Vast amounts of data are now being collected from census and surveys, scientific research, instruments, observation of consumer and internet activities, and sensors of many kinds. These data hold a wealth of information, however there is a risk that personal privacy will not be protected when they are accessed and used. This paper provides an overview of current and emerging approaches to balancing use and analysis of data with confidentiality protection in the research use of data, where the need for privacy protection is widely-recognised. These approaches were generally developed in the context of national statistical agencies and other data custodians releasing social and survey data for research, but are increasingly being adapted in the context of the globalisation of our information society. As examples, the paper contributes to a discussion of some of the issues regarding confidentiality in the service science and big data analytics contexts.

In this paper I document occupational mobility comparing the experiences of cohorts living one century apart: those captured in the Panel Study of Income Dynamics from 1968 to 2000 and a longitudinal census sample of individuals observed between 1880 and 1930. Considering different levels of aggregation with respect to occupations, as wellas different definitions of what an occupation is, I uncover several facts which are remarkably stable in time: (i) occupational mobility is higher for younger workers, (ii) it is closely related to geographical mobility and (iii) over the life-cycle, individuals tend towards occupations with high cognitive, non-routine taskrequirements and away from manual-heavy tasks. A major difference between cohorts is the fate of geographical movers: for the 1880 cohort, interstate migrants were characterized with worse occupational outcomes than stayers, while the opposite is true for the 1968 cohort; additionally, the data shows that routine occupations are more prominent for the 1968 cohort.

This dissertation studies the impact of World War II on the development of the American economy after 1940. Scholars have long-debated the economic consequences of the war, particularly with reference to the macroeconomy and often relying on standard measures of aggregate economic performance. The approach in this dissertation is to study the microeconomic implications of mobilization for World War II. Specifically, the three main chapters address the following questions: What were the human capital costs of the manpower mobilization for young women? Did industrial mobilization promote the growth and diversification of manufacturing in the American South? How much did government spending on supply contracts contribute to migration and the change in the structure of wages between 1940 and 1950? The first chapter provides an overview of America's twentieth century wars and surveys the literature on the impact of World War II. In the second chapter, I find that greater exposure to manpower mobilization decreased young women's educational attainment initially, with important implications for family formation and labor market performance. From the analysis of the third chapter I conclude that the war led to modest reallocation of manufacturing activity toward high value-added sectors, but the war most likely did not create the modern industrial South. In the final chapter I provide evidence that migration induced by World War II played a role in reshaping the structure of wages during the 1940s. Together, the chapters provide important nuance and revisions to our understanding of World War II.

The “stuff” of modern social science is data, but museums are often ill- equipped to allow visitors to appreciate and explore data as an “artifact” of science. Interpreting data visualizations can include both the ability to detect patterns and the ability to hypothesize possible causal mechanisms that would produce such patterns. We are utilizing embodied interaction as a means of augmenting visitors’ personal connection with a data subset in an effort to promote shared curiosity and hypothesis generation as multiple visitors explore the data collaboratively. Here we share initial findings of the impacts of embodiment on visitors’ sense-making around the display.

This dissertation explores how migration responds to economic conditions, particularly differences in responsiveness for various segments of the population. After a brief introduction and motivation of my work in Chapter One, Chapter Two estimates the responsiveness of households' interstate migration to origin state labor market conditions and surrounding state labor market conditions. Each percentage point increase in origin state unemployment insurance claims leads to a 3.2 percent increase in households' propensity to migrate interstate and each percentage point increase in the unemployment insurance claims rate of surrounding states reduces interstate migration propensity by 5.2 percent. I then examine how this responsiveness varies by demographics and how it has changed over time. I determine that the responsiveness of migration to labor market conditions is weaker for several groups at high poverty risk, including less educated, non-employed and rural households and households with children present. I also show that between the early 1980's and mid 1990's labor market conditions became became a smaller factor in household migration decisions, but since then labor market conditions have gained in importance. While Chapter Two examines short-run migration responsiveness, Chapter Three explores the size of the long-run outflow (or inflow) of skilled labor occurring in local areas in response to economic conditions, amenities and other area characteristics. I estimate the extent of this brain gain and brain drain within localities the United States between the early 1990's and late 2000's, describing both absolute changes (percentage growth in the stock of educated individuals) and relative changes )growth in the share of educated individuals). For each of three measures of brain gain estimated, I show substantially more positive flows of educated individuals towards local areas with strong initial economic conditions. I also show that non-metropolitan areas are more likely to experience all three measures of brain drain. I present evidence that nonmetropolitan areas' inability to attract and retain educated individuals stems primarily from labor market disparities including the urban-rural wage differential.

