If the goal of a census is to gather data about a country’s population, then the Trump administration’s recent procedural changes will set the 2020 census up for failure.
It is difficult to pinpoint the greatest concern for 2020, but a viable answer is certainly the return of a question about whether or not the respondent has U.S. citizenship. This has not appeared on a census since 1950.
Public reactions so far include concerns the census will be used to target undocumented immigrants, which will lead to under-counting when these immigrants opt not to respond out of fear of deportation.
Seventeen states and seven cities are now suing the Census Bureau and Commerce Department because they believe this question doesn't meet the Census Bureau's standards for new questions.
This is only the beginning.
Despite having identified changes in racial and ethnic categories, such as adding a Middle Eastern or North African category, as the best option for improving the accuracy of census data, a lack of support from the Office of Management and Budget has forced the Census Bureau to resort to other alternatives.
At a time when immigrants are facing particularly potent levels of discriminatory attitudes, census questions about national origin are becoming more specific. A Jan. 26 memo from the Census Bureau announced the introduction of write-in boxes under the black and white racial categories.
The intention of the write-in boxes is to gather more precise data about the countries from which current Americans and their ancestors emigrated, but there are doubts about whether this information will be accurate. For African American descendants of slaves, for example, it might be impossible to know which country their family originally came from.
"The decision to ignore years of research and the expert advice of scientists is a blow to science and the collection of the best data possible," said Arturo Vargas, executive director of the National Association of Latino Elected and Appointed Officials Educational Fund, in a written statement quoted by NPR.
There is a lot at stake for the 2020 census — $675 billion in federal funding, to be exact. These funds affect everything from health care to education to government representation, and inaccurate census data will lead to serious errors in allocating resources.
Some level of error in census data is inevitable. Even in 2010, the most accurate census to date, minorities were undercounted by at least 1.5 million.
More than in previous years, however, experts are concerned about the scale of inaccuracy that seems likely to plague the 2020 census.
Crucial field tests of collection methodologies and IT systems have been canceled. The director of the Census Bureau resigned in June 2017 and has not been replaced. The number of field workers who get in touch with unresponsive households has been cut by 200,000.
With so many changes unaccounted for, conducting a census under “such a level of uncertainty is literally unprecedented,” said Kenneth Prewitt, who directed the 2000 census, in a Mother Jones article written by Ari Berman.
The census will also now be entirely online, which the Census Bureau claims will save $5.2 billion, but the Editorial Board does not view these savings as beneficial or even legitimate. Who cares if an online census costs $5.2 billion less than previous versions when the results will be so inaccurate they’ll skew future budgets and policy?
If the 2020 census proceeds without any further changes, the American people — especially its most vulnerable populations — will suffer the consequences of an underprepared and politically irresponsible government institution.
