It’s a question that used to be on the national census every decade: whether you were a citizen of the United States.
But the Justice Department’s request to return it to the 2020 census for all respondents has unsettled demographic experts as well as advocates of voting rights and immigrants, who say it could lead Hispanic people to avoid being counted. Are they overreacting to a simple question?
We can’t say at this point what the electoral consequences would be, but it’s likely to lead to undercounting. The Census Bureau itself estimates that the 2010 census failed to find 1.5 percent of the Hispanic population. Research conducted that year suggests that Hispanic trust in the census may have been undermined. And from the start of his candidacy up through his reported vulgar remarks last week about Haiti and African countries, President Trump has been fanning anti-immigrant sentiment nationwide.
The Justice Department says it wants to add the question to aid its defense of Section 2 of the Voting Rights Act (which prevents the dilution of minority populations so their power cannot be weakened).
Supporters of adding the question say it shouldn’t be a problem because the citizenship question has since 2000 been asked on a smaller, recurring census-sponsored survey, the American Community Survey, and because the anonymity protections are strong. But the trouble is that today, everything even remotely political has become a battle over what it means to be an American.
Responding to both the census and the A.C.S. is the law of the land — you must do it or you could be fined. The data collected determines how a lot of money is allocated, as well as the allocation of House seats (and therefore Electoral College votes).
The more people who fail to respond, the more concern there is that we are missing some groups of people more than others, and that the failure to return the form among these group members is not random.
The government dedicates tremendous resources to reminding people to return their census form and even sends people to the doors of households from which no form has been filed. Mostly, it gets results. But if the reason for not filling it out is distrust of government, additional efforts at compliance by government might fall flat.
For the 2010 census, the Spanish-language television network Telemundo sought to improve census participation by writing a story line into one of its most popular telenovelas, “Más Sabe El Diablo.” In it, the character Perla meets a Latino census worker at her father’s empanada stand and is encouraged to apply for a job with the census. The plot shows Perla being trained and learning about why the census asks the questions it does and how it safeguards confidentiality. The idea was that a popular character on a TV show could do more to assuage the fears of a community than the government could.
Matthew Trujillo, currently at the Robert Wood Johnson Foundation, and Elizabeth Paluck, a Princeton University psychologist and recent MacArthur Award winner, studied the network’s efforts. Mr. Trujillo and Ms. Paluck asked 121 Spanish-speaking Latino adults across three states to watch either a four-minute clip from “Más Sabe” that showed Perla talking about the importance of completing the census or one that included Perla talking only about family. One of the states was Arizona, which had just passed a law requiring police officers, in the course of an unrelated investigation, to investigate a “reasonable suspicion” that a person was in the country illegally.
Subjects in the experiment completed a survey before and after watching one of the clips. (Which one they watched was determined at random.) Upon leaving the lab, they were able to take a flier about the census and choose either a generic “Latino Pride” or census-specific “Be Counted” sticker.
The results of the test showed that people who saw the census story line were more likely to have positive attitudes toward the government generally — unless they lived in Arizona. Latino residents there, under threat from the newly passed law, were not moved by Perla’s story line.
The study also revealed that, on average, seeing Perla’s experience with the census made people, including those in Arizona, more aware of it. They were also more curious: 86 percent took a flier about the census as they left, compared with 69 percent of people who saw the other clip. Finally, the census clip prompted more people to take and wear the “Be Counted” sticker as they left — if they lived in Texas or New Jersey. In Arizona, people in both groups avoided the “Latino Pride” sticker.
The results suggest that if Latinos in the United States feel generally threatened by the Trump administration, it may be hard to persuade them to overcome their negative views of government and return the 2020 census.
California officials are so worried about Latino nonparticipation — and the potential loss of a seat in Congress and billions of federal dollars — that they are discussing aggressive multilingual advertising campaigns.
In the 2010 Telemundo study, it mattered a bit how much people liked Perla as a character. This is both good news and bad news for the Census Bureau as it faces 2020. With TV content booming, there is no shortage of popular characters who could be seen talking about the census.
On the other hand, today’s networks may be reluctant to participate the way Telemundo did in 2010. Given the current climate, even they may be unsure of the government’s intentions, particularly as it relates to those who are undocumented or whose national origin may not be in keeping with the president’s view of being American.
Lynn Vavreck, a professor of political science at U.C.L.A., is a co-author of the coming “Identity Crisis: The 2016 Presidential Campaign and the Battle for the Meaning of America.” Follow her on Twitter at @vavreck.