CREAT: Census Research Exploration and Analysis Tool

Frequently Occurring Concepts within this Search

Viewing papers 1 through 4 of 4

• Working Paper

  
  
  Finding Needles in Haystacks: Multiple-Imputation Record Linkage Using Machine Learning
  

  November 2021

  Authors: Kristin McCue, John M. Abowd, Matthew D. Shapiro, Trivellore Raghunathan, Margaret C. Levenstein, Joelle Abramowitz, Dhiren Patki, Ann M. Rodgers, Nada Wasi, Dawn Zinsser

  Working Paper Number:

  CES-21-35

  This paper considers the problem of record linkage between a household-level survey and an establishment-level frame in the absence of unique identifiers. Linkage between frames in this setting is challenging because the distribution of employment across establishments is highly skewed. To address these difficulties, this paper develops a probabilistic record linkage methodology that combines machine learning (ML) with multiple imputation (MI). This ML-MI methodology is applied to link survey respondents in the Health and Retirement Study to their workplaces in the Census Business Register. The linked data reveal new evidence that non-sampling errors in household survey data are correlated with respondents' workplace characteristics.
  View Full Paper PDF

  ---

• Working Paper

  
  
  Optimal Probabilistic Record Linkage: Best Practice for Linking Employers in Survey and Administrative Data
  

  March 2019

  Authors: Kristin McCue, John M. Abowd, Matthew D. Shapiro, Trivellore Raghunathan, Margaret C. Levenstein, Joelle Abramowitz, Dhiren Patki, Ann M. Rodgers, Nada Wasi

  Working Paper Number:

  CES-19-08

  This paper illustrates an application of record linkage between a household-level survey and an establishment-level frame in the absence of unique identifiers. Linkage between frames in this setting is challenging because the distribution of employment across firms is highly asymmetric. To address these difficulties, this paper uses a supervised machine learning model to probabilistically link survey respondents in the Health and Retirement Study (HRS) with employers and establishments in the Census Business Register (BR) to create a new data source which we call the CenHRS. Multiple imputation is used to propagate uncertainty from the linkage step into subsequent analyses of the linked data. The linked data reveal new evidence that survey respondents' misreporting and selective nonresponse about employer characteristics are systematically correlated with wages.
  View Full Paper PDF

  ---

• Working Paper

  
  
  Planning Parenthood: The Affordable Care Act Young Adult Provision and Pathways to Fertility
  

  January 2017

  Authors: Joelle Abramowitz

  Working Paper Number:

  CES-17-65

  This paper investigates the effect of the Affordable Care Act young adult provision on fertility and related outcomes. The expected effect of the provision on fertility is not clear ex ante. By expanding insurance coverage to young adults, the provision may affect fertility directly through expanded options for obtaining contraceptives as well as through expanded options for obtaining pregnancy-, birth-, and infant-related care, and these may lead to decreased or increased fertility, respectively. In addition, the provision may also affect fertility indirectly through marriage or labor markets, and the direction and magnitude of these effects is difficult to determine. This paper considers the effect of the provision on fertility as well as the contributing channels by applying difference-in-differences-type methods using the 2008-2010 and 2012-2013 American Community Survey, 2006-2009 and 2012-2013 Centers for Disease Control and Prevention abortion surveillance data, and 2006-2010 and 2011-2013 National Survey of Family Growth. Results suggest that the provision is associated with decreases in the likelihood of having given birth and abortion rates and an increase in the likelihood of using long-term hormonal contraceptives.
  View Full Paper PDF

  ---

• Working Paper

  
  
  Considering the Use of Stock and Flow Outcomes in Empirical Analyses: An Examination of Marriage Data
  

  January 2017

  Authors: Joelle Abramowitz, Marcus Dillender

  Working Paper Number:

  CES-17-64

  This paper fills an important void assessing how the use of stock outcomes as compared to flow outcomes may yield disparate results in empirical analyses, despite often being used interchangeably. We compare analyses using a stock outcome, marital status, to those using a flow outcome, entry into marriage, from the same dataset, the American Community Survey. This paper considers two different questions and econometric approaches using these alternative measures: the effect of the Affordable Care Act young adult provision on marriage using a difference-indifferences approach and the relationship between aggregate unemployment rates and marriage rates using a simpler ordinary least squares regression approach. Results from both analyses show stock and flow data yield divergent results in terms of sign and significance. Additional analyses suggest prior-period temporary shocks and migration may contribute to this discrepancy. These results suggest using caution when conducting analyses using stock data as they may produce false negative results or spurious false positive results, which could in turn give rise to misleading policy implications.
  View Full Paper PDF

  ---