Just Launched and Growing: "Our Researchers"

Our researchers produce innovations that keep our economic and social statistics efficient and relevant. Census researchers engage in challenging and rewarding work in a wide variety of fields. Researchers present their findings in working papers, publications, and reports, and in contributions at professional meetings.

Meet our researchers and view their publications.

• "Our Researchers", a new Research @ Census feature, is a work in process
• If you do not find the researchers or publications you seek, keep checking back
• We will continue to add both research profiles and publications

Latest Publications

• Massey, Catherine G.. 2014. "Creating Linked Historical Data: An Assessment of the Census Bureau’s Ability to Assign Protected Identification Keys to the 1960 Census." CARRA WP No. 2014-12.
  
  • In order to study social phenomena over the course of the 20th century, the Census Bureau is investigating the feasibility of digitizing historical census records and linking them to contemporary data. However, historical censuses have limited personally identifiable information available to match on. In this paper, I discuss the problems associated with matching older censuses to contemporary data files, and I describe the matching process used to match a small sample of the 1960 census to the Social Security Administration Numeric Identification System.
• Johnson, David S., Catherine Massey, and Amy O'Hara. 2014. "The Opportunities and Challenges of Using Administrative Data Linkages to Evaluate Mobility." The ANNALS of the American Academy of Political and Social Science, 657 (1): 247-264.
  
  • Since Alan Krueger’s christening of the Great Gatsby curve, there has been increased attention given to the relationship between inequality and intergenerational social mobility in the United States. Studying intergenerational mobility (IGM) requires longitudinal data across large spans of time as well as the ability to follow parents and children over multiple generations. Few longitudinal datasets meet this need. This article surveys available data and the current and potential issues surrounding the use of administrative records to vastly extend the study of IGM. First, we describe the U.S. Census Bureau’s current uses of administrative records in the linkage of households across household surveys such as the Current Population Survey (CPS), American Community Survey (ACS), Survey of Income and Program Participation (SIPP), and the decennial censuses. Then, we describe the possibilities of creating additional parent-child linkages using the SIPP linked to decennial censuses and the ACS. Last, we outline our model to create linkages across earlier census data (e.g., 1980 and 1990) and contemporary surveys.
• Massey, Catherine G. and Amy O'Hara. 2014. "Person Matching in Historical Files using the Census Bureau’s Person Validation System." CARRA WP No. 2014-11.
  
  • The recent release of the 1940 Census manuscripts enables the creation of longitudinal data spanning the whole of the twentieth century. Linked historical and contemporary data would allow unprecedented analyses of the causes and consequences of health, demographic, and economic change. The Census Bureau is uniquely equipped to provide high quality linkages of person records across datasets. This paper summarizes the linkage techniques employed by the Census Bureau and discusses utilization of these techniques to append protected identification keys to the 1940 Census.
• Luque, Adela, Ron Jarmin, and C.J. Krizan. 2014. "Owner Characteristics and Firm Performance During the Great Recession." CES WP No. 2014-36.
  
  • Minority owned businesses are an increasing important component of the U.S. economy, growing at twice the rate of all U.S. businesses between 2002 and 2007. However, a growing literature indicates that minority-owned businesses may have been especially impacted by the Great Recession. As house prices declined, foreclosures fell disproportionately on urban minority neighborhoods and one of the sources of credit for business owners was severely constrained. Using 2002-2011 data from the Longitudinal Business Database linked to the 2002 Survey of Business Owners, this paper adds to the literature by examining the employment growth and survival of minority and women employer businesses during the last decade, including the Great Recession. At first glance, our preliminary findings suggest that black and women-owned businesses underperform white, male-owned businesses, that Asian-owned businesses outperform other groups, and that Hispanic-owned businesses outperform non-Hispanic ones in regards to employment growth. However, when we look only at continuing firms, black-owned businesses outperform white-owned businesses in terms of employment growth. At the same time, we also find that the recession appears to have impacted black-owned and Hispanic-owned businesses more severely than their counterparts, in terms of employment growth as well as survival. This is also the case for continuing black and Hispanic-owned firms.
• Bond, Brittany, J. David Brown, Adela Luque, and Amy O'Hara. 2014. "The Nature of the Bias When Studying Only Linkable Person Records: Evidence from the American Community Survey." Proceedings of the 2013 Federal Committee on Statistical Methodology (FCSM) Research Conference.
  
  • Record linkage across survey and administrative records sources can greatly enrich data and
    improve their quality. The linkage can reduce respondent burden and nonresponse follow-up
    costs. This is particularly important in an era of declining survey response rates and tight
    budgets. Record linkage also creates statistical bias, however. The U.S. Census Bureau links
    person records through its Person Identification Validation System (PVS), assigning
    each record a Protected Identification Key (PIK). It is not possible to reliably assign a PIK to
    every record, either due to insufficient identifying information or because the information does
    not uniquely match any of the administrative records used in the person validation process. Nonrandom
    ability to assign a PIK can potentially inject bias into statistics using linked data. This
    paper studies the nature of this bias using the 2009 and 2010 American Community Survey
    (ACS). The ACS is well-suited for this analysis, as it contains a rich set of person characteristics
    that can describe the bias. We estimate probit models for whether a record is assigned a PIK. The
    results suggest that young children, minorities, residents of group quarters, immigrants, recent
    movers, low-income individuals, and non-employed individuals are less likely to receive a PIK
    using 2009 ACS. Changes to the PVS process in 2010 significantly addressed the young children
    deficit, attenuated the other biases, and increased the validated records share from 88.1 to 92.6
    percent (person-weighted).
• Krizan, C.J., Adela Luque, and Alice Zawacki. 2014. "The Effect of Employer Health Insurance Offerings on the Growth and Survival of Small Business Prior to the Affordable Care Act." Center for Economic Studies Discussion Paper, CES-WP-14-22.
  
