-
Revisiting Methods to Assign Responses when Race and Hispanic Origin Reporting are Discrepant Across Administrative Records and Third Party Sources
May 2024
Working Paper Number:
CES-24-26
The Best Race and Ethnicity Administrative Records Composite file ('Best Race file') is an composite file which combines Census, federal, and Third Party Data (TPD) sources and applies business rules to assign race and ethnicity values to person records. The first version of the Best Race administrative records composite was first constructed in 2015 and subsequently updated each year to include more recent vintages, when available, of the data sources originally included in the composite file. Where updates were available for data sources, the most recent information for persons was retained, and the business rules were reapplied to assign a single race and single Hispanic origin value to each person record. The majority of person records on the Best Race file have consistent race and ethnicity information across data sources. Where there are discrepancies in responses across data sources, we apply a series of business rules to assign a single race and ethnicity to each record. To improve the quality of the Best Race administrative records composite, we have begun revising the business rules which were developed several years ago. This paper discusses the original business rules as well as the implemented changes and their impact on the composite file.
View Full
Paper PDF
-
Nonemployer Statistics by Demographics (NES-D):
Exploring Longitudinal Consistency and Sub-national Estimates
December 2019
Working Paper Number:
CES-19-34
Until recently, the quinquennial Survey of Business Owners (SBO) was the only source of information for U.S. employer and nonemployer businesses by owner demographic characteristics such as race, ethnicity, sex and veteran status. Now, however, the Nonemployer Statistics by Demographics series (NES-D) will replace the SBO's nonemployer component with reliable, and more frequent (annual) business demographic estimates with no additional respondent burden, and at lower imputation rates and costs. NES-D is not a survey; rather, it exploits existing administrative and census records to assign demographic characteristics to the universe of approximately 25 million (as of 2016) nonemployer businesses.
Although only in the second year of its research phase, NES-D is rapidly moving towards production, with a planned prototype or experimental version release of 2017 nonemployer data in 2020, followed by annual releases of the series. After the first year of research, we released a working paper (Luque et al., 2019) that assessed the viability of estimating nonemployer demographics exclusively with administrative records (AR) and census data. That paper used one year of data (2015) to produce preliminary tabulations of business counts at the national level. This year we expand that research in multiple ways by: i) examining the longitudinal consistency of administrative and census records coverage, and of our AR-based demographics estimates, ii) evaluating further coverage from additional data sources, iii) exploring estimates at the sub-national level, iv) exploring estimates by industrial sector, v) examining demographics estimates of business receipts as well as of counts, and vi) implementing imputation of missing demographic values.
Our current results are consistent with the main findings in Luque et al. (2019), and show that high coverage and demographic assignment rates are not the exception, but the norm. Specifically, we find that AR coverage rates are high and stable over time for each of the three years we examine, 2014-2016. We are able to identify owners for approximately 99 percent of nonemployer businesses (excluding C-corporations), 92 to 93 percent of identified nonemployer owners have no missing demographics, and only about 1 percent are missing three or more demographic characteristics in each of the three years. We also find that our demographics estimates are stable over time, with expected small annual changes that are consistent with underlying population trends in the U.S.. Due to data limitations, these results do not include C-corporations, which represent only 2 percent of nonemployer businesses and 4 percent of receipts.
Without added respondent burden and at lower imputation rates and costs, NES-D will provide high-quality business demographics estimates at a higher frequency (annual vs. every 5 years) than the SBO.
View Full
Paper PDF
-
Nonemployer Statistics by Demographics (NES-D): Using Administrative and Census Records Data in Business Statistics
January 2019
Working Paper Number:
CES-19-01
The quinquennial Survey of Business Owners or SBO provided the only comprehensive source of information in the United States on employer and nonemployer businesses by the sex, race, ethnicity and veteran status of the business owners. The annual Nonemployer Statistics series (NES) provides establishment counts and receipts for nonemployers but contains no demographic information on the business owners. With the transition of the employer component of the SBO to the Annual Business Survey, the Nonemployer Statistics by Demographics series or NES-D represents the continuation of demographics estimates for nonemployer businesses. NES-D will leverage existing administrative and census records to assign demographic characteristics to the universe of approximately 24 million nonemployer businesses (as of 2015). Demographic characteristics include key demographics measured by the SBO (sex, race, Hispanic origin and veteran status) as well as other demographics (age, place of birth and citizenship status) collected but not imputed by the SBO if missing. A spectrum of administrative and census data sources will provide the nonemployer universe and demographics information. Specifically, the nonemployer universe originates in the Business Register; the Census Numident will provide sex, age, place of birth and citizenship status; race and Hispanic origin information will be obtained from multiple years of the decennial census and the American Community Survey; and the Department of Veteran Affairs will provide administrative records data on veteran status.
