When one of the authors of this article began his professional journey in 1968, personal computers were nowhere in sight, a 10-key adding machine was the standard, and accountants literally hand-posted transactions to accounting journals. When someone told you to copy something, it meant writing it out longhand or using a typewriter. Analyzing government financial information was perilous at best, given the inability to readily correlate data and with limited assurance about its reliability and completeness. Audits seemingly took forever and were 100 percent labor intensive.

Tuple-independent and disjoint-independent probabilistic databases (TI-and DI-PDBs) represent uncertain data in a factorized form as a product of independent random variables that represent either tuples (TI-PDBs) or sets of tuples (DI-PDBs). When the user submits a query, the database derives the marginal probabilities of each output-tuple, exploiting the underlying assumptions of statistical independence. While query processing in TI-and DI-PDBs has been studied extensively, limited research has been dedicated to the problems of updating or deriving the parameters from observations of query results. Addressing this problem is the main focus of this paper. We first introduce Beta Probabilistic Databases (B-PDBs), a generalization of TI-PDBs designed to support both (i) belief updating and (ii) parameter learning in a principled and scalable way. The key idea of B-PDBs is to treat each parameter as a latent, Beta-distributed random variable. We show how this simple expedient enables both belief updating and parameter learning in a principled way, without imposing any burden on regular query processing. Building on B-PDBs, we then introduce Dirichlet Probabilistic Databases (D-PDBs), a generalization of DI-PDBs with similar properties. We provide the following key contributions for both Band D-PDBs: (i) we study the complexity of performing Bayesian belief updates and devise efficient algorithms for certain tractable classes of queries; (ii) we propose a soft-EM algorithm for computing maximum-likelihood estimates of the parameters; (iii) we present an algorithm for efficiently computing conditional probabilities, allowing us to efficiently implement Band D-PDBs via a standard relational engine; and (iv) we support our conclusions with extensive experimental results.

We use 1940 Census data to study the intergenerational transmission of human capital for children born in the 1920s and educated during an era of expanding but unequally distributed public school resources. Looking at the gains in educational attainment between parents and children, we document lower average mobility rates for blacks than whites, but wide variation across states and counties for both races. We show that schooling choices of white children were highly responsive to the quality of local schools, with bigger effects for the children of less-educated parents. We then narrow our focus to black families in the South, where state-wide minimum teacher salary laws created sharp differences in teacher wages between adjacent counties. These differences had large impacts on schooling attainment, suggesting an important causal role for school quality in mediating upward mobility.

Background: Federal surveys could play a role in measuring the association of rental assistance and health and in identifying the health needs of the assisted population. However, self-reports of rental assistance could be biased. Our objective was to assess the accuracy of reported rental assistance in the National Health Interview Survey (NHIS). Methods: We conducted a record-check study of reports of US Department of Housing and Urban Development rental assistance in the 2004–2012 NHIS, using survey responses linked to administrative records. Misclassification measures were limited to the false-negative rate because the survey ascertained participation in all rental assistance programs, but the administrative data pertained only to US Department of Housing and Urban Development. False-negative rates were calculated for the total population, for sociodemographic subgroups, across levels of self-reported health status, and for specific assistance types (Housing Choice Vouchers, Public Housing, and Multifamily Housing). Results: We estimated a false-negative rate of 22.6%. Misclassification was higher among Public Housing residents compared to those receiving other forms of assistance, even after controlling for sociodemographics. Rates varied across region and other demographics. Those self-reporting fair or poor health were less likely to misreport assistance compared with those in better health, but the difference was explained by covariates. Misreporting assistance had little independent impact on the adjusted association of assistance and health. Conclusions: False-negative reporting of rental assistance is moderately high in the NHIS, but we did not find evidence that it independently biased estimates of the association of health and rental assistance.

Using data from the 2000 U.S. census and the 2010 to 2014 waves of the American Community Survey, we examine the importance of occupational niches in explaining earnings disparities between U.S.-born blacks and black immigrants in the United States. Our results show that, relative to U.S.-born blacks, most black immigrant subgroups have similar or greater representation in occupational niches. Employment in a niche occupation has a small but positive association with earnings, and the returns to niche employment are greater for black immigrants, particularly black immigrant women. Niche employment does not, however, explain earnings disparities between U.S.-born and immigrant blacks.

