Comment on 2020 Census Federal Register Notice

Heidi Hartmann, Ph.D.

August 7, 2018
  • ID:

August 7, 2018


Jennifer Jessup

Departmental Paperwork Clearance Officer
Department of Commerce
14th and Constitution Avenue NW, Room 6616
Washington, DC 20230
Re: USBC-2018-0005-0001, 2020 Census Federal Register Notice


Dear Ms. Jessup:


The Institute for Women’s Policy Research (IWPR) writes to express its concern over proposed changes to the 2020 Census. IWPR conducts rigorous research and disseminates its findings to address the needs of women, promote public dialog, and strengthen families, communities, and societies.


IWPR relies on accurate and timely data from the federal statistical agencies, including the Census Bureau, for research across our program areas. A comprehensive, accurate decennial Census provides the primary source of information on the U.S. population and is the foundation on which many other data are based for gauging coverage, constructing population weights, and providing other inputs across the federal statistical system. As a result, we share a profound commitment with other census stakeholders to ensuring full public participation in the decennial Census. We not only rely directly on the decennial Census itself in our work, but we also rely on all the other federal and private data systems that use the decennial Census to ensure the accuracy of their data collection, sampling frames, population weights, and so on.


We are concerned that including an untested citizenship question on the 2020 Census will discourage participation in the decennial Census—especially among vulnerable, hard-to-count populations. Asking households to enumerate their members by an untested question designed to collect citizenship status is very likely to be perceived as a risk by many respondents, even those in the U.S. with proper visas/documentation, and will certainly increase the burden of response, suppress response rates, and reduce overall data quality. These negative outcomes are more likely to occur in urban areas that attract a greater number of single mothers (who give birth to majority of first-time births), immigrants, and racial minorities, and other vulnerable minorities, for example adherents of the muslin religions. The impact will be quite large when one considers that many families have members of different citizenship statuses.


For example in 2017, 16.4 million native-born children were living with one or more foreign-born parents and 2.4 million of these children were living with only their mother.


The failure to count fully the children of single mothers alone is sufficiently large to reduce future grants to the cities where many live, because poverty assistance dollars are distributed according to how the Census measures poverty.


Furthermore, we believe that an undercount will not be distributed evenly across the country and will therefore undermine the primary purpose of the Census to count all residents for apportionment of the seats in the U.S. House of Representatives. The enumeration total and quality in urban areas are likely to be most adversely affected by the decision to seek reporting of household members’ citizenship status.


This uneven undercount will, as noted above, reduce the efficient use of federal funds under many important government programs. In particular this uneven undercount will trigger a redistribution of resources away from cities and their vulnerable populations, such as families headed by a single mother or otherwise in poverty, families headed by racial minorities and immigrants (both documented and lacking documentation, as noted above), and veterans. It will also be detrimental to governments, at all levels, that try to provide services to their resident populations, such as education, employment supports, poverty reduction, sidewalks and roads, building permits, garbage collection, and other services.


We ask the administration to reconsider the proposed addition of the citizenship question to the 2020 Decennial Census. We believe the question on citizenship is not only not necessary to fulfill the Census’ mandate, but also the addition of the citizenship question will reduce data quality and lead to the serious misallocation of Congressional representatives and important economic resources to the most vulnerable people in our country: single mothers and their families, poor families, immigrants, both documented and undocumented, and their families, families of racial minorities, and veterans and their families. The inclusion of the citizenship status will also increase Census costs for both the Census Bureau and all the respondents in the United States and abroad (as we also try to include nationals living abroad).




Heidi Hartmann, Ph.D.


President & CEO
Institute for Women’s Policy Research
Senior Economist in Residence, American University
Editor, Journal of Women, Politics & Policy
MacArthur Fellow
Charlotte Perkins Gilman Fellow, American Academy of Political and Social Science