• Group homes

• Nursing facilities/skilled nursing facilities (nursing homes)

• Residential treatment facilities

• College/university or seminary student housing

• Religious group living quarters

• Correctional facilities

• In-patient hospice facilities

• Hospitals

• Workers’ group living quarters

• Job Corps centers or vocational training facilities

The Bureau uses multiple approaches to try to get an accurate enumeration of people living in group quarters including:

• Electronic Response Data Transfer (eResponse)

• In-person Interview

• Drop Off/Pick Up of Questionnaire

• Paper Response Data Collection

To read more about GQ enumeration, check out the following resources:

• 2020 Census Detailed Operational Plan for Group Quarters (GQ) Operation

• Group Quarters Enumeration (2018 Census Test)

• College Dorm Enumeration Fact Sheet (Census 2010)

• Ethnographic Study of the Group Quarters Population in the 2010 Census: Prisons for Adults and Juvenile Correctional Facilities (Michelle Inderbitzin and Anna Chan)

All address canvassers will carry an official badge with photo ID, a US Census Bureau bag, and a Census Bureau issued laptop.

CENSUS 🤓 ALERT: @uscensusbureau is sending workers to walk through neighborhoods this month through Oct. 18, to make sure it has a complete list of home addresses for #2020Census. Official canvassers carry 1) badge w photo ID, 2) bag & 3) laptop w logo👇https://t.co/AamUDRzoSd pic.twitter.com/sPRdylLdiT

— Hansi Lo Wang (@hansilowang) August 5, 2019

A nationwide map of neighborhoods where Census Bureau employees are canvassing can be found here. For more information on addressing canvassing, including a short video, check out the Bureau’s post about the process here.

What is your census story?  How does your story compare to the experiences of other Americans?  Check out the “NY Times: A Conversation on Race” short film series documenting the intersection between race/ethnicity and identity in America here.

Participating in the decennial census presents an important opportunity for us to stand up, be counted, and tell our stories. For many Americans, the story of where we come from and who we are is inextricably linked to race/ethnicity.

The U.S. Census Bureau asks individuals about race/ethnicity according to guidelines established by the Office of Management and Budget (OMB). Last updated in 1997, the race/ethnicity questions from Census 2010 and Census 2020 are shown below. There are two big picture takeaways:

1. The OMB definitions of race and ethnicity in Census 2020 will remain consistent with Census 2010. This includes Hispanic, Latino, or Spanish origin, which will remain an ethnicity, not a race.

2. While race/ethnicity definitions will remain consistent, Census 2020 will feature an important change.

   • Individuals who identify as White, Black/African American, and/or American Indian or Alaska Native will be asked to specifically identify their racial origins.

     • As outlined in the planned Census 2020 question, Black/African American individuals, for example, will be asked to print their specific origin (e.g. African American, Jamaican, Haitian, Nigerian, Ethiopian, Somali, etc.). Write in options will still be available for “Other Asian” and “Some other race” categories.

The story of race/ethnicity and Census 2020 also includes technical, behind-the-scene specifics. Following several years of testing, in 2017 the Bureau recommended that OMB adopt the following definitional changes for race/ethnicity:

1. Reassigning Hispanic, Latino, or Spanish origin to a racial category. This move would have ultimately collapsed the race/ethnicity questions into one question (i.e. race).

2. Include a new race category for individuals identifying as Middle East and North African (MENA).

The OMB did not respond by December 31, 2017 to these recommendations, effectively forcing the Bureau to retain the race/ethnicity definitions used in Census 2010.

Read more about the U.S. Census Bureau’s research on race/ethnicity here.

Details on how the U.S. Census Bureau will report race/ethnicity in the Public Law (PL) 94-171 file is available here in Appendix D.

A significant change from the 2010 Census however, is that for the first time in American history, married same-sex couples will be asked to identify themselves on the decennial census. Following the historic Obergefell v. Hodges case in 2015, which legalized same-sex marriage throughout the United States, the U.S. Census Bureau modified the American Community Survey (ACS) form (below) to count same-sex married households.

Prior to this landmark decision, in 2000 the Bush administration ordered the Bureau to officially reclassify individuals identifying as married same-sex couples as “unmarried partners”, citing the Defense of Marriage Act (DOMA). In 2010, the Obama administration reversed the earlier decision and allowed the Bureau to report self-identifying same-sex couples as married.

