Detecting Effect Heterogeneity: An application to the benefits of college in reducing poverty

Xu, Brand, Shinkre, and Jeon are working on a project titled “Flexibly Detecting Effect Heterogeneity: An Application to the Benefits of College in Reducing Poverty.”

RECENT DEVELOPMENTS IN CAUSAL INFERENCE AND MACHINE LEARNING

Jeon, Nanum and Jennie E. Brand. Causal machine learning for sociology. Revise & Resubmit, Cologne Journal of Sociology and Social Psychology (Kölner Zeischrift für Soziologie und Sozialpsychologie)

Jeon, Nanum. Deep-learning-based g-computation for life course sociology.

Brand, Zhou, and Yu Xie (Princeton University) published a review of recent developments in causal inference for Annual Review of Sociology.

Lundberg, Brand, and Jeon published a review of recent developments in machine learning for Social Science Research.

Uncovering Effect Heterogeneity with Causal Trees

Brand, Xu, Geraldo, and Koch use tree-based machine learning, i.e. causal trees, to recursively partition a sample to uncover sources of effect heterogeneity. A paper was published in Sociological Methodology.

COLLEGE EFFECT HETEROGENEITY

In her recently published book, Brand contends that individuals with different backgrounds differ in their benefits of college. Brand argues that those who are less likely to complete college have considerable gains, especially in circumventing lives marked by disadvantage.

COLLEGE direct and indirect EFFECT HETEROGENEITY

Ahearn, Brand, and Zhou employ a nonparametric causal mediation analysis to examine the total, direct, and indirect effects of college attendance on voting and how these effects differ across individuals with different propensities of attending college. A paper was published in Research in Higher Education.

DETERMINANTS OF DOWNWARD MOBILITY

Rouhani, Brand, and Xi Song (University of Pennsylvania) offer an analysis of several potential factors influencing declining absolute intergenerational income mobility.

Disruptive Event EFFECT Heterogeneity

Aquino, Brand, and Florencia Torche (Stanford University) review recent findings of heterogeneity in the impact of disruptive events. A paper was published in Sociology Compass.

INCOME VOLATILITY

Lundberg, Brand, and Aquino are working on effective data visualization tools for income volatility.