Wednesday, Aug 7: 10:30 AM - 12:20 PM
5148
Contributed Papers
Oregon Convention Center
Room: CC-E147
Main Sponsor
Government Statistics Section
Presentations
The Census Bureau's Demographic Frame is a comprehensive, person-level frame consisting of geographic, demographic, social, and economic characteristics. It can be used to identify addresses associated with each person and potentially to identify their correct residence. A person found on the Demographic Frame often has multiple person-address records. These multiple records create difficulty in placing a person at their correct residential address. The Demographic Frame has a person-place model process that assigns probabilities to person-address records. This analysis will evaluate addresses for people from these models on the Demographic Frame based on a reference date of July 1, 2021. As part of this analysis, we will compare these Demographic Frame addresses to addresses in the 2020 Census Enumeration and the 2021 American Community Survey frames. This comparison will allow us to examine the Demographic Frame addresses that are not found within these other Census products, which may provide information to help identify whether subsets of these addresses may be more likely to be residential or non-residential addresses.
Keywords
Demographic Frame
Census
Population Estimates
Sampling Frame
Administrative Records
American Community Survey
Past research has shown that survey misreporting can potentially bias income estimates. Measurement of Social Security income faces additional challenges as respondents may be confused about the concept of gross versus net value of Social Security. Building on past work, this paper links reference year 2017 data from the Survey of Income and Program Participation (SIPP) to administrative data from the Social Security Administration to explore the accuracy with which respondents report: 1) annual Social Security receipt; 2) number of months of income received; and 3) monthly payment amounts. This final component will explore whether respondents are reporting their gross Social Security income (as the survey item asks), or the size of the check that they actually receive (net of any deductions for Medicare premiums). Results will inform how questions could be reworded to improve measurement, and whether administrative records could be used to augment survey reports in advance of upcoming redesigns of the SIPP instrument and edits.
Keywords
Administrative Records
Social Security
Medicare
Record Linkage
Data Quality
Abstracts
This report develops a method using administrative records (AR) to fill in responses for nonresponding American Community Survey (ACS) housing units rather than adjusting survey weights to account for selection of a subset of nonresponding housing units for follow-up interviews and for nonresponse bias. The method also inserts AR and modeling in place of edits and imputations for ACS survey citizenship item nonresponses. We produce Citizen Voting-Age Population (CVAP) tabulations using this enhanced CVAP method and compare them to published estimates. The enhanced CVAP method produces a 0.74 percentage point lower citizen share, and it is 3.05 percentage points lower for voting-age Hispanics. The latter result can be partly explained by omissions of voting-age Hispanic noncitizens with unknown legal status from ACS household responses. Weight adjustments may be less effective at addressing nonresponse bias under those conditions.
Keywords
Citizenship
Administrative Records
Voting-Age Population
Nonresponse Bias
The U.S. Census Bureau's Demographic Frame (DF) is a comprehensive, person-level frame consisting of demographic, social, and economic characteristics of individuals derived from census, survey, administrative, and third-party data sources.
The DF is expected to provide information about the residential location of individuals to the Longitudinal Employer-Household Dynamics program; this information is currently generated from the production of the Residence Candidate File (RCF).
The current RCF needs to be assessed relative to the output generated by the DF, which establishes people's residence with its nascent Person-Place Model. This assessment will inform a plan for the eventual replacement of the RCF with a product derived from the DF. This analysis will be used to determine what improvements are necessary for the DF to meet the needs of the Job Frame for residential location of individuals.
This presentation will summarize results of this effort to evaluate and report out on the gaps between the RCF and the DF. Results will include overall comparison between these two data products, as well as geographic distance of residence location where there are differences.
Keywords
Linked Survey and Administrative Data
Official Statistics
Administrative Records
Modernizing Infrastructure
Government
We produce annual population estimates at low levels of geography in 2018-2021 using person-level administrative records (AR) from a large number of sources. Population change is decomposed into births, deaths, internal migration, and entry into and exit from the AR data of people not known to have been born or died during the period. The demographic distributions of each of these groups is shown. The results are compared to the Census Bureau's Population Estimates Program estimates, the American Community Survey, and Internal Revenue Service estimates. We examine the extent to which the entry and exit could be a result of inconsistent person coverage in AR data over time and discuss methods for addressing this.
Keywords
administrative records
population estimates
The 2020 Census and other data sources such as administrative records often disagree on the race and Hispanic origin of a person. If we assume that the values reported in the 2020 Census are more accurate, then we find that administrative records have a high rate of misclassification. Moreover, we find different definitions for race and Hispanic origin between data sources, which complicate comparisons. In this paper, we analyze the race and Hispanic origin classifications of people on the 2020 Census, the 2020 Post-Enumeration Survey (PES), and administrative records. We discuss methods for correcting misclassification of the race and Hispanic origin values on administrative records, including corrections to these characteristics for people in administrative records that match or do not match to the 2020 Census. We also discuss a potentially problematic scenario where the definitions of race and Hispanic origin are not the same for different data sources. Finally, we discuss the impact misclassification has on PES estimates of population.
Keywords
dual-system estimation
matching
misclassification
multiple-race identification
post-enumeration survey
small-area estimation
The Continuous Count Study (CCS) is an important research effort within Census to generate lower level geographic and demographic estimates throughout the decade. One part of the study involves the use of administrative records to generate these estimates. The Demographic Frame is a comprehensive database of person-level data containing demographic characteristics and addresses associated with each person. Its information is derived from administrative, third-party, decennial census, and survey data sources. Prior modelling work has focused on predicting the probability that a given individual is found at a particular address on Census Day.
This research will examine a triple-system estimation of person counts where the three systems are: 2020 Census, 2020 American Community Survey (ACS), and a particular vintage of the Demographic Frame. We follow the ideas of Van der Heijden et al (2018, 2021), who estimated populations in New Zealand where individuals may be found on any number of lists, including being missing from all lists. We will be making use of the R package cvam (Schafer 2021) to fit the models, including a latent class model to assign demographic information.
Keywords
Multiple Systems Estimation
Latent Class Modeling
Administrative Records