The use of blended data in this manner will make possible the production of NES-D, an annual series that will become the only source of detailed and comprehensive statistics on the scope, nature and activities of U.S. businesses with no paid employment by the demographic characteristics of the business owner. Using the 2015 vintage of nonemployers, initial results indicate that demographic information is available for the overwhelming majority of the universe of nonemployers. For instance, information on sex, age, place of birth and citizenship status is available for over 95 percent of the 24 million nonemployers while race and Hispanic origin are available for about 90 percent of them. These results exclude owners of C-corporations, which represent only 2 percent of nonemployer firms. Among other things, future work will entail imputation of missing demographics information (including that of C-corporations), testing the longitudinal consistency of the estimates, and expanding the set of characteristics beyond the demographics mentioned above. Without added respondent burden and at lower imputation rates and costs, NES-D will meet the needs of stakeholders as well as the economy as a whole by providing reliable estimates at a higher frequency (annual vs. every 5 years) and with a more timely dissemination schedule than the SBO.
View Full
Paper PDF
-
The Use of Administrative Records and the American Community Survey to Study the Characteristics of Undercounted Young Children in the 2010 Census
May 2018
Working Paper Number:
carra-2018-05
Children under age five are historically one of the most difficult segments of the population to enumerate in the U.S. decennial census. The persistent undercount of young children is highest among Hispanics and racial minorities. In this study, we link 2010 Census data to administrative records from government and third party data sources, such as Medicaid enrollment data and tenant rental assistance program records from the Department of Housing and Urban Development, to identify differences between children reported and not reported in the 2010 Census. In addition, we link children in administrative records to the American Community Survey to identify various characteristics of households with children under age five who may have been missed in the last census. This research contributes to what is known about the demographic, socioeconomic, and household characteristics of young children undercounted by the census. Our research also informs the potential benefits of using administrative records and surveys to supplement the U.S. Census Bureau child population enumeration efforts in future decennial censuses.
View Full
Paper PDF
-
Reporting of Indian Health Service Coverage in the American Community Survey
May 2018
Working Paper Number:
carra-2018-04
Response error in surveys affects the quality of data which are relied on for numerous research and policy purposes. We use linked survey and administrative records data to examine reporting of a particular item in the American Community Survey (ACS) - health coverage among American Indians and Alaska Natives (AIANs) through the Indian Health Service (IHS). We compare responses to the IHS portion of the 2014 ACS health insurance question to whether or not individuals are in the 2014 IHS Patient Registration data. We evaluate the extent to which individuals misreport their IHS coverage in the ACS as well as the characteristics associated with misreporting. We also assess whether the ACS estimates of AIANs with IHS coverage represent an undercount. Our results will be of interest to researchers who rely on survey responses in general and specifically the ACS health insurance question. Moreover, our analysis contributes to the literature on using administrative records to measure components of survey error.
View Full
Paper PDF
-
Medicare Coverage and Reporting
December 2016
Working Paper Number:
carra-2016-12
Medicare coverage of the older population in the United States is widely recognized as being nearly universal. Recent statistics from the Current Population Survey Annual Social and Economic Supplement (CPS ASEC) indicate that 93 percent of individuals aged 65 and older were covered by Medicare in 2013. Those without Medicare include those who are not eligible for the public health program, though the CPS ASEC estimate may also be impacted by misreporting. Using linked data from the CPS ASEC and Medicare Enrollment Database (i.e., the Medicare administrative data), we estimate the extent to which individuals misreport their Medicare coverage. We focus on those who report having Medicare but are not enrolled (false positives) and those who do not report having Medicare but are enrolled (false negatives). We use regression analyses to evaluate factors associated with both types of misreporting including socioeconomic, demographic, and household characteristics. We then provide estimates of the implied Medicare-covered, insured, and uninsured older population, taking into account misreporting in the CPS ASEC. We find an undercount in the CPS ASEC estimates of the Medicare covered population of 4.5 percent. This misreporting is not random - characteristics associated with misreporting include citizenship status, year of entry, labor force participation, Medicare coverage of others in the household, disability status, and imputation of Medicare responses. When we adjust the CPS ASEC estimates to account for misreporting, Medicare coverage of the population aged 65 and older increases from 93.4 percent to 95.6 percent while the uninsured rate decreases from 1.4 percent to 1.3 percent.
