Recommendations of the Census Bureau's African American Advisory Committee

This document includes the October 2003 recommendation of the Census Bureau's African American Advisory Committee that the Bureau count people in prison at addresses outside of the facility and the Census Bureau's January 2004 response.

This document, previously published on the Census Bureau's website at, disappeared sometime in 2007 and is reprinted here as a public service.

For an analysis of the Census Bureau's response, see Memo from Kirsten Levingston to John Flateu [PDF] and section IV A of our report "Why the Census Bureau can and must start collecting the home addresses of incarcerated people".

October 1-3, 2003

The Census Advisory Committee on the African American Population made the following recommendations to the Director, U.S. Census Bureau, during its meeting on October 1-3, 2003. Comments reflecting the responses and actions taken or to be taken by the Census Bureau accompany each recommendation.

Recommendation 1

Retain the “Some Other Race” Category

“We recommend the retention of the ‘some other race’ category on the race question. Since this group of 15 million persons comprises many immigrants who do not identify with the traditional racial categories, we also recommend that the Bureau develop a research agenda that enhances our understanding of their preferred racial/ethnic identities and their social and economic characteristics (such as nationality, language, education, income, year of immigration, employment status, etc.)—and that expands upon the pioneering studies by Professor John Logan at SUNY University of Albany.”

Census Bureau Response

Although we have used the “Some other race” (SOR) category in the past, that option is not one of the required Office of Management and Budget (OMB) race categories. To make census data consistent with other federal agencies’ data, the Census Bureau must reassign responses in the SOR category to one of the standard OMB race categories. Further, data for the SOR category are only shown in decennial census data products. For all other Census Bureau purposes, such as estimates, projections, and surveys, these responses must be reassigned into one of the OMB categories. These results are sometimes used to calculate indicators, such as cancer incidence rates for various racial and ethnic populations.

The SOR response category was not included in the response categories to the question on race for the 2004 Census Test, and we have not yet determined the wording and response categories for the 2010 census. In addition, we have not yet analyzed the 2003 National Census Test data to specifically determine the effect on the African American population of dropping the write-in line with the SOR category, and this is one of the areas we are continuing to examine. However, results from the 2003 National Census Test indicated that dropping the SOR category decreased SOR reporting by Hispanics by about 18 percentage points, and increased (by 6 percentage points) race nonresponse for Hispanics. Furthermore, the increase in Hispanic nonresponse was offset by lower imputation rates (resulting from a reduction in SOR reporting). This combination of factors leads to a substantial net improvement in the proportion of Hispanics reporting in the five standard OMB categories for the race question.

The Census Bureau plans to use these results, as well as the results of the tests planned as part of the 2004 Census Test, the 2005 National Content Survey, and additional cognitive and focus group tests to develop optimal wording for the race and Hispanic-origin questions and response categories. Our decision on the design of these questions for the 2010 census will be based on a careful analysis of all test results in order to select the wording that will provide the most complete and accurate data to meet the myriad needs for such information.

Recommendation 2

Add an “Other Black” Category to Questionnaire and Test in 2005

“We support the use of examples and instructions in the 2003 National Census Test in order to improve the enumeration of other Hispanics, Asians, Native Hawaiians and Pacific Islanders. Therefore, we strongly recommend that a comparable approach for Blacks and African Americans be provided in the 2005 Test by adding an ‘other Black’ category, which would include these examples: Dominicans, Ethiopians, Haitians, Jamaicans, Nigerians, and Trinidadians.”

Census Bureau Response

As we continue research and testing of the content and design of the 2010 census questionnaire, we are cognizant that we must balance the need to obtain an accurate count of a diverse and rapidly changing population with the practical considerations of questionnaire size and respondent burden. Currently, we are reviewing the multiple requests made by our various Advisory Committees to add more response categories, examples, and instructions to the race question. We are also aware of the OMB guidelines that emphasize that the number of categories should be kept to a manageable size, as determined by statistical concerns and data needs.

Recommendation 3

Retain the Revisions used in 2003 Census Test on the Hispanic Question

“We recommend retention of the revisions (such as use of commas, the addition of ‘origin’, and the listing of examples for ‘other Hispanics’) that were applied in the 2003 Census Test in order to more accurately identify Hispanic groups.”

Census Bureau Response

The revisions to the Hispanic-origin question noted in your recommendation are included on the final questionnaire approved by the OMB for use in the 2004 Census Test. As stated in our response to Recommendation 1, we have not completed our research aimed at developing optimal race and Hispanic-origin questions for the 2010 census. We will use the results of future tests, combined with the data we have already gathered, to determine the final wording for these questions. As required by law, we will submit the planned wording of the questions to Congress by April 1, 2008.

