Thanks to a Supreme Court decision in June, the 2020 Census won’t include a question about citizenship. But the hangover from the battle over a citizenship question leaves the upcoming Census uniquely challenged, according to a panel of experts.
NEW YORK — The panel, speaking at a speaking at a recent Reuters Newsmaker event on September 5 at Reuters’ New York headquarters, was introduced by Stephen Adler, Editor-in-Chief and President of Reuters, and moderated by Lauren LaCapra, Reuters’ Deputy Finance Editor, Americas.
While the 2020 Census has already seen its share of controversy, the 2030 Census may be even more surprising, said panelist Kenneth Prewitt, a Carnegie Professor at Columbia University and a former Director of the U.S. Census Bureau. “The design we want for 2030 is that the Census itself, the decennial Census, is nothing but counting the number of people who live in each state,” said Prewitt. “Nothing else. Not their race, not their ethnicity, not their height, not anything about them.”
Census in the spotlight
There is tremendous energy — both positive and negative — around the 2020 Census, the panelists agreed. The furor over a potential citizenship question has affected perceptions of the Census, and that could make it even harder to get an accurate count. “This country has never had a Census in which it didn’t have the general goodwill of the public and did not have the full support of leadership in the country,” explained Prewitt. “Our leadership has always wanted a good Census. This Census is not being launched with that trust.”
Another panelist, U.S. Rep. Carolyn Maloney (D-NY), was even more direct when referring to the current administration. “They’ve done everything to put fear into people,” Maloney said. “Making anti-immigrant, anti-certain population comments, and really increasing the activities of ICE.” The upshot, she says, is that “many dreamers and green card holders may be frightened to fill out the questionnaire.”
Panelist Justin Levitt, an Associate Dean for Research and professor at Loyola Law School, noted that there are very strong legal controls over the use of Census information, and that “there used to be better political controls.” In some places, a lot of effort is being made to help people feel comfortable answering Census questions, and in others, he said, not as much is being done. “You may see that reflected in the Census count,” he said.
But the possibility of a citizenship question highlighted the importance of the Census, and energized those who want a complete count, said Maloney. In New York and California, she said, millions of dollars have been allocated to reach out to community leaders and others that can help produce a more accurate count in hard-to-count areas. She referred to “a huge army of people” being hired as enumerators — to go door-to-door to get responses from those who don’t fill out the Census.
Levitt also pointed out that social media platforms have great potential to assist with the Census, should they choose to do so. “The risk is that they’re not as big a megaphone in the undercounted communities,” he added.
A radical change for 2030
Prewitt said he doesn’t expect the 2030 Census to look much like the 2020 Census. “An odd positive fallout [from the fight over a citizenship question] is that we don’t want to do it this way again,” he noted. The current form of Census-taking is necessary for counting people and deciding states’ representations in the electoral college, he said, adding that, as many of the participants noted, the American Community Survey, which is conducted on a rolling basis, already provides a wealth of detailed information. “That’s not politicized,” said Prewitt.
Levitt agreed, adding that the government already has data on sensitive topics such as individual disabilities, mortgages, and salary information “all collected in ways to minimize the intrusiveness of gathering the information and put people at ease in answering it.”
Indeed, most big economic data doesn’t come from the Census, Prewitt explained. Instead, researchers and policymakers increasingly rely on swipe data, consumer data, and banking data. “That stuff flows in every day — it’s recorded every day,” he said.
By 2030, said Prewitt, the U.S. Census Bureau will become more of a curator of data than a collector of it. The Bureau will concentrate more on the analysis of private data and will tell the providers of that data which parameters need to be followed if they want their data used by the Census Bureau.
That’s similar to a model already in use at Nielsen, said another panelist, Christine Pierce, Senior Vice President in Data Science Audience Measurement at Nielsen. She says her company uses data from its clients, who participate because they want to be measured. Then, Nielsen designs surveys to gather additional information. The next step is to make sure their results are fully representative of the relevant populations. If it’s not, they’ll calibrate it. But along the way, their measurements are benchmarked to the Census.
“In all of our operations, every day, we are using some aspect of the Census,” said Pierce. “If it’s off, it’s off for a decade.”