We explore the first period of decline in infant mortality in the U.S. and provide estimates of the independent and combined effects of clean water and effective sewerage systems on infant mortality. Our case is Massachusetts, 1880-1915, when state authorities developed a sewerage and water district for municipalities in the Boston Greater Metropolitan area. We find that the two interventions were complementary and together accounted for approximately 37 percent of the total decline in log infant mortality among treated municipalities during the 36 years considered. Considerable research has documented the importance of clean water interventions for improvement in population health, but there is less evidence on the importance of sewerage systems. Our findings are directly relevant to urbanization in the developing world and suggest that a dual-pronged approach of safe water and sewerage is important to improving infant and early child survival.

This dissertation examines the relationship between interest group access and American policymaking in three papers. Paper 1 suggests a key mechanism for interest group influence is in the opportunity for campaign contributors to frame information and dominate the belief diffusion process in congressional committees. The paper proposes hypotheses around the impact of interest group access on belief diffusion in a committee-level social network. It then employs social network analysis and Correlated Topic Modeling to determine whether members of Congress with the same interest group donors develop statistically similar cognitive maps with respect to a complex policy issue.

My research will use the Minnesota Population Center's Integrated Demographic and Health Series (IDHS) data to examine the characteristics of women who do not know about contraceptives in seven Sub-Saharan African countries. Preliminary research shows great variation in contraceptive knowledge across African countries: 35.34% of married women in Nigeria had heard of no contraceptive method (in 2008), compared to only 0.60% in Zimbabwe (in 2010-11). It also reveals a lack of contraceptive knowledge among married women who wish to limit their fertility, most notably in Nigeria and Mali. In Nigeria in 2008, 20.66% of married women who did not want another child had never heard of any contraceptive method, which suggests an unmet need for family planning. I will explore the characteristics of these women, who may have a "latent demand" for contraceptives rather than a demonstrably active demand (Freedman 1997).

BACKGROUND Empirical research describes retirement migration to Mexico as a viable option for some older Americans. However, far less research examines this phenomenon among Mexican immigrants in the United States. The literature that does address this topic treats international migration as a singular occurrence and does not examine the possibility of return and subsequent reentry between countries. This omission creates an important gap in our knowledge of international retirement migration, considering the strong transnational ties that Mexican immigrants maintain to their home and destination countries. OBJECTIVE Using a multistate life table approach, this study examines the rate of return to Mexico and reentry back into the United States among Mexican males aged 50 and older with U.S. migration experience, as well as the number of years spent in both countries. RESULTS Results show that the rate of reentry from Mexico into the United States declined from 3.33% at ages 5054 to less than 1% at age 70 and older (p<0.05). By contrast, the rate of return to Mexico from the United States increased from 3.19% at ages 50 to 54 to 4.44% at ages 65 to 69 and dropped to less than 2% at age 70 and older (p<0.05). CONCLUSIONS Although rates of return and reentry among this population are relatively low, they provide insight into the potential life course factors driving the migration patterns of a population of increasing size and relevance in the United States.

A model based on proximity breach for numerical sensitive attributes is proposed.At first，it divides numerical sensitive value into several intervals on the premise of protecting the internal relations between quasi-identifier attributes and numerical sensitive attributes.Secondly，it proposes a (k，ε)-proximity privacy preserving principle to defense proximity privacy.In the end，a maximal neighborhood first algorithm (MNF) is designed to realize the (k，ε)-proximity.The experiment results show that the proposed model can preserve privacy of sensitive data well meanwhile it can also keep a high data utility and protect the internal relations.

This paper examines unionism’s relationship to the size of the middle class and its relationship to intergenerational mobility. We use the PSID 1985 and 2011 files to examine the change in the share of workers in a middle-income group (defined by persons having incomes within 50% of the median) and use a shift-share decomposition to explore how the decline of unionism contributes to the shrinking middle class. We also use the files to investigate the correlation between parents’ union status and the incomes of their children. Additionally, we use federal income tax data to examine the geographical correlation between union density and intergenerational mobility. We find: 1) union workers are disproportionately in the middle-income group or above, and some reach middle-income status due to the union wage premium; 2) the offspring of union parents have higher incomes than the offspring of otherwise comparable non-union parents, especially when the parents are low-skilled; 3) offspring from communities with higher union density have higher average incomes relative to their parents compared to offspring from communities with lower union density. These findings show a strong, though not necessarily causal, link between unions, the middle class, and intergenerational mobility.

