116 3rd St SE
Cedar Rapids, Iowa 52401
University of Iowa professor had close-up view of 2020 census trials, triumphs
‘It takes a whole decade to plan,’ Scientific Advisory Committee member says

Sep. 24, 2021 7:30 am, Updated: Sep. 24, 2021 7:52 am
IOWA CITY — It’s like launching an Apollo program every 10 years, and a University of Iowa professor who was an adviser to the U.S. Census Bureau thinks that despite political and pandemic challenges, the 2020 census landed successfully.
“It is a big deal,” Juan Pablo Hourcade said of the recently completed 2020 census that’s estimated to cost $14 billion, and “it takes a whole decade to plan.”
That speaks to it being “super important for the democratic processes,” said Hourcade, an associate professor of computer science at the UI.
Advertisement
The once-a-decade head count determines how seats in the U.S. House and Iowa Legislature are apportioned and how state and federal funds are allocated.
“It affects decision-making by all kinds of actors from government entities to companies that are figuring out where to put their next business,” he said.
Before joining the UI in January 2006, Hourcade worked in the Census Bureau’s research division. He’s now finishing his second term on the Census Scientific Advisory Committee.
As the name implies, the committee advises the bureau when it’s considering changes in its processes. One change in 2020 was the use of differential privacy that allowed the bureau to share information about a data set without disclosing information about individuals in that data set.
That comes into play in small data sets, such as small towns and neighborhoods where census information combined with other available data could identify individuals.
“Facebook and Google probably know a lot more about you than the Census Bureau,” Hourcade said, but, by law, the bureau has to protect the privacy of respondents. “So they’ve taken it very seriously.”
Despite the challenges inherent in trying to count every resident, the 2020 census faced additional obstacles.
“I think anything, everything that could have gone wrong for this decennial went wrong,” Hourcade said. In addition to the challenges of new technology and doing much of the census online, there were political issues, particularly the Trump administration’s attempts to include a citizenship question.
“And the timing of the pandemic couldn’t have been worse,” Hourcade said, noting that Census Day was supposed to be April 1, about two weeks after schools, businesses and industries started closing down because of COVID-19.
One of the consequences of the pandemic was people living in places they normally wouldn’t have been living, Hourcade said. College students moved home, and some people moved to isolated areas hoping to escape the pandemic. Some of them may stay in the location where they were counted. Others will move back to their previous addresses.
“So that has all kinds of implications for funding that’s based on census counts,” such as decisions where to put a road or build a school, he said.
His time on the advisory committee is ending, so Hourcade can only speculate on how the 2030 census may be different. He expects there will be greater use of technology, and some proposed changes are floating around.
It’s too soon to know what changes will be made because, in part, “it’s also true that, in general, government agencies don’t like dramatic change.”
Comments: (319) 398-8375; james.lynch@thegazette.com
A census form is seen in March 2020 in Idaho. (Idaho Post-Register via AP)
Juan Pablo Hourcade, an associate professor of computer science at the University of Iowa, was part of the Census Scientific Advisory Committee that was part of the 2020 census. (Submitted)