Pandemic, hurricane, fire and politics seemed a recipe for a wrong count. But the preliminary investigation did not find any clear indication of lapse.
WASHINGTON – The problem-plagued 2020 census won a vote of confidence on Monday from experts who said the count appeared accurate enough for its constitutional purpose: to reallocate 435 seats in the House of Representatives.
But experts, eminent statisticians who provided access to census procedures, narrowed their search to overall national tallies and counts in 50 states and the District of Columbia. He said much more study would be needed to assess the totals of local populations and the reliability of characteristics such as race and ethnicity, which are important parts of every census.
The findings were presented in a 59-page analysis of the 2020 count commissioned and reviewed by the American Statistical Association. A task force set up by the group recommended the investigation in October – and the Census Bureau agreed – amid growing concerns about the accuracy of counting among businesses, state and local governments and others who need to plan for the future. rely on the census.
By any measure, the census faced unprecedented hurdles: the coronavirus pandemic forced the counting to stop as it did in April 2020, forcing the bureau to extend its work by nearly two months . Last autumn, western wildfires and coastal storms made the Census Bureau’s job just as door knockers were heading out to survey the millions of homes that hadn’t filled out forms.
In addition, the Trump administration pushed the deadline to finish the count, throwing a political chasm into an already entrenched tally. The flood of problems caused many experts, including some senior Census Bureau officials, to worry that the final count would be so flawed as to be unusable.
The report released on Tuesday indicated, at least, that those fears were overcame. Despite being written in the cautious language of scientists, it said there was no persuasive evidence that either the nationwide count or the individual state count was less reliable than the 2010 census results, widely considered the most accurate to date. appreciated as
John H. Thompson, former director of the Census Bureau, said, “Career workers took this census in the most difficult circumstances ever, and they did their best.”
The analysis examined 10 important tasks in the 2020 count, from compiling a national list of addresses to be surveyed to creating statistical estimates of people living in households who refused to fill out census forms.
The study uncovered wide variations between states in the results of those operations. For example, one in 20 Louisiana homes were linked not by answers to written surveys, but by data from federal records showing who lived there. In Hawaii, the figure was one in 60.
Logic says the aerial count, with more individual responses, will be more reliable than Louisiana’s. But the report cautioned against making hasty decisions: using more records to count homes, for example, could reduce the need to make less accurate statistical estimates about them.
Those statistical estimates, called imputations, were among the most closely watched and controversial aspects of the 2020 census. In the last census in 2010, only .4 percent of households were counted by stigma. Many experts forecast a sharp jump in that share – and a less accurate census – as the troubled 2020 count forced the bureau to speculate about the composition of more homes.
In fact, the results were more mixed, the study suggested. Overall, the share of the population count that was imposed came down slightly. The Bureau, on the other hand, was forced more often to guess who lived in group quarters such as college dormitories, long-term care facilities, and prisons. In part, this was because the pandemic forced many university students to return home, making it difficult to count them in dorms or apartments, where they were usually taller.