In much of the literature focusing on the growth and structure of the urban system, the difference between contagious and hierarchical interrelations across cities comprised in the urban system are obfuscated. In this paper, we clearly distinguish and quantify the effects of both. In other words, we focus on how the structure of the urban system influences population growth by using central place theory as a theoretical basis for addressing the research question: what natural and man-made locational characteristics influence population growth? We make three major contributions to the existing literature. First, we utilize a unique dataset of urban areas with decennial observations from 1990 to 2010 which captures the agglomerated economic activity and built extent of urban locations with at least 2,500 inhabitants, to include all but the smallest rural communities. Second, our analysis includes both the hierarchical relationship among cities of differing sizes and the continuous nature of proximity to other cities. The novel use of a spatially-lagged hierarchical linear model allows us to include both these critical aspects of the urban system in our analysis. Third, we include man-made amenities and characteristics of cities, which have been omitted from previous studies in an effort to avoid endogeneity in the analysis. By focusing on the intercept and lagged population variables in the urban area equation, we use this model to empirically explore the debate on whether there is random or deterministic growth in the distribution of cities in the United States.

Decades of social science research demonstrate the impact of education on civic participation. However, prior scholarship generally assumes that the returns to education do not change over time or geographic space. Absolute educational effects are not always plausible; as more individuals obtain a college degree, there are a greater number of qualified individuals competing for the same social resources. The present study tests the impact of education on civic participation, and whether this effect changes as the number of college-educated individuals increases over birth cohorts. The present findings suggest that in some cases, a college degrees impact on civic participation declines as more individuals obtain college degrees. The findings challenge commonly held assumptions about the effect of higher education on civic participation. Furthermore, the findings suggest that while sending an individual to college may increase civic participation, encouraging all high school students to go to college may undercut the benefits of college attendance.

With the ARX data anonymization tool structured biomedical data can be de-identified using syntactic privacy models, such as k-anonymity. Data is transformed with two methods: (a) generalization of attribute values, followed by (b) suppression of data records. The former method results in data that is well suited for analyses by epidemiologists, while the latter method significantly reduces loss of information. Our tool uses an optimal anonymization algorithm that maximizes output utility according to a given measure. To achieve scalability, existing optimal anonymization algorithms exclude parts of the search space by predicting the outcome of data transformations regarding privacy and utility without explicitly applying them to the input dataset. These optimizations cannot be used if data is transformed with generalization and suppression. As optimal data utility and scalability are important for anonymizing biomedical data, we had to develop a novel method.

Research on intergenerational mobility has often treated outcomes such as schooling and earnings as being imperfectly transmitted from one parent to a child. But because the characteristics of both parents are important in shaping children's outcomes, the way in which a generation of parents is sorted into couples is likely to be a key determinant of intergenerational persistence. Mating patterns are assortative-that is, individuals tend to partner with people similar to themselves-and this is typically measured by similarity of educational attainment. I present the first estimates of the effect of assortative mating on intergenerational persistence of schooling and earnings. I measure assortative mating as the rank correlation of cou-ples' educational attainment, that is, the degree to which the most highly-educated men partner with the most highly-educated women. Using data on parents and children in the United States, I find that assortative mating explains about one quarter of the observed intergenerational persistence of schooling and earnings.

We examine theoretically and empirically how wage subsidies and minimum wages interact regarding to employment opportunities, and how these interactive effects vary across different groups of workers. Assuming a neoclassical labor market and heterogeneous work, subsidized low-skilled worker displace less-skilled workers. The effect on non-subsidized low-skilled workers remains theoretically open. The empirical examination for the US shows that increasing minimum wages decreases less-skilled employment, haven’t an effect on non-subsidized low-skilled employment and induce a hump-shape of subsidized low-skilled employment. For Germany, however, several simulation studies indicate that the provided minimum wage and existing wage subsidies do not cause substitution effects, but lower employment of all workers in the low wage sector regardless of a grant funding.

Immigrants have been found to promote exports to their countries of origin. In this study, we show that immigrants can also increase exports to other countries–specifically, countries that we term “proximate" to the country of origin. The proximity measure for a country pair is based on a geographical factor of sharing a common border and a cultural/linguistic factor of the probability of sharing a common native language; these factors reflect business networks, foreign market information and communication facilitation as channels of trade facilitation. The expected number of non-proximate immigrants is termed “distant" and is found to have a negative effect on exports to a given country. The trade effects of different immigrant groups are more pronounced among state-country pairs with a larger number of immigrants from the export destination country. We obtain state-country level estimates of immigration-export elasticity in the United States that are preferable to previously available estimates due to the use of proper trading-pair fixed effects and more recent data, covering 2003-2013. Analysis is done both at the aggregate state-country level and at the industry level, with the within- industry immigrant effect on industry exports being estimated for the first time for the case of the United States as the host country.