A new sample containing rental price and characteristic data for over 15,000 New York City units was collected from historical newspapers for the period 1880 to 1910. These units were geolocated to the historical map of Manhattan Island to explore their geographic coverage, using Geographic Information System (GIS) software. This paper presents the new sample and discusses its representativeness of the New York City housing market during the sample period, with reference to the (limited) previous measures available in the literature and an analysis of the summary statistics of various subsamples of the data which can highlight selection biases. Finally, an analysis of the social status and ethnic composition of individuals located in the sample units in Census year 1880 is presented. Understanding the biases that might be present in this new sample will inform its usefulness in uncovering the workings of historical housing markets and in contributing to the scarce available information on historical housing costs.

How does corporate innovation investment affect employee departures to entrepreneurship (spawning)? Research and development (R&D) investment may generate growth options for the firm or make it a more interesting workplace, which could decrease spawning. Conversely, R&D investment could increase spawning if employees can appropriate some of the new growth options, or if engaging with the R&D process makes them more entrepreneurial. Using U.S. employer-employee matched Census data, we show that R&D investment increases spawning. We identify the causal effect of R&D with changes in federal and state tax incentives. The effect is driven by high-tech parents and by departures to high-growth and venture capital-backed entrepreneurship. Intellectual rather than human capital seems to explain the spawning (i.e., new ideas rather than skills). The effect does not impose observable costs on the parent, leading us to conclude that entrepreneurial spawning is a source of knowledge spillovers from corporate R&D.

Insurers have begun to submit their rate request for 2019 for the Health Insurance Marketplaces, which were created as part of the Patient Protection and Affordable Care Act (ACA). Insurers in Virginia and Maryland are requesting rate increases of as much as 90%, prompting concern that this linchpin of the ACA will not survive. In May 2018, Al Redmer Jr, insurance commissioner of Maryland, stated that “we’ve been in a death spiral for a year or 2.”1 These comments echo a statement from Tennessee insurance commissioner Julie McPeak, who in 2016 expressed concern that the Health Insurance Marketplace was “very near collapse.”

BACKGROUND The use of personal health care management (PHM) is increasing rapidly within the United States because of implementation of health technology across the health care continuum and increased regulatory requirements for health care providers and organizations promoting the use of PHM, particularly the use of text messaging (short message service), Web-based scheduling, and Web-based requests for prescription renewals. Limited research has been conducted comparing PHM use across groups based on chronic conditions. OBJECTIVE This study aimed to describe the overall utilization of PHM and compare individual characteristics associated with PHM in groups with no reported chronic conditions, with 1 chronic condition, and with 2 or more such conditions. METHODS Datasets drawn from the National Health Interview Survey were analyzed using multiple logistic regression to determine the level of PHM use in relation to demographic, socioeconomic, or health-related factors. Data from 47,814 individuals were analyzed using logistic regression. RESULTS Approximately 12.19% (5737/47,814) of respondents reported using PHM, but higher rates of use were reported by individuals with higher levels of education and income. The overall rate of PHM remained stable between 2009 and 2014, despite increased focus on the promotion of patient engagement initiatives. Demographic factors predictive of PHM use included people who were younger, non-Hispanic, and who lived in the western region of the United States. There were also differences in PHM use based on socioeconomic factors. Respondents with college-level education were over 2.5 times more likely to use PHM than respondents without college-level education. Health-related factors were also predictive of PHM use. Individuals with health insurance and a usual place for health care were more likely to use PHM than individuals with no health insurance and no usual place for health care. Individuals reporting a single chronic condition or multiple chronic conditions reported slightly higher levels of PHM use than individuals reporting no chronic conditions. Individuals with no chronic conditions who did not experience barriers to accessing health care were more likely to use PHM than individuals with 1 or more chronic conditions. CONCLUSIONS The findings of this study illustrated the disparities in PHM use based on the number of chronic conditions and that multiple factors influence the use of PHM, including economics and education. These findings provide evidence of the challenge associated with engaging patients using electronic health information as the health care industry continues to evolve.

This paper studies the effect of bank branch closures on lending in the Digital Age. Previous studies found that bank branch closings have either no effect or small negative impact on small business lending especially in the low-to-moderate income (LMI) communities. However, decisions to close bank branches are not random and are likely related to local business conditions. As a solution to this endogeneity, and building on work by Nguyen (2018), I use mergers between two large banks with no less than $10B in pre-merger assets with overlapping branching networks as a source of exogenous variation. Using instrumental variable methods, I find that branch closures significantly reduce small business lending (22% reduction in total volume) even in areas with alternative local branches. The total volume of mortgage lending is unaffected. These findings contribute to the debate about the importance of relationship lending, suggesting that physical branches will continue to play an important role mitigating informational asymmetries inherently present in small business lending.

