome weighting procedures, making them difficult to use. Our method provides a more practical and accessible approach for researchers by utilizing simpler regression-based techniques.

Our approach allows researchers to estimate mediation effects more efficiently and accurately, even when treatment-induced confounders are present. By simplifying the estimation process, this method makes causal mediation analysis more accessible and robust.

We introduced “residual balancing” as an alternative method for constructing weights in marginal structural models (MSMs) to estimate time-varying treatment effects, aiming to address issues inherent in inverse probability weighting (IPW).

We sought to overcome limitations of IPW, such as inefficiency, sensitivity to model misspecification, and difficulty with continuous treatments.

Residual balancing improves the efficiency and robustness of estimating marginal treatment effects in the presence of time-varying confounders. By offering a more reliable and user-friendly alternative to IPW, it enhances causal inference accuracy in observational studies.