Differential Undercounts in the U.S. Census

Abstract

SpringerBriefs in Population Studies presents concise summaries of cutting-edge research and practical applications across the field of demography and population studies. It publishes compact refereed monographs under the editorial supervision of an international Advisory Board. Volumes are compact, 50 to 125 pages, with a clear focus. The series covers a range of content from professional to academic such as: timely reports of state-of-the art analytical techniques, bridges between new research results, snapshots of hot and/or emerging topics, and in-depth case studies. The scope of the series spans the entire field of demography and population studies, with a view to significantly advance research. The character of the series is international and multidisciplinary and includes research areas such as: population aging, fertility and family dynamics, demography, migration, population health, household structures, mortality, human geography and environment. Volumes in this series may analyze past, present and/or future trends, as well as their determinants and consequences. Both solicited and unsolicited manuscripts are considered for publication in this series. SpringerBriefs in Population Studies will be of interest to a wide range of individuals with interests in population studies, including demographers, population geographers, sociologists, economists, political scientists, epidemiologists and health researchers as well as practitioners across the social sciences.

Trends in fatal occupational injuries in Latino/a workers relative to other groups, North Carolina 2000–2017

“…Bias may be present in our denominator due to sampling limitations of the decennial and ACS population estimates. We also note that prior research suggests that undercounting of Latino/a workers may be present in Census data, especially in areas of larger Latino/a populations 40–42 . Systematic undercounting of the Latino/a workforce could yield an overestimate of the yearly rate; however, we note that estimates of trend are robust to misclassification of the population at risk, provided that misspecification of the workforce is consistent over time.…”

“…We also note that prior research suggests that undercounting of Latino/a workers may be present in Census data, especially in areas of larger Latino/a populations. [40][41][42] Systematic undercounting of the Latino/a workforce could yield an overestimate of the yearly rate; however, we note that estimates of trend are robust to misclassification of the population at risk, provided that misspecification of the workforce is consistent over time. Our aim in conducting statistical analysis of these observational data is summarization, smoothing, and pattern recognition, 43 and we therefore focus on reporting events, rates, and trends, rather than statistical null hypothesis testing and p-values.…”

Coming To Our Census: How Social Statistics Underpin Our Democracy (And Republic)

“…Undercounts, especially of a minority population, are of immense interest ( O’Hare, 2018b ). In the political back-and-forth around discussion of undercounts, it is easy for the errors to be attributed to the Census Bureau, their procedures, and their personnel.…”

Seed-Fill-Shift-Repair: A redistricting heuristic for civic deliberation

“…First, the decennial census is not a perfect count. Undercounts tend to be of renters, young males, immigrants, and African-Americans, and overcounts of homeowners, middle-aged females, and college students [30,31]. Births, deaths, and migration between the census date and the release of redistricting data mean that even if an exact count were possible, it would no longer be exact when used for redistricting.…”