In the U.S., a substantial number of children live with stepparents or adoptive parents. Empirical research has found strong and robust correlations between family structure and child outcomes in modern data, indicating that family structure matters for the welfare of children. Although childrens living arrangements varied widely across households also in historical times, few empirical studies have examined how nonbiological children fared compared to biological children in the past. In this study, I use federal census microdata (IPUMS) in 1880-1930 and 2000 to compare socioeconomic conditions and educational outcomes of adopted, step, and biological children in the U.S. I find that: (1) for both whites and blacks, step households were negatively selected from the population of married two-parent households in 1880-1930; (2) even after controlling for household characteristics, stepchildren had significantly lower educational outcomes than biological children during the same period; (3) although there was little evidence that adoptive households were negatively selected, for both whites and blacks, adopted children were also educationally disadvantaged in the early twentieth century; and (4) the conditions of adoptive and step households relative to biological households have improved substantially by 2000, reducing or reversing the earlier educational disadvantages.

Background Regular short and long sleep durations are associated with increased mortality and morbidity. While previous research shows significant sleep disparities between people with and without disabilities, less is known about the association between different types of disability and high-risk sleep using nationally representative data. Objective We examine the association between short and long sleep durations and having a work disability or an impairment in sensory, cognitive, or physical functioning among a nationally representative sample of working-age adults in the United States. Methods We estimate multinomial logistic regression models using data from the 20032012 American Time Use Survey to identify how different types of disabling conditions net of other sociodemographic factors relate to the likelihood of reporting short (6 h or fewer) or long (9 h or more) sleep, versus mid-range (between 6 and 9 h) sleep. Results For respondents with work disabilities versus those without work disabilities, the relative risk of short and long sleep is 1.4 and 1.5 times (respectively) that of those with mid-range sleep. The risk of short and long sleep durations is also higher among respondents with cognitive, physical, or multiple impairments. Conclusions Individuals with disabilities are less likely than those without disabilities to have optimal sleep durations. These results demonstrate the importance of health promotion services among this population, with specific attention to sleep hygiene interventions.

Analyses of infant and child mortality (ICM) show traditionally a mortality gap between sexes that favors females in Latin America, being magnified to the extent that higher levels of life expectancy are modeled. However, evidence found in Brazil (Wong et al, 2013) shows a different trend, where infant male mortality is lower among girls than boys at early ages. The objective of this study is to evaluate whether the pattern found for Brazil also prevails in other Latin American countries, since they also have shown a steady decline in the ICM levels. Our hypothesis is that in countries where there has recently been a significant reduction in ICM, the gender mortality gap has also reduced. The analysis is based on data from censuses and DHS for recent years, following two stages: (a): to show the trend of ICM from the proportion of surviving children by age of mother. (b): to apply the Brass indirect technique to estimate the MI level by sex, based on Coale-Demeny Model Tables (1966). The findings for the Latin American countries corroborate, to some extent, observed Brazilian trend. The male to female mortality ratio (MFMR) are smaller than those defined by model tables which are still used to define current ICM levels in the Region.

