Background GDP per capita is arguably the single most predictive determinant of health at the national level, yet such a measure fails to capture the complexity of wealth (and thus health) distribution within a population. Health at the individual level has been shown to be associated with asset-based measures of wealth, but the relationship hasn’t been demonstrated consistently at the national level. Currently existing wealth and inequality measures are not comparable across surveys, across countries, or across time. In this thesis, I quantify wealth inequality using an asset-based measure by country and year, from 1990 to 2010, and study associations between inequality and several measures of health through the lens of the Millennium Development Goals. Methods I generate a global asset-based wealth index that is comparable across countries and over time using 461 surveys from 140 countries, anchoring this scale . . .

Exploiting cross-state variation in mortality from flu/pneumonia, tuberculosis, and maternal death, together with the time variation emanating from medical breakthroughs toward the middle of the twentieth century, this research examines the economic repercussions of positive health shocks within US states. It establishes that these shocks, causing the second US mortality revolution, impacted income differently. Specifically, states with higher levels of flu/pneumonia mortality prior to the onset of the era of big medicine experienced faster per capita GDP growth after its onset, while states with higher mortality rates from tuberculosis faced the opposite development. In addition, the relationship between the maternal-mortality shock and GDP per capita is positive, but it appears not to be statistically robust. Taken together, the findings suggest that the total effect of the mortality revolution on income was close to zero. Moreover, as these causes of death had different age distributions, the evidence here instigates the argument that health's impact on economic growth is contingent upon when in life it changes.

We present a theoretical and empirical analysis of the relationship between U.S. migration experience and earnings in the Mexican labor market. We use our model to analyze the effects of self-selection and endogeneity on OLS estimates of the return to migration experience in the Mexican labor market. Under plausible assumptions, OLS estimates provide a lower bound on the true average return to migration experience among return migrants. Using Mexican Migration Project (MMP) data, we find a return to migration experience of about 2.2 percent per year. Our estimates are robust to the inclusion of proxies for unobserved skill. A comparison with patterns in the 1995 Mexican Population and Dwelling Count suggests that our results are robust across data sets and are driven by a relationship between migration experience and wages, not hours worked. We also explore the plausibility of multiple mechanisms that could explain this relationship. We find the most evidence for the theory that individuals are acquiring occupation-specific work experience in the United States. The return to a year of occupation-specific migration experience is estimated to be as high as 8.7 percent for some occupations.

BACKGROUND: The American Community Survey (ACS) in the United States (US) collects detailed demographic information on the US population. Pressures to use year-to-year population estimates to analyze trends (i.e., between-year differences on the characteristics of interest) have motivated the need to explore how single- or multi-year estimates can be used to investigate changes in US population over time. OBJECTIVE: The specific aim of this manuscript is to provide empirical evidence that between-year differences in population characteristics have difference levels of uncertainty around point-estimates. METHODS: Six ACS Public Use Microdata Sample (PUMS) single year files from 2005 through 2010 are used to empirically show the heterogeneity of uncertainty in between-year differences on level of education, for a birth cohort born between 1960 and 1970 of non-Latino-whites and Mexican Latinos/as. RESULTS: The data show the precision of the education estimate decreases as the specificity of the population increases. For example, Mexicans 99% confidence intervals have wider and more time-varying bandwidths than non-Latino-whites. CONCLUSIONS: Inferring meaningful population change requires the challengeable assumption that between-year differences are not the product of data artifacts. Harvesting reputable ACS data demands further research before between-year differences can be treated as real change.

We link federal contracting data to ACS industry data, provided by IPUMS-USA, to estimate wages and employment supported by federal contracting.

Data publishing is an easy and economic means for data sharing, but the privacy risk is a major concern in data publishing. Privacy preservation is a major task in data sharing for organizations like bureau of statistics, and hospitals. While a large number of data publishing models and methods have been proposed, their utility is of concern when a high privacy requirement is imposed. In this paper, we propose a new framework for privacy preserving data publishing. We cap the belief of an adversary inferring a sensitive value in a published data set to as high as that of an inference based on public knowledge. The semantic meaning is that when an adversary sees a record in a published data set, s/he will have a lower confidence that the record belongs to a victim than not. We design a method integrating sampling and generalization to implement the model. We compare the method with some state-of-the-art methods on privacy-preserving data publishing experimentally, our proposed method provides sound semantic protection of individuals in data and, provides higher data utility.