Louisiana is at a pivotal crossroads. Although the economy is improving and producing a lower unemployment rate than the state has seen in years, not everyone is benefitting from this comeback. Workers of color continue to be significantly overrepresented among both the unemployed, and the underemployed. These inequities for workers of color persist regardless of educational attainment. Black men in particular have the highest unemployment rate of any other group, with more than one in 10 Black male workers unable to find work. Without increased access to opportunity, these men are unable to support their families or contribute to the local economy. Employment equity is essential to creating economic prosperity for all Louisianans. Achieving employment equity would mean that everybody in Louisiana who wants to work can find a good job that pays family-supporting wages. It would also mean that Black men and other workers of color are not . . .

This dissertation contains three self-contained chapters on human capital and altruism. The first two chapters explore why women used to lag behind but now exceed men in college enrollment. Chapter 1 shows that examining occupations that require only a high school degree ("non-college" occupations) can help resolve two puzzles. First, why do women attend college at greater rates than men today, when men work more and earn more than women? I document that non-college occupations for men are both more plentiful and higher paying than those for women. Next, I link the occupational inequality in the non-college labor market to the gap in college enrollment, by employing two empirical exercises to show that non-college jobs dramatically affect college-going decisions. Using employment changes in the oil and gas industry, I demonstrate that increases in men's non-college job opportunities lead male high school graduates to forego college enrollment. Using the automation of the office, I demonstrate that declines in the non-college employment opportunities of women lead female college enrollment to grow over time. Thus, women's lower non-college job prospects contribute to their higher college enrollment. This leads to the second puzzle: why did women initially attend college at lower rates than men, when women have always had worse non-college job . . .

The razor thin margins separaƟng the DemocraƟc and Republican candidates for senator and governor in Florida; in the Georgia governor’s race; in the senatorial elecƟon in Arizona; and the strong showing by Beto O’Rourke in the senatorial race in Texas, mask the extraordinary problems faced by the DemocraƟc Party as it turns to the presidenƟal elecƟon scheduled for November 2020.1 In each of these state elecƟons DemocraƟc‐leaning demographic consƟtuencies – African Americans, LaƟnos, and young voters between 18 and 29 years of age – voted at significantly lower rates than the principal Republican consƟtuencies, older non‐Hispanic white men and women. If DemocraƟc demographic consƟtuencies would have voted at even slightly higher rates, elecƟons in these four states each DemocraƟc candidate would have won by comfortable margins. Journalists and poliƟcal commentators have noted over and again that the problem Democrats will face in the 2020 presidenƟal elecƟon is how they can nominate a candidate who will aƩract the white working‐class vote and/or ‘peel off’ some of the Republican support in predominantly white rural areas of key swing states. This theory for victory may be somewhat accurate . . .

“Like The People”: Three Essays on Race, Gender, and Political Participation This dissertation presents three essays, all written with the goal of better understanding mass political behavior in American politics, with a specific focus on experimental approaches and the role of race and gender in political outcomes. The first essay presents the results of a field experiment testing the marginal effects of direct mail on individual vote choice. I fail to find effects of these mailers on vote choice or turnout. The second essay presents the results of the first field experiment to measure the effects of candidate training sessions on political ambition. I find that pre-existing attitudes largely drive the decision to participate in a candidate training session, and that participants started with significantly higher levels of political ambition. The third essay presents a new analysis of a unique dataset of voter registration during and after the Voting Rights Act. I find some evidence of the power threat hypothesis, and a complex interaction between black political agency, white backlash, and institutional barriers.

This report summarizes data from a national survey on the experiences, needs and potential solutions for mHealth technology by people with physical, cognitive, sensory and emotional disabilities. Convenience sampling was used to draw a sample of 377 adults with disabilities. Data were collected from February to August 2017. The survey was conducted by the Rehabilitation Engineering Research Center for Community Living, Health and Function (LiveWell RERC). The survey instrument includes items on user experiences and needs for a wide range of mHealth solutions. This paper focuses on mHealth apps: 1) types of health/wellness mobile apps currently used by people with disabilities; 2) satisfaction levels with the use of health/wellness apps; 3) ease/difficulty in finding usable and effective health/wellness apps; 4) interest in an online repository of information/reviews of mHealth apps; 5) specific problems or challenges using health/wellness apps; and 6) “wish list” for health/wellness apps that currently do not exist.

