"Why does Costa Rica outperform the United States in life expectancy?  A tale of two inequality gradients"

Abstract: Costa Rica is among the few low or middle income countries with high quality adult vital statistics mortality data. We link these mortality records with census data to create a Costa Rican National Longitudinal Mortality Study, and compare adult mortality patterns to those in the United States. We find that mortality in the U.S. is 18% higher than in Costa Rica among adult men and 10% higher among middle-aged women, despite the several times higher income and health expenditures of the U.S. The U.S.’s underperformance is strongly linked to its much steeper socioeconomic (SES) gradients in health. Although the highest SES quartile in the U.S. has better mortality than the highest quartile in Costa Rica, U.S. mortality in its lowest quartile is markedly worse than in Costa Rica’s lowest quartile. Further examination of cause-specific mortality and risk factors suggest that these patterns are strongly related to behaviors leading to lung cancer and heart disease.

View Podcast Here!

"Kernel balancing: a weighting approach for causal inference and sample adjustment"

Abstract: When making causal inferences from observational data under the assumption of no unobserved confounders, matching and weighting estimators are used to adjust the joint distribution of observed covariates for treated and control units to be the same. Similarly, investigators often have data from an observed sample, which they wish to adjust to make more similar to a target sample or known population. However, existing weighting and matching approaches for both problems have important limitations: matches are generally not exact, and standard weighting approaches ensure that the observed sample is similar to the target sample/population only on a finite set of pre-specified moments. I introduce kernel balancing, first in the context of causal inference and then as a solution to the general sample-adjustment problem. The method works by taking a high-dimensional expansion of the observed covariates, and choosing weights on the control group (or observed sample) such that it has equal means to the treated group (or target sample) on this high-order expansion of the covariates. By using kernels, it is possible to choose an expansion such that all continuous functions of the covariates are linear in that expansion. This proves very desirable, as the weighting then ensures that any unspecified but plausibly important continuous function of the covariates (such as a ratio of two variables) will automatically have the same means for the two groups as well. I provide empirical examples, and show that this method also implies that a particular estimator of the entire multivariate density of covariates is the same for the two samples at every observed location in the covariate space. An R package implementing the procedure is available from the author.

"Immigrant Generation and Adolescent Childbearing in the United States"

Abstract: Despite recent declines, the teen birth rate in the United States is still the highest among high-income countries. Immigrant youth can be expected to increasingly shape US trends in adolescent childbearing as their share of the youth population continues to grow. About one in four US children has foreign-born parents currently, up from 6% in 1960; this share is projected to rise to one-third by 2050. In this study, I use data from the National Longitudinal Study of Adolescent Health (Add Health) to examine how the risk of early childbearing varies by immigrant generation; to what extent generational variation reflects discrepancies in the timing of sexual onset (versus post onset factors); and what family, neighborhood, and individual-level social factors underlie generational differences. I will also describe a new data collection project called the mDiary Study of Adolescent Relationships, which pairs a year-long diary study with an ongoing birth cohort study to increase understanding of the content and quality of teen partnerships over time, and of the childhood precursors and health and developmental consequences of teen relationship behavior.

"Beyond LATE: Economic Choices and the Identification of Multiple Treatment Effects "

Abstract: “Monotonicity” refers to a condition in choice models with instrumental variables in which a local variation of an instrument shifts all agents toward or against a choice. This paper presents a useful framework to investigate the role of monotonicity in the identification of causal effects in multiple choice models with categorical instrumental variables.  I first examine a new monotonicity condition that applies to unordered choice models with multiple treatments.  Like its analogous property in the binary choice model, I show that unordered monotonicity imply and is implied by additive separability in observables and unobservables in choice equations.  I show that unordered monotonicity may arise from preference properties of choice behavior. I then exemplify the use of preference properties to identify causal effects in choice models where monotonicity does not hold. I show that identification and equivalence results flow from simple properties of binary matrices.

"The Shifting Structure of Intragenerational Inequality"

Abstract: Traditional stratification and inequality research often treats individuals as single points of observation in the stratification system. This paper extends current scholarship on economic inequality by invoking the life course perspective to study the intragenerational pattern of wage inequality, focusing particularly on how its structure has changed across cohorts. Using over 40 years of national representative data from CPS and PSID, I found that inequality increases over the life course for all cohorts born between 1941 and 1970. Further, cross-cohort comparison reveals that the amount of intragenerational growth of inequality has increased from earlier to later cohorts, suggesting that the labor market plays a more important role in generating inequality in recent years. Microlevel decomposition analysis suggests that the relative importance of the underlying mechanisms for intragenerational inequality has also shifted across cohorts, with a growing amount of intragenerational growth of inequality attributable to education-based cumulative advantage and residual inequality.

"Death and Desirability: Retrospective Reporting of Unintended Pregnancy after a Child’s Death"

Abstract: Social scientists have long debated how to best measure pregnancy intentions. The standard approach in domestic and international surveys is to use mothers’ retrospective reports of their intentions at the time of conception. Because women have already given birth at the time of this report, their children’s health may influence their responses. Some fertility scholars have argued that mothers will be more likely to recall “lower quality”—that is, unhealthy—children as unintended. What about children who are no longer alive? In contemporary sub-Saharan Africa, where one in ten African children die before their fifth birthday, this research aims to understand whether women are susceptible to recalling deceased children, and the pregnancies from which they resulted, in a more positive light. Leveraging Demographic and Health Survey data from 31 sub-Saharan African countries, I will demonstrate how children’s health and vital status is associated with women’s retrospective reports of pregnancy intentions. The results challenge the reliability of standard measures of pregnancy intentions in high mortality settings, and thus our current knowledge of the levels and consequences of unintended pregnancies in these contexts.