ncome segregation produces unequal social outcomes and has steadily increased since the 1970s. High-poverty neighborhoods suffer from low performing schools, fewer jobs, an evaporation of local role models (Wilson 1987; Reardon and Bischoff 2011a). Recent evidence suggests growing income inequality influences the segregation of affluence more than the segregation of poverty (Reardon and Bischoff 2011b). Metropolitan areas that display strong population and economic growth are susceptible to higher levels of income inequality. I use three unique quantitative approaches to measure the segregation of affluence and poverty in a comparison of four metropolitan areas exhibiting strong growth to four metros with weaker growth. I find the increase in income segregation between 1990 and 2010 is attributable to the increase in the segregation of affluence. Weaker metropolitan areas exhibit higher levels of income segregation than strong metros due to their significantly higher levels of segregation of poverty; however, strong metros exhibit higher levels of segregation of affluence.

Exposure has amplified rapidly over the past half century and is one of the primary drivers of increases in disaster frequency and consequences. Previous research on exposure change detection has proven limited since the geographic units of aggregation for decennial censuses, the sole measure of accurate historical population and housing counts, vary from one census to the next. To address this shortcoming, this research produces a set of gridded population and housing data for the Chicago, Illinois, region to evaluate the concept of the expanding bulls-eye effect. This effect argues that targetspeople and their built environmentsof geophysical hazards are enlarging as populations grow and spread. A collection of observationally derived synthetic violent tornadoes are transposed across fine-geographic-scale population and housing unit grids at different time stamps to appraise the concept. Results reveal that intensifying and expanding development is placing more people and their possessions in the potential path of tornadoes, increasing the likelihood of tornado disasters. The research demonstrates how different development morphologies lead to varying exposure rates that contribute to the unevenness of potential weather-related disasters across the landscape. In addition, the investigation appraises the viability of using a gridded framework for assessing changes in census-derived exposure data. The creation of uniformly sized grid data on a scale smaller than counties, municipalities, and conventional census geographic units addresses two of the most critical problems assessing historical changes in disaster frequencies and magnitudeshighly variable spatial units of exposure data and the mismatch between spatial scales of population/housing data and hazards.

The implications of immigration legislation for U.S. agriculture have been a subject of interest for scholars and policymakers for several decades. The U.S. farm industry spends an average of 17% of its total variable production costs on hired labor. Certain farm sectors are even more labor-intensive such as vegetable, nursery, and fruit farms with hired labor expense shares of 35%, 46%, and 48%, respectively (Zahniser et al., 2012). The National Agricultural Workers Survey from the U.S. Department of Labor indicates that, over the last 15 years, about half of agricultural crop farm workers have been undocumented (Carroll, Georges and Saltz, 2011). Most efforts to regulate immigration in the United States have been at the federal level. However, since 2003, the nation has experienced a surge of enforcement efforts at the state and sub-state levels. Pham and Van (2010) discuss the legal and economic significance of such decentralization. For agriculture, this is unchartered territory; previous enforcement efforts at the federal level have typically employed a balanced approach, whereby measures that reduce labor were counterbalanced by measures that provide agriculture special consideration. For example, the 1986 Immigration Reform and Control Act (IRCA) imposed new sanctions on employers who knowingly hired undocumented workers and approved steps to provide legal status to around 3 million illegal workers. Due to the high reliance of agriculture on immigrant workers, IRCA also took measures designed to give special consideration for farm labor. It revised the H-2 guest worker program, which provides work visas to a limited number of immigrants (66,000 in 2013), to establish the H-2A agricultural guest worker program with no numerical limits. States and counties have newfound authority on enforcement, but little control over other tools for regulating immigration (e.g., the H2A-program), making it harder for them to follow a balanced approach.

Linking multiple databases to create longitudinal data is an important research problem with multiple applications. Longitudinal data allows analysts to perform studies that would be unfeasible otherwise. We have linked historical census databases to create longitudinal data that allow tracking people over time. These longitudinal data have already been used by social scientists and historians to investigate historical trends and to address questions about society, history and economy, and this comparative, systematic research would not be possible without the linked data. The goal of the linking is to identify the same person in multiple census collections. Data imprecision in historical census data and the lack of unique personal identifiers make this task a challenging one. In this paper we design and employ a record linkage system that incorporates a supervised learning module for classifying pairs of records as matches and non-matches. We show that our system performs large scale linkage producing high quality links and generating sufficient longitudinal data to allow meaningful social science studies. We demonstrate the impact of the longitudinal data through a study of the economic changes in 19th century Canada.

