Census Bureau practice test for 2030 uses citizenship question, experts say
The U.S. Census Bureau is conducting a practice test for the 2030 census that includes survey questions with a citizenship question, raising concerns among experts about how the testing is being run and whether it signals a broader push to change how people are counted, according to the Associated Press. The test is being conducted in Huntsville, Alabama, and Spartanburg, South Carolina, and it uses questions from the American Community Survey rather than questions from recent census forms.
The American Community Survey is described in the report as the comprehensive survey of American life, and one question on the form asks whether “this person a citizen of the United States.” Questions for the census, the report says, are not supposed to ask about citizenship and have not done so for 75 years, a point raised in the discussion of how the bureau’s practice test differs from past census practice.
Terri Ann Lowenthal, a former congressional staffer who consults on census issues, said the citizenship-question approach represented a shift in what the bureau is testing. She said the American Community Survey questions have never been used for a census field test before, and she criticized changes to the overall test plan, saying the 2026 test—which the report says was pared down from six locations to two—has become “a shell of what the Census Bureau proposed and should do to ensure an accurate 2030 Census.”
The Associated Press report also notes that the Census Bureau did not respond to inquiries seeking comment about why the citizenship-question-containing American Community Survey questions were being used for the 2026 test. The practice test gives the bureau an opportunity to learn how to better tally populations that were undercounted during the last census in 2020 and to refine methods that will be used for the 2030 count, the report said.
Among the new methods described in the report is the use of U.S. Postal Service workers to perform tasks that were previously handled by census workers. The Associated Press report adds that the test originally was supposed to take place in six locations, but that the Trump administration earlier this week eliminated four sites, including Colorado Springs, Colorado, western North Carolina, western Texas and tribal lands in Arizona.
Mark Mather, an associate vice president at the Population Reference Bureau, said he could not speculate on political motivations behind the decision to use the American Community Survey questions. But he said the more fundamental issue is methodological, telling the AP: “The ACS form wouldn’t provide a valid test of 2030 census operations.” In the same quote, he characterized the survey form as “a completely different animal.”
The AP report places the current practice test in context of efforts during Donald Trump’s first term to add a citizenship question to the 2020 census form and to pursue policies tied to citizenship status. It says that Trump instructed the Commerce Department last August to begin work on a new census that would exclude people in the U.S. illegally from the headcount, and it describes prior attempts: the citizenship-question bid for the 2020 census form was blocked by the U.S. Supreme Court, and both associated orders were rescinded after President Joe Biden took office in January 2021 before the 2020 census figures were released.
Republican lawmakers, the report says, have also introduced legislation that would exclude some noncitizens from apportionment figures, and several Republican state attorneys have filed federal lawsuits in Louisiana and Missouri seeking to add a citizenship question to the next census and exclude people in the U.S. illegally from the apportionment count.
Where the practice-test question intersects with apportionment is the constitutional framework described in the report. It says the Constitution’s 14th Amendment requires counting “the whole number of persons in each state” for apportionment, including congressional seats and Electoral College votes, and it describes the Census Bureau’s interpretation as counting anyone living in the U.S. regardless of legal status.