Recommendation 4

Retain the Space Provided to Identify Tribes in the 2003 Census Test

“We recommend retention of the space provided in the 2003 Census Test to identify the ‘tribes’ for Americans Indians.”

Census Bureau Response

As in the 2000 census questionnaire and the 2003 National Census Test questionnaire, the 2004 Census Test includes space to write the name of each respondent’s enrolled or principal tribe. Although we have not completed our research aimed at developing optimal race and Hispanic-origin questions for the 2010 census, we plan to retain the space for “tribes” in the American Indian/Alaska Native response category of the race question.

Recommendation 5

Sequence the NHOPI to Follow the AIAN as in the 2003 Census Test

“We recommend the sequencing in the 2003 Census Test that allows the Native Hawaiians and Pacific Islanders to follow American Indians or Alaskan Natives.”

Census Bureau Response

We are considering a wide range of questionnaire design options for future research and testing, including a variety of race response category sequencing options in the 2005 National Content Survey. We will keep the Committee apprised of our progress as we develop the final panel designs for the 2005 National Content Survey.

Recommendation 6

Do Not add “Caucasian” to the White Response Category

“We do not recommend any tests that add ‘Caucasian’ to the White response category.”

Census Bureau Response

We conducted cognitive research in the summer of 2003 to determine the feasibility of including “Caucasian” as part of the “White” category to the question on race. The resulting data are inconsistent. Some individuals view the term "Caucasian" as “referring to people of Northern European descent” while others thought that "Caucasian" was a more general term. In contrast, some focus group participants viewed the term "White" as a general term, whereas other respondents felt that "White" referred exclusively to descendants of Europeans (Jiles et al. 2003).

Given these mixed results, we may include "Caucasian" as part of the “White” response category to the question on race in the 2005 Census Test, in our continuing efforts to gather data about terminology that is meaningful to the greatest number of respondents.

Recommendation 7

Opposition to Research for Testing of the Shortened Questions on Hispanic Origin and Race

“We strongly oppose any research resources being provided for testing of the shortened questions on Hispanic Origin and Race, because we believe that such efforts would undermine the years of progress in improving this nation’s understanding of the diversity of the Hispanic populations and various racial groups.”

Census Bureau Response

We are currently considering including a test of shortened race, Hispanic-origin, and ancestry questions in the 2005 National Content Survey. This is due to two issues that have been raised by our stakeholders. First, some ethnic groups that were not listed in the race and Hispanic- origin questions on previous questionnaires have asked why they were not specifically included. They believe a shortened format would result in a more balanced treatment for all population groups for the three concepts of race, Hispanic-origin, and ancestry. Secondly, because these questions overlap, some stakeholders believe that respondents have trouble answering them. They believe that shortening the questions could minimize some of the confusion.
Recommendation 8

Establish Formalized Partnerships with Historically Black Colleges and Universities

“That the Census Bureau establish more formalized partnerships with Historically Black Colleges and Universities (HBCU’s)—similar to the relationship between the Bureau and the University of Puerto Rico—to facilitate access to and participation in career and academic opportunities so that the Census Bureau can more effectively fulfill its commitment to racial and ethnic diversity. Consequently, we request the following information related to such partnerships:

a. List of targeted HBCU’s under consideration
b. Timeline for project implementation
c. Draft of Memorandum of Understanding (MOU’s)
d. The number of census staff who will be assigned to this project”

Census Bureau Response

The Census Bureau welcomes opportunities to cultivate sustained relationships with a variety of colleges and universities. This is a cornerstone of our recruitment program and critical to our efforts to promote career and student internship opportunities at the Census Bureau. Toward that end, the Census Bureau regularly visits college campuses twice a year to conduct on-campus recruiting activities. These activities include formal lectures on career opportunities at the Census Bureau, rigorous promotion of student internship opportunities, and recruitment and interviewing for full-time professional positions.

Our college recruitment activity focuses on those campuses that produce high numbers of graduates qualified for our major occupational series (Mathematical Statisticians, Statisticians/Analysts, IT Specialists, and Geographers). In developing this list, we have taken great care to ensure that a large number of minority-serving institutions are included in our ongoing campus activities. Attached for your consideration is a complete listing of those colleges and universities that the Census Bureau regularly visits that have sizeable African-American student populations (see Attachment 1).

In addition to these efforts, the Census Bureau has cultivated closer relationships with certain schools and organizations that serve the African American communities. Since 1998, the Census Bureau also has been a member of the Norfolk State University Cluster, with the principal goal of preparing NSU graduates for careers in the private and public sectors. Over the past year, the Census Bureau has engaged in a pilot program with the Atlanta University Center, a consortium of six historically black colleges and universities in Atlanta. The purpose of the pilot was to identify opportunities for the Census Bureau to increase its involvement in campus activities as a means to supplement traditional recruitment trips. The pilot program also tested the concept of utilizing resources from the Regional Office to establish or expand upon relationships between the Regional Office and the campuses. This pilot is presently under evaluation.

