Some Fear the Census May Have Been Affected by the Pandemic and Political Turmoil

U.S.|Some fear the pandemic and political turmoil may have affected census the count.

• Aug. 12, 2021, 4:48 p.m. ET

Perhaps no census has been as fraught as the one that led to the data released on Thursday, a count pummeled by the pandemic and hobbled by a White House that sought to use it as a tool to permanently shift the balance of national political power.

Theoretically, those crises could have opened big holes in the data the Census Bureau gathered last year, as some people shied from being counted and others refused to tell the government everything it wanted to know. How big those holes are, and how they were plugged, won’t be known until the bureau publishes the results of its own quality check later this year.

But one longtime expert preaches caution. “Early returns on every census cause people to jump to conclusions that may not be supported on further research,” Steve Jost, a census consultant and former bureau official, said in an interview.

In this case, Mr. Jost said, the numbers could even show that a census many expected to be wildly inaccurate was actually pretty close to the mark.

There was plenty to worry about. The national count unfolded amid a contentious effort by the Trump administration to exclude from the census count millions of people living in the county without authorization, despite a constitutional mandate to count everyone.

Not until July 2020 did it reveal why: President Donald J. Trump wanted to exclude them from population totals used to divvy up seats in the House of Representatives, creating an older, whiter and presumably more Republican base for reapportionment.

That effort failed, but one early indicator suggests the anti-immigrant crusade may have scared some ethnic groups: The share of households that declined to answer at least one of the nine questions on the 2020 census form was exponentially higher than in the last census in 2010. And questions about race and ethnicity were the ones most likely to be skipped.

On the other hand, the early results show a much larger move to the cities than many expected, and a substantial jump in the Hispanic population — all suggesting that maybe ethnic and racial groups were not as deterred as was thought likely.

Experts also fretted after the coronavirus shut down the nation in April 2020, just as the nationwide tally was getting underway. The crucial final phase of the census, in which door-knockers tracked down the millions who had not voluntarily filled out a form, was delayed to autumn — peak hurricane season, when storms battered much of the South. Frightened residents refused to open doors to census-takers; census-takers proved harder to recruit and quit more often for fear of getting sick. In a final, frantic push, the bureau literally airlifted its best door-knockers into its hardest-to-count regions, a logistical move reminiscent of an army campaign.

That led many experts to worry that the bureau would miss counting so many households that it would have to fill in data on huge swaths of some areas by making statistical educated guesses about who lived there. But in fact, those guesses, called count imputations, are actually lower than in 2010, because the bureau sifted through federal records to identify who was in those missed households.

In theory, that could lead to a more accurate census than anyone expected — if the records were accurate. That won’t be clear until the bureau issues its report card. In the meantime, one expert suggests that people simply be thankful that, in a year of historic social and political upheaval, a decent count happened at all.

“The mere fact that we’re getting the data and beginning to get back on track is a big accomplishment,” said Margo J. Anderson, a historian and census expert at the University of Wisconsin-Milwaukee.