Breaking new ground in its innovative blend of quantitative and qualitative methods, the book essentially argues that another sort of growth is indeed possible. While offering specific insights for regional leaders and analysts of metropolitan areas, the authors also draw a broader – and quite timely – set of conclusions about how to scale up these efforts to address a U.S. economy still seeking to recover from economic crisis and ameliorate distributional divisions.

This study explored factors that affect access to and use of sexual and reproductive health services including family planning among immigrant Latino men residing in rural Oregon communities that have experienced a high growth in their Latino population. In-depth interviews were conducted with 49 sexually active men aged 18 to 30 years who recently immigrated to the United States. Findings from content analysis identified multiple overlapping individual- level barriers, including lack of knowledge, perception of personal risk for unintended pregnancy and STIs, and fear of disease. On a service delivery level, structural factors and the importance of confianza when interacting with providers and clinic staff were dominant themes.The majority of these themes were grounded in a cultural context and linked to men’s cultural background, beliefs, and experiences. Examining the needs of immigrant Latino men through this cultural lens may be critically important for improving access and use of sexual and reproductive health services.

I use variation in childhood exposure to the Dust Bowl, an environmental shock to health and income, as a natural experiment to explain variation in adult human capital. I find that the Dust Bowl produced significant adverse impacts in later life, especially when exposure was in utero, increasing rates of poverty and disability, and decreasing rates of fertility and college completion. Dependence on agriculture exacerbates these effects, suggesting that the Dust Bowl was most damaging via the destruction of farming livelihoods. This collapse of farm incomes, however, had the positive effect of reducing demand for child farm labor and thus decreasing the opportunity costs of secondary schooling, as evidenced by increases in high school completion amongst the exposed.

This paper examines the effect of early childhood investments on college enrollment and degree completion. We used the random assignment in Project STAR (the Tennessee Student/Teacher Achievement Ratio experiment) to estimate the effect of smaller classes in primary school on college entry, college choice, and degree completion. We improve on existing work in this area with unusually detailed data on college enrollment spells and the previously unexplored outcome of college degree completion. We found that assignment to a small class increases students' probability of attending college by 2.7percentage points, with effects more than twice as large among black students. Among students enrolled in the poorest third of schools, the effect is 7.3percentage points. Smaller classes increased the likelihood of earning a college degree by 1.6percentage points and shifted students toward high-earning fields such as STEM (science, technology, engineering, and mathematics), business, and economics. We found that test-score effects at the time of the experiment were an excellent predictor of long-term improvements in postsecondary outcomes.

Political behavioral research on geographic variation typically employs national surveys and rarely digs below the congressional district level. The data used in such analyses are not suitable for detecting geographic patterns of shared political behavior, or communities of interest, in geographic data. Using large databases containing micro-level voter data, we identify the communities in which different types of voter behavior cluster geographically, without relying on the assumptions associated with survey research. As a motivating example, we examine variation in income-based voting across and within states. Using block-group-level party registration data and precinct-level election returns, we employ a combination of nonparametric and multi-level models to demonstrate that much state-by-state variation in income-based voting is driven by differences in geographic clusters that rarely encompass states and often cross state boundaries.

Recent college graduates have had a difficult time finding employment to match their skills. Indeed, unemployment is now higher among 18 to 24 year olds than any other age group. Despite the current labor market, a college degree can drastically change an individual’s earning power. But what impact does a college education have 40 years later during retirement?

With over 40 percent of U.S. children born out-of-wedlock, understanding how parents navigate their legal relationship contract options is important. This paper examines the consequences of lowering the cost of paternity establishment a contract granting unmarried fathers partial parental rights and responsibilities within a framework where parents trade-off access to children with match quality. I show that the resulting increases in paternity establishment rates are partially driven by reductions in parental marriage rates. Although unmarried fathers become more involved with their children along some dimensions, the net effects on father involvement, family, and child welfare are either negative or zero.

The twentieth century United States provides a natural experiment to measure the strength and persistence of entrepreneurial cultures. Assuming immigrants bear the cultures of their birth place, comparison of revealed entrepreneurial propensities of US immigrant groups in 1910 and 2000 reflected these backgrounds. Two measures of entrepreneurial culture are employed; the first is simply the chance that a member of the migrant group will be an employer and the second is the origin country effect on this probability, conditional upon personal characteristics. The preferred second measure shows persistence of some cultures and change of others over the twentieth century. Among the more stable cultures North-western Europe, where modern economic growth is widely held to have originated, did not host unusually strong entrepreneurial propensities. Instead such cultures were carried by persons originating from Greece, Turkey and Italy, together with Jews.

