On August 7, 2018, the AHA signed on to a letter from the Consortium of Social Science Associations (COSSA) and 24 other research organizations urging the Department of Commerce to remove a controversial citizenship question from the upcoming census. The letter criticizes the science and research implications of the question, arguing that “the inclusion of a question on citizenship in the 2020 Census will increase the burden on respondents, add unnecessary costs to the operation, and negatively impact the accuracy and integrity of one of the most valuable data resources the government produces.”

Download the Letter as a PDF

August 7, 2018

Ms. Jennifer Jessup
Departmental Paperwork Clearance Officer
Department of Commerce, Room 6616
14th and Constitution Avenue, NW
Washington, DC 20230

Re: Docket No. USBC-2018-0005 (Comments submitted electronically via regulations.gov)

Dear Ms. Jessup:

Thank you for the opportunity to comment on the Census Bureau’s proposed information collection activities for the 2020 Census. We, the 26 undersigned organizations, represent the scientific research community, which relies on data derived from the decennial census to advance scientific knowledge and produce valuable findings about the U.S. population that can be used to inform evidence-based policies. We are gravely concerned that the inclusion of a question on citizenship in the 2020 Census will increase the burden on respondents, add unnecessary costs to the operation, and negatively impact the accuracy and integrity of one of the most valuable data resources the government produces. We urge that the Department of Commerce remove the question from 2020 Census forms.

Many organizations have identified a number of potential negative impacts of this question—on representation and apportionment, on the allocation of over $800 billion in federal funds to state and local programs, and on the private sector’s ability to make informed business decisions, to name a few.

The concerns expressed in this letter, however, address the Census as a scientific enterprise: as an undertaking that is guided by science in its design and operation and as an invaluable source of data for researchers whose discoveries improve the lives of countless Americans.

The decision to add the citizenship question circumvented the extensive research and testing the Census Bureau routinely conducts in the years leading up to a decennial census and was made against the advice of experts at the Census Bureau, including Chief Scientist John M. Abowd, who wrote in a memo to Commerce Secretary Ross that adding such a question is “very costly” and “harms the quality of the census count.”¹

As a matter of course, the Bureau spends years before a decennial census carefully researching all proposed changes to the design and wording to ensure that they do not affect the quality of the responses received. In this case, agency experts were not given the opportunity to comprehensively evaluate the potential impact of the citizenship question. The evidence that does exist, however, suggests that the information collected is likely to be flawed and will harm the overall Census operation by deterring self-response. Research conducted by the Census Bureau indicates that a citizenship question would likely increase the number of households that do not respond to the Census at all—particularly in already hard-to-count immigrant communities.² This would increase the overall costs of the operation, as more enumerators would need be sent to collect responses in person, at far greater expense than planned mail or internet outreach. A decline in response from immigrant communities could result in an undercount of these populations, affecting the accuracy and integrity of Census data overall. In addition, evidence from the American Community Survey suggests that as many as 30 percent of non-citizens incorrectly identify themselves as citizens when responding to the survey,³ suggesting that the citizenship data collected at great expense during the 2020 Census could very well be inaccurate.

The costs of an inaccurate decennial Census to the scientific community—and the American public—are incredibly high. The decennial census is an irreplaceable source of data for researchers in varying fields who use it to produce evidence about the U.S. population that can be used to inform policies. In addition, information from the decennial census undergirds numerous other surveys and data sets at the Census Bureau and beyond, so a problem at the source would have far-reaching implications across the statistical system.

Given the high degree of risk and uncertainty and the likely poor quality of the citizenship information collected during the 2020 Census, the benefits of including a citizenship question on the 2020 Census do not outweigh the costs—both in terms of increased respondent burden and on the resulting Census data. In his memo to Secretary Ross, John Abowd noted that information on citizenship could be obtained for all but two percent of the U.S. population using administrative records.⁴ Collecting citizenship information from every person to fill in missing information for only two percent of the population should meet any reasonable definition of “unnecessary burden.”

The Census is a once-a-decade undertaking; if we allow the integrity of the data to be jeopardized by an untested, unresearched citizenship question, we will be living with the harmful consequences for years. We have no way of knowing what future insights will be lost if this data is compromised.

For the reasons discussed above, we strongly oppose including citizenship in the 2020 Census data collection and urge the Department of Commerce to remove it. Thank you for your consideration. If we may be of any assistance or can provide additional information, please contact Julia Milton at jmilton@cossa.org.

Sincerely,

American Academy of Political and Social Science
American Anthropological Association
American Association for Public Opinion Research
American Association of Geographers
American Educational Research Association
American Evaluation Association
American Historical Association
American Political Science Association
American Psychological Association
American Sociological Association
American Statistical Association
Association for Behavioral and Cognitive Therapies
Association of Population Centers Association of Public Data Users (APDU)
Association of Research Libraries
Consortium of Social Science Associations
Federation of Associations in Behavioral and Brain Sciences
Institute for Social Science Research, University of Massachusetts, Amherst
Knowledge Alliance
Midwest Political Science Association
National Communication Association
Population Association of America
Social Science Research Council
Society for Prevention Research
Society for Research in Child Development
Southern Political Science Association

1 http://www.osec.doc.gov/opog/FOIA/Documents/AR - FINAL FILED - ALL DOCS [CERTIFICATION-INDEX-DOCUMENTS] 6.8.18.pdf, pg 1289
2 http://www.osec.doc.gov/opog/FOIA/Documents/AR - FINAL FILED - ALL DOCS [CERTIFICATION-INDEX-DOCUMENTS] 6.8.18.pdf, pg 1293
3 http://www.osec.doc.gov/opog/FOIA/Documents/AR - FINAL FILED - ALL DOCS [CERTIFICATION-INDEX-DOCUMENTS] 6.8.18.pdf, pg 1295
4 http://www.osec.doc.gov/opog/FOIA/Documents/AR - FINAL FILED - ALL DOCS [CERTIFICATION-INDEX-DOCUMENTS] 6.8.18.pdf, pg 1324