Michigan officials have dwindling faith in an accurate census count

The 2020 Census has had all the cards stacked against it: a pandemic that has caused confusion on where to respond and stopped most in-person events, a shortened timeline, an executive order calling for the exclusion of undocumented residents, an increased mistrust of government and so much more. So it should come as no surprise that many people, including local officials, are starting to lose faith in how accurate the 2020 Census results will be. 

Because of these factors and others, only 5 percent of local officials in Michigan surveyed have expressed strong confidence in the Census’ completeness or accuracy, according to the Spring 2020 Michigan Public Policy Survey (MPPS), which is conducted annually by the Center for Local, State, and Urban Policy (CLOSUP) at the University of Michigan’s Ford School of Public Policy.

Factors like the pandemic and the shortened timeline could affect the “really vital work of processing the data correctly and making sure people are counted in the right place and not counted more than once, according to  Terri Ann Lowenthal, a former staff director of the House census oversight subcommittee and nationally recognized expert on the U.S. Census. “it could severely undermine the accuracy of the final numbers.”

This year, the Michigan Public Policy Survey reached almost 1,350 leaders of counties, municipalities and townships in Michigan. Jurisdictions with larger minority populations and rural jurisdictions have less confidence in the Census. 

So, are miscounts really possible? 

According to Lowenthal, they absolutely are and “in the past sometimes a whole neighborhood was missed or a group facility was counted in the wrong county.”  

Brad Wisniewski, a Michigan enumerator, stated that there were some issues with the software when he went to follow up on responses because he got multiple cases where people who had already filled out the census were showing up as “nonresponsive.” He knew the residents he followed up with had filled it out because they were able to provide a screenshot or printout of their census response confirmation.

“There is some problem apparently with the software where people are completing the census and yet we would get a case generated sent out to our phones to go knock on that door,” Wisniewski explained. “I don’t know if that means that somebody is going to be counted twice or if there’s a possibility of  not being counted at all. But people who completed the census didn’t show up as completed and that is an issue.” 

There were 8 million duplicates in the 2010 Census according to the U.S. Census Bureau’s own evaluation.  Moreover, the 2010 Census undercounted renters by 1.1 percent. Homeowners were overcounted in both the 2000 and 2010 censuses, and 2.1 percent of the Black population and 1.5 percent of the Hispanic population were also undercounted.

“If we are relying on Michigan residents to complete the census online, this may be a problem in communities where internet access is a challenge,” said Debra Horner, a project manager for MPPS.

This could greatly affect Detroit’s overall numbers and the funding the city should receive being that Detroit is a predominantly Black city with more renters than homeowners who have limited access to the internet. 

What’s worrisome to many is that the first set of final numbers for the apportionment portion of the Census is due on Dec. 31 of each Census year. Under the pre-pandemic schedule, the major data collections operations would have been completed before July 31, giving the Bureau five months to complete the quality check, which spots falsification, poor-quality enumerator work, duplications and software errors. But now the Bureau is forced to do the quality check and processing work in only three months, giving the Bureau less time to catch and rectify these miscounts. 

According to a 2020 Census research report by the Urban Institute, “the overall accuracy of the national population count in 2020 could range from an undercount of 0.27 percent in the low-risk scenario to an undercount of 1.22 percent in the high-risk scenario.” While these percentages may seem small, said Diana Elliott, one of the authors of the study, 900,000 to  more than 4 million people could be missed.

This study was conducted in 2019 so the miscounts could possibly be higher. 

Optimism in Detroit’s response

On top of Detroit’s already low response rate, 50 percent as of Tuesday, “the U.S. Census Bureau still has to conduct special operations, in other words, count populations who don’t live in households or are experiencing homelessness which of course would affect a big city like Detroit,” Lowenthal explained. 

However, Victoria Kovari, who is leading Detroit’s 2020 Census initiative, has been working hard to get Detroiters to respond. She remains optimistic to make Detroit count. 

“Of the people who open doors, about 70 percent of them have said they already filled out the census and that has been consistent among any event we’re at —at food sites, at a rally or a store — over 70 percent of everybody that we meet say they’ve already filled it out, Kovari said. “So I’m convinced that even though our number says 49 percent, it’s closer to 70.”

The reason behind Kovari’s thinking is because of the number of vacancies in Detroit and all the people who have filled out the form without their Census ID — the Census Bureau has to recheck and verify those addresses before they can put them in the system which she’s hoping should bring up that percentage by Sept. 30, the Census deadline. 

However, given the importance of the Census for political representation, redistricting and allocation of federal funds and resources, ideally confidence in the accuracy should be high, but pessimism is increasing, according to the MPPS. 

Although Michigan’s 2020 self-response rate, 70.6  percent as of Monday, is above the national average of 64.2 percent, making Michigan fifth in the national ranking, the completeness and accuracy of the Census can and is still being questioned.