Recommendation 9

Provide a Comprehensive Employment Audit of African American Employees at All Grade Levels

“We recommend that a comprehensive employment audit of African American employees at all grade levels be prepared for our committee. Specifically, such an audit should provide information in the following areas:

a. The extent of participation of African Americans in grades 12 through SES on key assignments and programs areas (e.g., ACS, MAF/TIGER, Geography, 2010 Planning, etc.).

b. A description of the scope and content of responsibility of the positions held by African American employees in grades 12 through SES.

c. The average length of time African Americans have been in their current grades by series and organization or directorate.

d. The number of African Americans participating in leadership training and mentoring programs.

e. Succession planning for employees of all races at all grade levels.

f. Description of promotions policy and the process for new hires of all races at the Bureau.”

Census Bureau Response

While the Census Bureau appreciates the committee’s interest in the welfare of our African American employees, we do not agree that this information request contributes to the official discourse that is within the scope of the REAC Committee. Nevertheless, be assured that the Census Bureau is a vigorous proponent of equal opportunity for employees from all backgrounds.

With respect to African American employees in particular, there are a number of organizations which have proven to be effective forums for receiving feedback from our employees. The Equal Employment Opportunity Advisory Committee and the Diversity Council both provide opportunities for employees to voice concerns about employment issues. In addition, the Census Bureau formally recognizes two voluntary employee organizations that promote the interests of our African American employees—the Census African American Managers Group and the Census Chapter of Blacks in Government.

Recommendation 10

Provide a Count of Prisoners in Their Pre-incarceration Address

“We recommend that prisoners, including those housed outside their states, be counted as residents of their pre-incarceration addresses for purposes of Congressional apportionment, state redistricting, the distribution of financial aid to their home communities, and to permit those communities to have more resources to meet their needs upon their return.”

Census Bureau Response

Currently, we have no plans to reevaluate the methodology that we use for counting prisoners at the institution where they are incarcerated. Census Bureau residence rules are a complex set of interdependent rules for making sure that every resident of the United States is counted only once and in the correct place.

Census Bureau residence rules state that individuals should be counted at their “usual residence.” The concept of "usual residence" as the main principle for determining where people are to be counted was established for the first census in 1790. This concept has been followed in all subsequent censuses. Currently, usual residence is defined as the place where the person lives and sleeps most of the time. This place is not necessarily the same as the person's voting residence or legal residence. According to the current formulation of the residence rules, all residents of institutional group quarters are counted where they sleep or live most of the time. Consequently, we do not collect data about the home community of the institutionalized population.

In 1992, the United States District Court for the District of Columbia confirmed the principle of usual residence (District of Columbia v. U.S. Department of Commerce). In that case, prisoners who were residents of the District of Columbia were incarcerated in a prison located in the State of Virginia and, consequently, counted as residents of Virginia in the 1980 census. The District of Columbia claimed that this action violated the Administrative Procedures Act, Article 1, Section 2, Clause 3 of the Constitution (the authority for taking the decennial census for the purpose of apportionment), and the “due process clause” of the fifth amendment, and that it would cost the District $60 million over the decade.

The court ruled that the application of the Census Bureau’s “usual residence rules” to prison inmates was a rational decision that was neither arbitrary nor capricious, and that it did not violate the constitutional command of the census clause. In other words, the court decided the Census Bureau could count the inmates at their usual place of residence as it is defined above.

Recommendation 11

Conduct Further Research to Accurately Count/Report West Indian Ancestry and Sub-Saharan African Populations

“The Census Bureau should conduct further research to determine the best ways to accurately count and report West Indian ancestry and sub-Saharan African populations.”

Census Bureau Response

Counting an increasingly diverse population is a challenge that grows with the passage of each successive decade. Through the testing cycle for the 2010 census, we are attempting to make many improvements in the coverage of all populations, including those of West Indian ancestry and Sub-Saharan African populations.

Source: U.S. Census Bureau,
Census Advisory Committee Office

Created: 01, 26, 2005

Tweet this page Follow @PrisonPolicy on Twitter Get our newsletter Donate


  • May 15, 2018:
    Our Policy Analyst Lucius Couloute will be at the LEDA Summit on Race and Inclusion in Holland, Michigan, presenting his research on the challenges and disadvantages people face when they are released from prison. Tickets are available on LEDA’s website.

Not near you?
Invite us to your city, college or organization.