Once Maryland students graduate from high school, they have the opportunity to remain in-state for postsecondary education or leave Maryland for further education or employment. This loss of graduates to other states has been termed “brain drain” and is a concern for state policymakers (Zheng & Ness, 2010). Several states have adopted merit-based programs designed to retain college-educated individuals in order to support the state workforce, but outcomes have been inconsistent across the high school-to-college and collegeto-workforce transition points. This study uses data from the Maryland Longitudinal Data System (MLDS) to link student high school records to college and employment information in order to determine the extent of brain drain in Maryland. Findings indicated that brain drain does exist in Maryland to some degree, as students who attended out-of-state colleges were less likely to return to the Maryland workforce when compared to students who attended instate colleges (80% of students who enrolled at Maryland colleges had post-college workforce records, compared to 57% of students who enrolled outside Maryland). Additionally, the students lost to brain drain tended to be higher achieving students. Policy implications and directions for future research on brain drain in Maryland are discussed.

We introduce and provide the first comprehensive comparative assessment of the Federal Reserve Bank of New York Consumer Credit Panel (CCP) to demonstrate the utility and unique advantages of these data for research on internal migration in the United States. Relative to other data sources on U.S. internal migration, the CCP permits highly detailed cross-sectional and longitudinal analyses of migration, both temporally and geographically. After introducing these data, we compare cross-sectional and longitudinal estimates of migration from the CCP to similar estimates derived from the American Community Survey, the Current Population Survey, Internal Revenue Service data, the National Longitudinal Survey of Youth, the Panel Study of Income Dynamics, and the Survey of Income and Program Participation. Our results firmly establish the comparative utility and advantages of the CCP. We conclude by identifying some profitable directions for future research on U.S. internal migration using these data.

Purpose: This study describes the association between frequency of bicycling for active transportation and modifiable cardiometabolic risk factors in the Minneapolis–Saint Paul (MSP) metropolitan area. Design: We conducted an online survey from a convenience sample of likely bicycle commuters. Setting: Participants were e-mailed invitations including a link to the online survey. Participants: Participants (n = 1450, 15.6% response rate) were residents of the 7-county MSP metropolitan area aged 20 to 64 years. Measures: Participants reported frequency of commute and destination cycling from April to September and October to March, prevalent cardiometabolic risk factors, and demographic covariates. Analysis: We performed logistic regressions for associations between average weekly transportation bicycling and self-reported prevalent obesity, high blood pressure, high cholesterol, low high-density lipoprotein cholesterol, and high triglycerides. We used Poisson regression for the count of cardiometabolic risk factors. All models were adjusted for age, sex, race, smoking, alcohol consumption, and other physical activity, with random intercepts by zip code. Results: Self-reported bicycling frequency is associated with significantly lower odds of prevalent obesity, hypertension, and high triglycerides. Three bicycling trips per week is associated with 20% fewer (95% confidence interval: 12%-28%) cardiometabolic risk factors. Conclusion: Bicycling frequency is inversely associated with cardiometabolic risk factors. These results provide empirical support for the promotion of active transportation as public health policy.

Top income inequality in the United States has increased considerably within occupations as diverse as bankers, managers, doctors, lawyers and scientists. The breadth of this phenomenon has led to a search for a common explanation. We show instead that increases in income inequality originating within a few occupations can “spill over” into others, driving broader changes in income inequality. We develop an assignment model where consumers with heterogeneous income buy services from doctors with heterogeneous ability. In equilibrium the highest-earning consumers match with the highest-ability doctors. Increases in income inequality among the consumers feed directly into the doctors’ income inequality. To test our theory, we identify occupations for which our consumption-driven theory predicts spillovers and occupations for which it does not. Using a Bartik-style instrument, we show that an increase in general income inequality causes higher income inequality for doctors, dentists and real estate agents, and in fact accounts for most of the increases in inequality within these occupations. Physician pricing and insurance network data support our mechanism.

Care-providing parents of immigrants fill a labor market niche in the receiving economy. My research examines the intergenerational support for working women in immigrant families in the United States. Using panel data derived from the Current Population Survey (CPS), I find that having a coresiding parent increases the labor force participation probability of foreign-born women with children by about 7.4 percent, and that the effects differ by birth region and educational level. I use a difference-in-differences approach to reinforce the finding that coresiding parents significantly alleviate the short-term labor force participation decline of foreign-born females after their transition to motherhood.