Adolescents face daily tradeoffs between human capital investment, labor, and leisure. This paper exploits state variation in the repeal of Sunday closing laws to examine the impact of a distinct and plausibly exogenous rise in the quantity of competing diversions available to youth on their educational attainment. The results suggest that the repeals led to a significant decline in both years of education and the probability of high school completion. I explore increased employment and risky behaviors as potential mechanisms. Further, I find a corresponding decline of the repeals on adult wages.

With nearly half of U.S. births occurring out of wedlock, understanding how parents navigate their relationship options is important. This paper examines the consequences of lowering the cost of paternity establishment - a contract granting unmarried fathers partial parental rights and responsibilities - within a framework where parents trade off joint child investment against interaction with each other. My empirical results show that increases in paternity establishment are partially driven by reductions in parental marriage. Although unmarried fathers invest more in their children along some dimensions, the overall effects on father involvement and child welfare are either negative or zero.

During the first few decades of the twentieth century, the number of doctors setting up practice in rural areas dropped dramatically. Many contemporaries attributed this decline to Abraham Flexner's 1910 report and the accompanying reforms in medical education which resulted in many medical schools closing or merging with each other or with major research universities. Others argued that it reflected the falling income of country doctors due to increasing competition brought about by the automobile and decline in the population in rural areas. We use data from the American Medical Directories (AMD) for 1909, 1914, 1918, and 1923 to examine the location decisions of physicians and determine the factors leading to the shift of medical practice out of rural areas. We find that much of the movement out of rural areas was due to changes in the locations recent medical school graduates chose to set up practice. While the rural share of all physicians in the AMD sample fell from 28 percent in 1909 to 20 percent in 1923, the rural share for recent medical school graduates fell from 29 to 8 percent over the same period. We further find that the location choices of new physicians were strongly influenced by the quality of the medical schools they attended. Graduates of schools receiving unfavorable reviews from Flexner were on average 12 percent more likely to locate in rural areas than their peers who graduated from higher quality schools. For a subset of the AMD data, we have information on city or county of birth. These data reveal that the physicians most likely to set up practice in rural areas were those who were born in rural areas. Rural-born doctors were also more likely to attend medical schools Flexner had criticized in his report. As those schools closed or merged with more prestigious institutions, the supply of physicians to rural areas declined.

Regularly releasing the aggregate statistics about data streams in a privacy-preserving way not only serves valuable commercial and social purposes, but also protects the privacy of individuals. This problem has already been studied under differential privacy, but only for the case of a single continuous query that covers the entire time span, e.g., counting the number of tuples seen so far in the stream. However, most real-world applications are window-based, that is, they are interested in the statistical information about stream- ing data within a window, instead of the whole unbound stream. Furthermore, a Data Stream Management System (DSMS) may need to answer numerous correlated aggregated queries simulta- neously, rather than a single one. To cope with these requirements, we study how to release differentially private answers for a set of sliding window aggregate queries. We propose two solutions, each consisting of query sampling and composition. We first selectively sample a subset of representative sliding window queries from the set of all the submitted ones. The representative queries are an- swered by adding Laplace noises in a way satisfying differential privacy. For each non-representative query, we compose its answer from the query results of those representatives. The experimental evaluation shows that our solutions are efficient and effective.

Aims To investigate the direction and degree of potential bias introducedto analyses of drinking and health status which exclude former drinkers from exposure groups. Design: Pooled analysis of 14 waves (1997–2010) of the U.S. National Health Interview Survey (NHIS). Setting General population-based study. Participants 404,462 participants, from 14 waves of the NHIS, who had knownself-reported health status and alcohol consumption status. Measurements Self-reported health status was used as the indicatorof health. Two approaches were used to classify alcohol consumption: (i)separation of former drinkers and current drinkers, and (ii) combined former and current drinkers. The prevalence of fair/ poor health by alcohol use, gender and age with 95% confidence intervals was estimated. The difference in prevalence of fair/ poor health status for lifetime abstainers, former drinkers, current drinkers and drinkers (former drinkers and current drinkers combined) were compared using Poisson regression with robust estimations of variance. Findings Excluding former drinkers from drinker groups exaggerates the difference in health status between abstainers and drinkers, especially for males. Conclusions In cohort study analyses, former drinkers should be assigned to a drinking category based on their previous alcohol consumption patterns and not treated as a discrete exposure group.