Selection bias is a central problem for causal inference in the social sciences. Quite how central a problem it is, however, is often obscured by ambiguous terminology, needlessly technical presentations, and narrow rules of thumb. This paper uses directed acyclic graphs (DAGs) to advance a precise yet intuitive global definition of endogenous selection bias and argue its theoretical and practical centrality for causal inference.
Past macro-level research linking population and economic development has emphasized national averages. Studies in this vein focus on national fertility rates and age structure (on the population side) and average per capita incomes or national school enrollments for instance (on the economic side). As a complement to this literature, our research draws attention to population influences on economic differentiation, both within and between countries.
A growing literature suggests that stressful events in pregnancy can have negative effects on birth outcomes. Some of the estimates in this literature may be affected by small samples, omitted variables, endogenous mobility in response to disasters, and errors in the measurement of gestation, as well as by a mechanical correlation between longer gestation and the probability of having been exposed. We use millions of individual birth records from Texas to examine the effects of exposure to hurricanes during pregnancy.
This presentation will focus on a program of research that seeks to identify and better understand the multiple biological pathways through which social factors impact on trajectories of aging. Major social factors of interest include the broad impact of socio-economic status as well as the influence of patterns of social interaction with others (including both the positive and negative features of such interactions and their respective impacts on physiology).
This paper analyzes the impact of baseline household income and scholastic ability on post-secondary enrollment in South Africa. Using longitudinal data from the Cape Area Panel Study (CAPS), we analyze the large racial gaps in the proportion of high school graduates who enroll in post-secondary education. Our results indicate that baseline income and ability (measured in CAPS¿ literacy and numeracy evaluation) are strong predictors of post-secondary enrollment and statistically account for all of the black-white difference in enrollment.
The 1994 discovery of arsenic in ground water in Bangladesh prompted a massive public health effort to test all tubewells in the country and convince nearly one-quarter of the population to switch to arsenic-free drinking water sources. According to numerous sources, the campaign was effective in leading the majority of households at risk of arsenic poisoning to abandon backyard wells in favor of more remote tubewells or surface water sources, a switch widely believed to have saved numerous lives.
We have reserved the last DuPRI slot on Thursday, April 26 for a practice session for students presenting at PAA this year during the regular DuPRI time slot. We hope there will be a critical mass of faculty in addition to other students to provide feedback on the presentation. For all three students, this will be the first PAA presentation.
In a very real way, the rise of undocumented migration and the growth of America's undocumented population are a product of poorly conceived immigration and border policies, which in the course of a few decades transformed Mexico-U.S. migration from a stable, circular flow of male Mexican workers going to three states into a much larger settled population of Mexican families living in 50 states.
Little is known about the health effects of the economic downturn, with foreclosure one of its most visible signs. Research in economics suggests that the impact of an economic downturn is first felt through emotional well-being. Hence we explore onset of depression over the interval of the economic downturn with a unique data source, the National Social Life Health and Aging Project (NSHAP). Fortuitously NSHAP wave 1 was collected in 2005-2006 and wave 2 in 2010-2011, bounding the economic downturn and foreclosure crisis.