Interstate migration has decreased steadily since the 1980s. We show that this trend is not primarily related to demographic and socioeconomic factors, but instead appears to be connected to a concurrent secular decline in labor market transitions. We explore a number of reasons for the declines in geographic and labor market transitions, and find the strongest support for explanations related to a decrease in the net benefit to changing employers. Our preferred interpretation is that the distribution of relevant outside offers has shifted in a way that has made labor market transitions, and thus geographic transitions, less desirable to workers.

The great recession hit young workers hard, leaving roughly five million young adults unemployed five years after the downturn officially ended. However, nearly 50 million Millennials (young adults aged 18 to 34) are currently working across the nation, comprising a third of the workforce today. Eventually, that will grow to half by 2020. Where young people work and how their earnings have fared since the economic downturn have enormous implications for our economys future. In some states, most Millennials work in sectors with real wage growth, such as healthcare, while other common jobs in the service industry fail to provide reliable income that our generation desperately needs. Below is a snapshot of the present economic landscape for Americas young working adults, illustrating where our generation works by sector and geography as well as how the most popular sectors have fared since the economic downturn.

Good morning, Chairman Robert Menendez (D-NJ), Ranking Member Jim Moran(D-VA), and members of the committee. My name is Julia Gordon, and I direct the Housing Finance team at the Center for American Progress, a nonpartisan think tank dedicated to improving the lives of Americans through progressive ideas and action. Thank you so much for convening this hearing on the critical topic of inequality and opportunity in the housing market. I greatly appreciate the opportunity to testify today about the state of our housing recovery and its relationship to the well-being of families and the broader economy.

Is the high degree of gender inequality in developing countriesin education, personal autonomy, and moreexplained by underdevelopment itself? Or do the societies that are poor today hold certain cultural views that lead to gender inequality? This article discusses several mechanisms through which, as countries grow, gender gaps narrow. I argue that while much of the GDP/gender-inequality relationship can be explained by the process of development, society-specific factors are also at play: Many countries that are poor today have cultural norms that exacerbate favoritism toward males. Norms such as patrilocality and concern for women's "purity" help explain the male-skewed sex ratio in India and China and low female employment in India, the Middle East, and North Africa, for example. I also discuss why the sex ratio has become more male-skewed with development. Finally, I lay out some policy approaches to address gender inequality.

Since 2000, the American suburb has emerged as a principal destination for new immigrants to the United States, both documented and undocumented. Whereas some suburban communities have responded to perceived undocumented immigrants with hostility in the form of exclusionary local immigration policies, others have introduced policies designed to welcome immigrants independent of federal legal status. In this article, I employ a qualitative comparative case study analysis of four local immigration policies in the Chicago and Washington DC metropolitan areas to explain how suburbs justify their policy positions. I find that these suburban communities relied on conceptions of American identity and the American Dream in support of their policies, but leveraged these tropes in vastly different ways depending on the broader strategic purposes of the policies. These divergent suburban immigration policies both challenge traditional notions of suburban political and cultural homogeneity and reveal how such heterogeneity has produced a distinct unevenness in contemporary local policy responses to undocumented immigration within metropolitan regions.

The field of education has been and will continue to be essential to the survival and sustainability of the Black community. Unfortunately, over the past five decades, two major trends have become clearly evident in the Black community: (a) the decline of the academic achievement levels of Black students: and (b) the disappearance of Black teachers, particularly Black States was teaching (Ruggles, Alexander, Genadek, et al., 2010). Approximately 36% of the working Black male population was standing in front of the classroom delivering instruction to young Black men and women. Today, only 8% are Black teachers, and approximately 2% of these teachers are Black males (NCES, 2010). . .