"Restless Denominators"  

Abstract: This paper explores how denominators are used, misused, and—especially—how often they are missing, and to what effects. The paper makes three arguments. First, that denominators are essential in domains far beyond the quantitative disciplines that presently attend to them. In ethnography, in practical politics, in cultural studies, in everyday decision-making, we need to think much more about the pools of possible chances out of which emerge the events we observe. Second, denominators in the social sciences are much more intently theoretical objects than their usual treatment suggests, both in the sense that populations are not naturally bounded in the ways that many statistics imply, and in the sense that people do not merely find themselves randomly in certain populations facing certain risks, but rather participate in a variety of ways in their location. Finally, I argue that a classical demographic way of thinking about the denominator—as exposure to risk—offers an elegant way of integrating contemporary theory about uncertainty, agency, and habitus into formal quantitative research.

Vimeo Podcast

"Thirty-Five Years Later: Long-Term Impacts of the Matlab Maternal and Child Health Program on Migration and Labor Market Outcomes"

Abstract: Despite global proliferation of reproductive health and family planning programs, little is known about their long‐term effects. This talk introduces a project that provides causal evidence on the impact of the Matlab Maternal and Child Health and Family Planning Program (MCH‐FP) in Bangladesh, over thirty‐five years after it began. The Matlab Health and Socioeconomic Survey 2 (MHSS2) collected surveys and objective tests from a sample of 36,000 participants, including beneficiaries and their descendants, with followup of ~90% of migrants living outside the study area or abroad. I introduce the larger study in the context of evaluating MCH/FP program impacts on migration and labor market success. To the extent that family planning reduced competition for resources, the program may discourage migration among treated populations. However, to the extent that child health services increased human capital, treated individuals may be better positioned to migrate successfully. The results illustrate the importance of intensive migrant followup for reducing attrition bias.

"Health Equity in Family Planning and Family Planning Care: Putting the Focus back on Women"

Abstract: Differences in contraceptive use and unintended pregnancy by race/ethnicity are well described. Interpretation of these differences, and how they relate to the desire to achieve health equity, is complex due to the nature of reproduction, including the personal, social and historical context in which reproductive and contraceptive decisions are made. Lack of attention to these contextual factors has the potential to interfere with the ultimate goal of optimizing women’s reproductive health and to exacerbate health inequities. This talk will review data regarding women’s reproductive outcomes and how they vary by their sociodemographic characteristics, and discuss conventional approaches in both public health efforts and clinical family planning care to thinking about and responding to these data. I will then make the case for a woman-centered approach that focuses on individual’s preferences and conceptualizations of reproductive health as the best strategy to meet women’s needs and promote health equity.

View Podcast Here

"Social Network Influences on and for Behavior Change"

Abstract: In this presentation, Valente will show how social networks influence behaviors across a wide variety of applications.  Recent research on the diffusion of innovations will also be presented along with results from the study of the diffusion of the Framework Convention on Tobacco Control (FCTC).  Models for how social networks can be used for program implementation and network interventions will also be detailed.

View Podcast Here

"Precarious employment and fertility: Insights from Japan’s “Lost 20 Years”"

Abstract: In this paper, we examine relationships between precarious employment and fertility. We focus on Japan, a country characterized by a prolonged economic downturn, significant increases in both unemployment and non-standard employment, a strong link between marriage and childbearing, and pronounced gender differences in economic roles and opportunities. Analyses of retrospective employment, marriage, and fertility data for the period 1990-2007 indicate that changing employment circumstances for men are associated with lower levels of marriage while changes for women are associated with higher levels of marital fertility. These two offsetting relationships combine to limit the overall association between changes in employment circumstances and declining fertility. Results of counterfactual standardization analyses suggest that Japan’s total fertility rate (TFR) would have been slightly lower than observed if aggregate- and individual-level employment conditions had remained unchanged from the 1980s. We discuss the implications of these results in light of ongoing policy discussions about fertility promotion and academic debates about the changing nature of gender relations within the family.

"Race Matters: Income Inequality and Mobility from 2000 to 2013"

Abstract: Using unique linked Internal Revenue Service and U.S. Census Bureau data matched at the individual level, we examine the differences in levels and trends for income inequality across racial and ethnic groups in the United States. Our data span 2000 to 2013, a period including the Great Recession, and will thus inform us on how financial crises affect inequality for important sub-groups of Americans. Previous research has focused on the increasing concentration of income and assets in the top decile of tax filers; this analysis will provide information on the entire distribution of income by decile, highlighting the circumstances of those at the lower end of the distribution as well. Because our data include both administrative and census data, our research is the first to provide detailed income and wage inequality information for racial and ethnic groups. Finally, we focus on short-run measures of mobility and document this measure over the decade and post-Great Recession era.

View Podcast Here

"Measuring neighborhoods in the new data economy."

Abstract:  The data economy in the United States has changed dramatically in the past 5-10 years.  Naively one might argue that this new data economy holds particular promise for academics, because revolutions in science are often preceded by revolutions in measurement.   But for social scientists who study cities in the United States these changes are mixed.  The new data economy is complex complicates the study of neighborhoods.  In this talk I'll describe one such complication - the replacement of the long form of the decennial census with the American Community Survey in 2010.  The ACS produces estimates for thousands of variables at a variety of geographic scales.  However, estimates from the ACS are terribly imprecise, for many policy relevant variables ACS estimates are almost unusable.  In this talk I’ll describe the quality of the ACS and use its shortcomings to motivate a discussion of changing the way we measure neighborhoods.  Rather than just talk about alternatives I’ll present results from two novel computational methods that leverage new ways of thinking about the measurement of neighborhoods.  One of these methods can be used to process existing public domain ACS estimates to dramatically reduce the margin of error.