Mayor Wu challenges 2020 US Census count of Boston

Sep 13, 2022

Mayor Michelle Wu today announced that the City of Boston is challenging the United States Census Bureau’s 2020 Census Population count of Boston’s total population. Based on research by the Boston Planning and Development Agency (BPDA), the population count of 675,647 for Boston undercounts significant segments of the City’s college and university student and foreign-born population, and individuals in correctional facilities. Mayor Wu also announced that the City of Boston has applied for a review of the City’s group quarters count through the Post-Census Group Quarters Review (PCGQR) program.

The City’s challenge centers on population count concerns, count case coverage issues, and objections to the racial and ethnic classification changes for the 2020 census. An accurate count of Boston’s population will help inform the needs of Boston’s residents as well as guide the City’s planning for allocation of resources across communities.

“Boston deserves an accurate census count across every neighborhood and community,” said Mayor Michelle Wu. “This count is the foundation to assess the needs of all of our communities, ensuring that Boston receives crucial federal resources, and it should reflect our full numbers.”

"Our Census count is immensely consequential for determining the allocation of public resources, federal grants and assistance programs, and understanding the needs of our communities. I am appreciative of the administration's formal submission of a challenge to our 2020 Census results through Census Bureau programs because we must set the record straight," said Councilor Liz Breadon. "I am particularly concerned by Allston’s reported 5.9 percent loss in total population and 40 percent decline in group quarters population, severely impacted by the early pandemic evacuation of colleges and universities. Correcting the count will help inform the needs of our communities and ensure every Bostonian is reflected."

“I’m relieved that we are challenging the 2020 Census results. My office met with the Administration early on to sound the alarm on how an undercount would unjustifiably lead to our receipt of fewer federal dollars to tackle issues like housing & education. It is vitally important that we are getting accurate census data to help ensure we measure what matters, and what matters most is every person. Undercounting our immigrant population, our students, and our incarcerated residents is a civil rights issue that we must forcefully address. An accurate count of Boston residents will ensure that all our communities, especially our Black & Brown communities that have suffered from decades of disinvestment, get the resources and attention needed to thrive,” said Councilor Ruthzee Louijeune. “Additionally, the U.S. Census Bureau must respect how people identify themselves racially and ethnically. The Census must consider simplifying categorization, particularly for Latino, Brazilian, Middle Eastern and North African respondents, to correctly capture our demographics.”

The City’s challenge centers on concerns related to the group quarters population count, household population count, and objections to the racial and ethnic classification of Boston’s population in the 2020 census; all three of which heavily impacted an accurate and reflective count of Boston’s population.

Group Quarters Population

The City of Boston’s concerns regarding the group quarters population count from the April 1, 2020 Census are due to undercount issues amongst colleges and universities and with Boston’s two correctional facilities.

The COVID-19 pandemic in mid-March 2020 led colleges and universities to evacuate their campuses and move to remote operations before the April count. Data collected from colleges and universities under the University Accountability Ordinance for Fall 2019 show that approximately 6,000 additional students were not accounted for by the 2020 Census redistricting data.

The Suffolk County Department of Corrections records show that their two correctional facilities on April 1, 2020 housed approximately 500 additional residents than reported by the 2020 Census redistricting data.

With an undercount of the group quarters count by approximately 6,500 residents, the City has applied for a review of Boston’s group quarters count through the Post-Census Group Quarters Review (PCGQR) program. Additionally, the City has provided the Census Bureau with administrative records from educational and correctional institutions to support these undercount issues.

Household Population

The City of Boston also has concerns related to the Household Population Count based on low self-response rates that were exacerbated by the pandemic disruptions, an issue that the Census’ operation has failed to adequately address. In the 2010 census, all of Boston’s census tracts' self-response rate exceeded 50 percent. However, in 2020, 29 census tracts – 15 percent of populated Boston tracts – had a self-response rate between 30.0 percent and 49.8 percent.

Some of these census tracts with lower response rates either have a large share of off-campus students or foreign-born residents. Off-campus students may have left the city due to universities shifting to remote operations, which would have led to them not being able to participate in the Census’ Non-Response Follow-up activity. Other census tracts with lower response rates had larger shares of foreign born residents. Issues such as language barriers and government mistrust, in particular a citizenship question and prevalent anti-immigrant sentiment when count was administered, may have resulted in an undercount.

Currently, the Census Count Question Resolution Program (January 3, 2022 - June 30, 2023) only accepts 2020 census challenges for boundary issues and census processing errors which exclude valid housing and associated population data. Additionally, the Census CQR operation provides no avenue to review an increase in apparent housing unit vacancies. Vacancy rates of housing units increased significantly in Allston, Chinatown, Fenway, Longwood, and South Boston Waterfront, neighborhoods that also have significantly lower Census self-response rates. Thus, pandemic disruptions could have led to many of Boston’s housing units being incorrectly identified as vacant, resulting in an undercount of the City’s population.

Race Classification

In 2020, the Census reclassified the collection and processing of race and ethnicity data which led to large increases in the “some other race” and “two more races” categories independent of actual demographic or cultural changes. Following this self-reported data on the population’s race and Hispanic origin, the Census Bureau recategorizes this information following prescribed definitions developed in 1997 by the Office of Management and Budget (OMB). This recent race and ethnicity reclassification, paired with the Census’s recategorization process, heavily impacted Boston’s data.

As a result of this Census coding, 76 percent of Hispanics in Boston chose (or were assigned) the “some other race” category, either by itself or in addition to other racial categories, up from 45 percent of Hispanics in 2010. Additionally, respondents listing Brazilian or Cape Verdean origin were reassigned by the U.S. Census Bureau to the “some other race” and respondents that stated a Middle Eastern or North African origin were assigned to the White category respectively.

Implications of the Census’s (Re)Classification include the following:

• Multiracial or “other” categories are too heterogeneous to be lumped together for data analysis or policy/program creation.
• Black and White populations appear smaller while multiracial and “other” populations appear larger. Using single race alone categories may underestimate the number of people who identify as White or Black.
• A false impression may result that changes are a result of actual demographic or cultural changes in the population not, in great part, a result of reclassification.
• Respondents who thought their self-identification was recorded by the Census may have been misled.
• “Origin” and “race” are conflated as respondents listing certain origins are assigned to certain races. For example, Hispanic, Brazilian, and Cape Verdean origins are classified as some other race.

To address these implications, an alternative combined single race/ethnicity was created, tested and recommended in 2015 by the Census Bureau that allows for “Hispanic” and “Middle Eastern and North African” as distinct choices. In 2015, the Census Bureau’s National Content Test Race and Ethnicity Report found that 70% of Hispanics only chose to mark the Hispanic box with the combined question, however this recommended change was not approved by OMB. The City of Boston agrees with the U.S Census Bureau that an alternative single race/ethnicity question would be an overall improvement and will advocate for its adoption in future census and the American Community Survey.

However, the City also urges the Census Bureau to implement two changes on top of adopting a recommended single race/ethnicity question with the aim of improved data collection. First, the City believes that the Census needs to commit to respecting self-identification of respondents since the currently proposed change of a combined alternative would not alter the OMB’s prescribed definition classification. Secondly, the City encourages the Census to facilitate discussion regarding the Bureau’s recategorization, which is rooted in OMB’s prescribed definitions from 1997, with the goal of reflecting real demographic processes.