Accurately reporting the number of same-sex unmarried partners and spouses has been difficult. Upon evaluating responses to separate questions about respondent’s sex and their relationship to the head of household (e.g. spouse or unmarried partner), the Bureau discovered an error. Some opposite-sex couples were inadvertently checking the same sex as their spouse, meaning they were counted them a same-sex couple. To address this issue, the Bureau has modified the household relationship to count opposite and same-sex relationships separately. More information about this issue is available in this report from the Pew Research Center.

There is a unavoidable truth in American democracy, which is that you don’t count unless you’re counted. Recognizing this situation, many LGBTQ civil rights groups have advocated that the U.S. Census Bureau ask respondents about sexual orientation and gender identity. As the first decennial census to ask respondents about same-sex marriages, Census 2020 represents progress towards increasing visibility and addressing inequity for some in the LGBTQ community. As a self-identifying gay man who is married, I too celebrate this important step forward. However, it’s important to remember that these data will enumerate some, but not all gay and lesbian spouses, and by no means will be representative of the larger LGBTQ community.

Read more about same-sex marriages on the upcoming census here.

To provide some local context, consider the following language statistics for the Portland, OR metropolitan area in 2013-2017:

• Roughly 82% of Portland metro residents (1,835,000 of 2,240,000) only speak English at home;

• Almost one-half million (405,000) Portland metro residents speak a language other than English. The top 5 spoken languages, other than English, in the Portland metro area include:

  • 1. Spanish (46%)

  • 2. Russian, Polish, or Other (10%)

  • 3. Other Asian or Pacific (9%)

  • 4. Other Indo-European (8%)

  • 5. Vietnamese (6%)

Why are hard-to-count populations undercounted?

According to Dr. William O'Hare, there are multiple reasons why individuals are undercounted.  These include:  home address not included in census address roster, a fear of government and privacy, language issues, complex household relationships, and highly mobile populations with multiple addresses (e.g. renters) (presentation).  

What was the Census 2010 undercount?

The census undercount varies by race/ethnicity, age, housing tenure, immigrant status, and for urban and rural communities.  To provide some context, U.S. Census Bureau reported the following undercounts in Census 2010:

• Children (ages 0 to 4), -4.6%

  • More than 4 per 100 children, ages 0 to 4

• Black/African American, -2.1%

  • More than 2 per 100 Black/African-American individuals

• Black/African American men (ages 30-49), -10.1%

  • More than 10 per 100 Black/African-American men (ages 30-49)

• Renters, -1.1%

  • More than 1 per 100 renters

• More detail on the Census 2010 undercount (including by U.S. state) from the U.S. Census Bureau is available here and this is another great resource

What areas of my city/town are more like to be undercounted?

For more information, see this blogpost

What resources are available for addressing undercount in Census 2020?

• The Leadership Conference has great information about hard-to-count populations:

  • Hard to Count: Young Children and Their Communities (PDF)

    • Hard-to-Count State, City, and Congressional District Tables for Children under Age 5 here

  • Will You Count? Latinos in the 2020 Census (PDF)

    • View Hard-to-Count State, City, and Congressional District Tables for Latinos here

  • Will You Count? Asian Americans and Native Hawaiians and Pacific Islanders (NHPIs) in the 2020 Census (PDF)

    • View Hard-to-Count State, City, and Congressional District Tables for Asian Americans and NHPIs here

  • Will You Count? African Americans in the 2020 Census (PDF)

    • View Hard-to-Count State, City, and Congressional District Tables for African Americans here

  • Will You Count? American Indians and Alaska Natives in the 2020 Census (PDF)

    • View Hard-to-Count State, City, and Congressional District Tables for American Indians and Alaska Natives here

• Resources on immigrants

  • Welcoming America: How Municipalities Can Improve Census 2020 Accuracy and Address Immigrant Undercount (link)

  • Edward Kissam, "Differential Undercount of Mexican Immigrant Families in the U.S. Census (link)

• Resources on rural undercount

  • Census Project blogpost

• Resources on child undercount

  • Dr. William O'Hare (presentation)