Skip Navigation

Citizenship Questions on the Census Have No Historical Pedigree

Never in the census’ 230-year history has the decennial questionnaire asked for the citizenship status of everyone in the country.

The article first appeared in the Los Angeles Times.

Later this month, the Supreme Court will take up Commerce Secret­ary Wilbur Ross’ contro­ver­sial decision to collect the citizen­ship status of every­one in the coun­try in the 2020 census. In court filings, the Trump admin­is­tra­tion has defen­ded the citizen­ship ques­tion as normal and inof­fens­ive, part of an “unbroken tradi­tion” whose “pedi­gree dates back nearly 200 years.”

But look closer, as we did, and history tells a differ­ent story. Over the last year, we pored over archival mater­ial includ­ing 19th century census-taker instruc­tions and decades-old papers on govern­ment stat­ist­ics. We discovered that the Trump admin­is­tra­tion’s history is mislead­ing, where it’s not outright false.

Never in the census’ 230-year history has the decen­nial ques­tion­naire asked for the citizen­ship status of every­one in the coun­try. In real­ity, when a citizen­ship ques­tion was asked at all, it was direc­ted to small segments of the popu­la­tion, such as foreign-born men 21 or older (1890–1910) or foreign-born people (1930–1950), mainly to gauge how well they were assim­il­at­ing.

There’s no justi­fic­a­tion for a citizen­ship ques­tion in 2020.

What’s more, these ques­tions had a bad track record. They were part of an approach that the Census Bureau ulti­mately rejec­ted as incom­pat­ible with its consti­tu­tional duty to count every person.

History isn’t on Ross’ side. But one federal court already bought the admin­is­tra­tion’s “we’ve done this before” narrat­ive, so it’s vital to set the record straight before the nation’s highest court is simil­arly misled and the census is imperiled.

Before 1960, the Census Bureau tried to pursue two goals at once with the decen­nial census: to count every­one and to collect other inform­a­tion the govern­ment needs, such as mortal­ity or employ­ment stat­ist­ics. This led to major prob­lems. These censuses were tremend­ously long, enorm­ously expens­ive, often took years to complete and — most import­ant — were plagued by inac­cur­acy. Census takers were bogged down seek­ing out answers to ques­tions such as, is anyone in the house­hold “a pris­oner, convict, home­less child, or pauper.”

By the 1950s, the Census Bureau was able to apply emer­ging tech­no­logy and new stat­ist­ical meth­ods to eval­u­ate how accur­ate its popu­la­tion count had been. It was deeply flawed. Later analyses put the under­count of the 1950 census at 5 million to 5.5 million people, includ­ing a 12% to 13% under­count of people that the bureau labeled “nonwhites.” The bureau also learned that it could collect better data more cheaply by asking most ques­tions only to portions of the popu­la­tion and extra­pol­at­ing from there. 

So the Census Bureau slimmed down its head­count form to a hand­ful of ques­tions. Everything else it put on a longer survey that went out to small sample groups. It has used that model ever since.

Citizen­ship-related ques­tions were among the first to go. The bureau took them off the census completely in 1960 (except for New York and Puerto Rico, which had unusual redis­trict­ing needs). And, from 1970 onward, citizen­ship-related ques­tions were only ever on the smal­ler surveys, like the long form (through 2000) or the Amer­ican Community Survey (from 2005 on). But census offi­cials never let similar ques­tions back on the head­count form, because they knew that would hurt the count.

The bureau has long recog­nized that anti-immig­rant envir­on­ments make citizen­ship ques­tions prob­lem­atic. In the 1970s, as waves of immig­rants from Latin Amer­ica and Asia arrived here, lawmakers and activ­ists began search­ing for ways to limit their polit­ical power. In the lead-up to the 1980 and 1990 censuses, anti-immig­rant groups and their allies pushed the bureau to exclude immig­rants in this coun­try illeg­ally from the head count. To do so, the bureau would have to try to collect every­one’s citizen­ship status.

The bureau repeatedly resisted these pres­sures. Census Bureau director John Keane warned the Senate in 1985 that citizen­ship ques­tions posed a huge risk because “the Census Bureau could be perceived as an enforce­ment agency.” Efforts to gather this inform­a­tion, he said, would lead people to refuse to parti­cip­ate in the census for fear that their answers could be used against them.

In the lead-up to the 2020 census, Pres­id­ent Trump’s rhet­oric and policies have already created head­winds for an accur­ate head­count. Bureau research in 2018 found that 35% of Asian respond­ents, 34% of black/African Amer­ican respond­ents, and 32% of Latino respond­ents feared inform­a­tion about them would be illeg­ally shared with other govern­ment agen­cies (as compared with 24% of respond­ents over­all). Adding a citizen­ship ques­tion will not help matters.

When the issue lands before the Supreme Court on April 23, the history of the census will be cent­ral to the argu­ments. That history should make it clear: There’s no justi­fic­a­tion for a citizen­ship ques­tion in 2020.

Brianna Cea is a research and program assist­ant in the demo­cracy program at the Bren­nan Center for Justice at the New York Univer­sity School of Law. Thomas Wolf is coun­sel in the same program.