  • Whether or not small businesses offer health insurance to their employees is a critical factor in the health care coverage of many Americans, yet many entrepreneurs and decision makers fear that the cost of offering health care coverage to their employees will diminish the growth and survival of small firms. While there is an emerging consensus among economists that small businesses both create and destroy a disproportionately large number of jobs, less is known about the relationship between health care costs and business growth. We examine this latter issue prior to the passage of the Affordable Care Act in 2010. We first review information on the relationship between small employers’ decisions to offer health insurance prior to 2010 and i) financial factors (including premium variability and tax advantages), ii) labor markets with a focus on employee characteristics and the demand for employer's health insurance, iii) insurance markets and products, discussing access and insurance options with lower premium costs, and iv) the health insurance regulatory environment with an examination of state-level reform and health insurance mandates. We then discuss employer reactions to rising health care costs, followed by a review of factors other than rising health care costs that often affect the growth and survival of a small business. In the remaining sections, we describe our longitudinal analysis of how health insurance offering (HIO) affected the expansion and survival of small businesses. Using 2001-05 linked data from the Longitudinal Business Database (LBD) and the Insurance Component of the Medical Expenditure Panel Survey (MEPS-IC), we look at how HIO affected four measures of business performance (growth in employment, growth in payroll, growth in average wage, and survival) – after controlling for business characteristics and relevant state-level variables. We employ instrumental variable two-stage least squares estimation to address the endogeneity that permeates the question at hand. We find that young businesses (both large and small) that offer health insurance grow at not significantly different rates as those that do not, possibly due to selection effects. Older businesses offering health insurance – both small and large - seem to have higher employment and payroll growth. Survival is strongly and positively correlated with HIO for older establishments at both large and small firms. However, these results should be interpreted with extreme caution due the concerns we raise ab
• Leach, Mark A.. 2014. "A Burden of Support? Household Living Arrangements and Economic Resources Among Mexican Immigrant Families with Children." Journal of Family Issues, 35(1):28-53.
  
  • The children of Mexican immigrants face formidable barriers to achieving socioeconomic mobility due to their parents’ precarious economic position and high rates of unauthorized status. In the short term, Mexican immigrants often coreside in extended household living arrangements with extended kin and unrelated friends and associates to shelter themselves from economic deprivation and insecurity of unauthorized status. Using individual-level Census data, the present study examines how family economic resources relate to household living arrangements. The results are consistent with various theories of immigrant household formation, especially those that explain household structure in terms of economic need and processes of immigration. Families residing in extended arrangements are unique, however, in terms of how often they include a householder and how much they contribute to total household resources, indicators that families may hold more supportive roles within extended households. The implications of the findings for the well-being of immigration children are discussed.
• Bean, Frank D., James D. Bachmeier, Susan K. Brown, Jennifer Van Hook, and Mark A. Leach. 2014. "Unauthorized Mexican Migration and the Socioeconomic Integration of Mexican Americans." In Diversity and Disparities: America Enters a New Century, edited by John Logan. New York: Russell Sage Foundation..
  
  • This study analyzes the sources of unauthorized Mexican immigration to the United States, the disadvantages faced by unauthorized immigrants in the labor market, and especially the long term impact on educational achievement by their US-born children and grandchildren Children of legal Mexican immigrants averaged two more years of schooling compared to children of illegal immigrants, the equivalent of the difference between having some college and not finishing high school. Clear pathways to legalization can boost Mexican American educational attainment even as late as the third generation.
• Rastogi, Sonya, Leticia Fernandez, James Noon, Ellen Zapata, and Renuka Bhaskar. 2014. "Exploring Administrative Records Use for Race and Hispanic Origin Item Non-Response." United Nations Economic Commission for Europe Conference of European Statisticians Working Paper.
  
  • Race and Hispanic origin data are required to produce official statistics in the United States. Data collected through the American Community Survey and decennial census address missing data through traditional imputation methods, often relying on information from neighbors. These methods work well if neighbors share similar characteristics, however, the shape and patterns of neighborhoods in the United States are changing. Administrative records may provide more accurate data compared to traditional imputation methods for missing race and Hispanic origin responses. This paper first describes the characteristics of persons with missing demographic data, then assesses the coverage of administrative records data for respondents who do not answer race and Hispanic origin questions in Census data. The paper also discusses the distributional impact of using administrative records race and Hispanic origin data to complete missing responses in a decennial census or survey context.
• Bhaskar, Renuka , Adela Luque, Sonya Rastogi, and James Noon. 2014. "Coverage and Agreement of Administrative Records and 2010 American Community Survey Demographic Data." CARRA WP No. 2014-14.
  
  • The U.S. Census Bureau is researching possible uses of administrative records in decennial census and survey operations. The 2010 Census Match Study and American Community Survey (ACS) Match Study represent recent efforts by the Census Bureau to evaluate the extent to which administrative records provide data on persons and addresses in the 2010 Census and 2010 ACS. The 2010 Census Match Study also examines demographic response data collected in administrative records. Building on this analysis, we match data from the 2010 ACS to federal administrative records and third party data as well as to previous census data and examine administrative records coverage and agreement of ACS age, sex, race, and Hispanic origin responses. We find high levels of coverage and agreement for sex and age responses and variable coverage and agreement across race and Hispanic origin groups. These results are similar to findings from the 2010 Census Match Study.