Seminar Series

This paper provides a systematic analysis of identification in linear social networks models. This is both a theoretical and an econometric exercise in that it links identification analysis to a rigorously delineated model of interdependent decisions. We develop a Bayes-Nash equilibrium analysis for interdependent decisions under incomplete information in networks that produces linear strategy profiles of the type conventionally used in empirical work and which nests linear social interactions models as a special case.

There has been great interest in one dimension of mortality change, aggregate human life expectancy. I focus on a distinct dimension, the variance in the age at adult death. I explain why this measure matters, discuss historical trends in this variance, and compare trends across countries. I discuss the relationship between the pattern of adult death and socioeconomic inequalities, in factors such as education and income, using data from the US. Finally I examine the sources of variance in the context of models of the life cycle.

ABSTRACT: Scholars question whether Latino, especially Mexican American, communities contain social, financial, and ethnic resources that foster mobility and integration. This research examines Latino ethnobanks in Los Angeles that are chartered by Mexican American elites. We find that Mexican American elites charter ethnobanks to make a profit, but they operate them within a "banktivist" model. They view Latino ethnobanks as community capital institutions that have the potential to grow the Latino middle class by providing access to capital for Latino entrepreneurs.

ABSTRACT : Life expectancy at birth in the United States is currently among the lowest of all high-income countries. Previously, research and policy discussions focused largely on cross-national mortality differences at older ages (e.g., at ages 50 and above). This talk highlights several key dimensions of Americans' mortality disadvantage relative to a set of 16 high-income comparison countries, with an emphasis on estimating the contribution of mortality differences at younger ages to the U.S.

ABSTRACT: Despite widespread success in using Twitter data to explain what people are doing or talking about, little attention has been paid to developing systematic ways of gathering demographic information from this data source. This paper develops a scalable, sustainable toolkit for social science researchers interested in using Twitter data to examine behaviors and attitudes, as well as understand the populations expressing them. We begin by describing how to collect Twitter data on a particular population ¿ in this case, individuals who did not plan to vote in the 2012 U.S.

ABSTRACT: In analysis of longitudinal data about migration behavior in rural northeastern Thailand, we examine the association of migration behavior with patterns of NDVI signals to estimate exposure to non-modal clusters over varying periods of retrospective time. Our hazard model approach indicates significantly different associations between rural out migration and return migration depending on the length of the temporal lens.

ABSTRACT: We carry out an analysis of societal variations in the process of educational attainment using a multilevel modeling strategy to assess how societal modernization, educational expansion, social inequality, a world-wide secular trend toward greater equality of opportunity, and communist educational policies affect the dependence of educational attainment on parental status and the gender gap in educational attainment. Using data from 541 sample surveys conducted in 54 countries, we define five-year birth cohorts ranging from the late 19th century through the late 20th century.

ABSTRACT: How does living in a violent environment "get into the minds" of children to affect them as they engage in daily life at home and in school? The research I will discuss develops new methods to identify the causal effect of exposure to incidents of extreme violence in children's neighborhoods. By exploiting the timing of events that occur in children's neighborhoods, this research examines how violence affects children's cognitive functioning, their ability to maintain attention and control impulses, and their performance in school.

ABSTRACT: This paper critically evaluates available data on divorce and the dissolution of cohabiting unions. We find that both vital statistics and retrospective survey data on divorce after 1990 are deeply flawed, and have greatly underestimated recent marital instability. These flawed data led many analysts to conclude that divorce risk has been stable or declining for the past three decades.

Incarceration research catalogues its direct negative impact on former inmates and their families, though the effects of punishment clearly spill over to affect broader economic and political institutions as well. To further expand the scope of incarceration research, this study examines spillover effects between state-level incarceration rates and the functioning of the U.S. health care system.