View Full
Paper PDF
-
Response Error & the Medicaid undercount in the CPS
December 2016
Working Paper Number:
carra-2016-11
The Current Population Survey Annual Social and Economic Supplement (CPS ASEC) is an important source for estimates of the uninsured population. Previous research has shown that survey estimates produce an undercount of beneficiaries compared to Medicaid enrollment records. We extend past work by examining the Medicaid undercount in the 2007-2011 CPS ASEC compared to enrollment data from the Medicaid Statistical Information System for calendar years 2006-2010. By linking individuals across datasets, we analyze two types of response error regarding Medicaid enrollment - false negative error and false positive error. We use regression analysis to identify factors associated with these two types of response error in the 2011 CPS ASEC. We find that the Medicaid undercount was between 22 and 31 percent from 2007 to 2011. In 2011, the false negative rate was 40 percent, and 27 percent of Medicaid reports in CPS ASEC were false positives. False negative error is associated with the duration of enrollment in Medicaid, enrollment in Medicare and private insurance, and Medicaid enrollment in the survey year. False positive error is associated with enrollment in Medicare and shared Medicaid coverage in the household. We discuss implications for survey reports of health insurance coverage and for estimating the uninsured population.
View Full
Paper PDF
-
When Race and Hispanic Origin Reporting are Discrepant Across Administrative Records and Third Party Sources: Exploring Methods to Assign Responses
December 2015
Working Paper Number:
carra-2015-08
The U.S. Census Bureau is researching uses of administrative records and third party data in survey and decennial census operations. One potential use of administrative records is to utilize these data when race and Hispanic origin responses are missing. When federal and third party administrative records are compiled, race and Hispanic origin responses are not always the same for an individual across sources. We explore different methods to assign one race and one Hispanic response when these responses are discrepant. We also describe the characteristics of individuals with matching, non-matching, and missing race and Hispanic origin data by demographic, household, and contextual variables. We find that minorities, especially Hispanics, are more likely to have non-matching Hispanic origin and race responses in administrative records and third party data compared to the 2010 Census. Minority groups and individuals ages 0-17 are more likely to have missing race or Hispanic origin data in administrative records and third party data. Larger households tend to have more missing race data in administrative records and third party data than smaller households.
View Full
Paper PDF
-
An outside view: What do observers say about others' races and Hispanic origins?
August 2015
Working Paper Number:
carra-2015-05
Outsiders' views of a person's race or Hispanic origin can impact how she sees herself, how she reports her race and Hispanic origins, and her social and economic experiences. The way outsiders describe non-strangers in terms of their race and Hispanic origin may reveal popular assumptions about which race/Hispanic categories are salient for Americans, which kinds of people are seen as multiracial, and the types of cues people use when identifying another person's race. We study patterns of observer identification using a unique, large, linked data source with two measures of a person's race and Hispanic origin. One measure (from Census 2000 or the 2010 Census) was provided by a household respondent and the other (from the other census year) was provided by a census proxy reporter (e.g., a neighbor) who responded on behalf of a non-responsive household. We ask: Does an outsider's report of a person's race and Hispanic origin match a household report? We find that in about 90% of our 3.7 million (nonrepresentative) cases, proxy reports of a person's race and Hispanic origin match responses given by the household in a different census year. Match rates are high for the largest groups: non-Hispanic whites, blacks, and Asians and for Hispanics, though proxies are not very able to replicate the race responses of Hispanics. Matches are much less common for people in smaller groups (American Indian/Alaska Native, Pacific Islander, Some Other Race, and multiracial). We also ask: What predicts a matched response and what predicts a particular unmatched response? We find evidence of the persistence of hypodescent for blacks and hyperdescent for American Indians. Biracial Asian-whites and Pacific Islander-whites are more often seen by others as non-Hispanic white than as people of color. Proxy reporters tend to identify children as multiple race and elders as single race, whether they are or not. The race/Hispanic composition of the tract is more powerfully predictive of a particular unmatched response than are tract-level measures of socioeconomic status; unmatched responses are often consistent with the race/Hispanic characteristics of the neighborhood.
View Full
Paper PDF
-
Evaluating Race and Hispanic Origin Responses of Medicaid Participants Using Census Data
April 2015
Working Paper Number:
carra-2015-01
Health and health care disparities associated with race or Hispanic origin are complex and continue to challenge researchers and policy makers. With the intention of improving the measurement and monitoring of these disparities, provisions of the Patient Protection and Affordable Care Act (ACA) of 2010 require states to collect, report and analyze data on demographic characteristics of applicants and participants in Medicaid and other federally supported programs. By linking Medicaid records to 2010 Census, American Community Survey, and Census 2000, this new large-scale study examines and documents the extent to which pre-ACA Medicaid administrative records match self-reported race and Hispanic origin in Census data. Linked records allow comparisons between individuals with matching and non-matching race and Hispanic origin data across several demographic, socioeconomic and neighborhood characteristics, such as age, gender, language proficiency, education and Census tract variables. Identification of the groups most likely to have non-matching and missing race and Hispanic origin data in Medicaid relative to Census data can inform strategies to improve the quality of demographic data collected from Medicaid populations.
View Full
Paper PDF