This dissertation consists of two original empirical studies on the determinants of teenage childbearing in the United States. The first study examines the impact of educational attainment on teenage childbearing, using school entry laws as an instrument for education and a highly detailed North Carolina administrative dataset that links birth certificate data to school administrative records. I show that being born after the school entry cutoff date affects educational success in offsetting ways, with a negative impact on years of education but positive impact on test scores. Using an IV regression strategy to distinguish the impacts of years of education and test scores, I show that both educational measures have negative impacts on teenage childbearing. The second study examines potential causes of the decline in the U.S. teenage birth rate between 1991 and 2010. Using age-period-cohort models with Vital Statistics birth data and Census population counts, I show that the decline was driven by period changes in the early 1990s but by cohort changes between the mid-1990s and mid-2000s. I also use a difference-in-differences model to investigate the extent to which social policies in the 1970s-1980s can explain these cohort changes. The evidence suggests that while legalization of abortion for adult women and unilateral divorce laws had a significant impact on teenage birth rates in the 1990s-2000s, abortion legalization is unlikely to be a major explanation for the observed decline.

This paper provides a detailed empirical assessment of the evolution of income inequality and the redistributive effects of tax and transfer policies following the 2007-2008 crisis. It focuses on the US case, drawing on data from the Current Population Survey for the period 2007-2013. Contrary to most existing studies, it looks at several parts of the income distribution, giving special emphasis to the bottom and middle income groups. Furthermore, it performs a detailed analysis of the effects of several redistributive mechanisms, as opposed to focusing solely on the overall effects of transfers and taxes. Finally, it provides estimates covering the recovery period while most of the literature so far has mainly focused on the Great Recession years. Results show that although the crisis implied income losses across the whole income distribution, the burden was disproportionately born by low to middle income groups. Income losses experienced by richer households were relatively modest and transitory, while those experienced by poorer households were not only strong but highly persistent. Tax and transfer policies had a crucial role in taming an increase in income inequality in the immediate aftermath of the crisis, and during the GR years. Since 2010, however, these policies have been weakened and income inequality has experienced a new surge. The findings of this paper contribute to a better understanding of the distributional consequences of aggregate crisis and the role of tax and transfer policies in stabilising the income distribution in a crisis aftermath.

Many census occupational classification systems have been developed over the last 150 years. Availability of digital census data sets now means such classifications can be systematically analyzed. Examination of heavy industrial workers in the full count U.S. 1880 census, and other censuses, has revealed major problems in the attribution of occupations to industrial sectors. This is traceable to the original enumeration process, and it particularly affects generic tradesmen such as blacksmiths and carpenters, who worked in numerous industrial sectors. As a result, the imputation of industrial sector codes from recorded occupations by the North Atlantic Population Project (NAPP) is substantially in error, suggesting that re-coding of existing census records using non-census sources would be necessary for such industrial sector codes to have empirical validity. A new occupational/industrial coding system, incorporating the NAPP-modified HISCO scheme, is presented. This system is capable of supporting both future re-coding work, in a structured data warehouse environment, and the systematic coding of occupational data from a range of archival sources such as company records and city directories.

The 1986 Immigration Reform and Control Act was supposed to encourage a smaller, legal agricultural workforce. However, it actually led to a larger labor force, increased by newimmigrant SAWs and unabated illegal immigration. Previous papers have failed to econometrically identify the drop in wages and increase in labor supply reported by farmers and workers, leading to a discrepancy between the economic literature and other fields literatures on the IRCA. Using a linear spline approach between two regions in California, I find that employment of agricultural workers in the SAW-heavy Los Angeles regions increased relative to that in the North Coast region due to a relative increase in farm labor contractor employment, suggesting an increase in labor supply caused by the IRCA. I also use difference-in-differencein-differences equations to show that wages decreased post-IRCA, particularly for laborintensive industries and for Mexican male immigrants.

The standard portrayals of economic life for ordinary Americans and their families paint a picture of stagnancy, even decline, amidst rising income inequality or joblessness. But rarely does the public conversation about the changing economic fortunes of Americans and their families look at questions of family structure. This is an important oversight because, as this report shows, changes in family formation and stability are central to the changing economic landscape of American families, to the declining economic status of men, and to worries about the health of the American dream.

Prior research has demonstrated the importance of physical activities for health and understanding how estimates differ by measurement strategy, including self-reported measures of frequency and intensity, time diary data, and accelerometer data, is thus an important contribution to the limited research base about physical activity measurement. Using nationally representative data from the 2003-2006 National Health and Nutrition Examination Survey (NHANES), the 2003-2013 National Health Interview Survey (NHIS), and the 2003-2013 American Time Use Survey (ATUS), we propose to investigate the following questions. First, how do recall estimates of time spent in physical activity in the ATUS and NHIS compare to NHANES estimates based on accelerometer data? Second, how do ATUS estimates of time spent in activities above different metabolic equivalent (MET) value thresholds compare to NHANES estimates based on accelerometer data? And finally, to what extent do the measurement differences in these estimates vary by BMI and self-reported health?

+ Older workers typically experience longer bouts of unemployment than their younger counterparts. + Once among the ranks of the long-term unemployed, the chances a worker will return to work have been slim. + Prolonged unemployment has obvious consequences for current financial well-being, but its impact can extend well into the retirement years.