IntroductionAfter years of preparation, Census Day 2000 has finally arrived. In preparing for this day, the Census Bureau, state and local governments, and thousands of partners in the private and public sectors have taken steps to ensure the success of Census 2000.
This report details the findings from a series of joint field observations conducted in recent months. Specifically, the Board has undertaken case studies of Bureau operations in both the Dallas and New York census regions, conducted observations of critical training sessions at the Census Bureau's National Processing Center in Jeffersonville, Indiana, and sent staff to observe census operations in the state of Alaska.
With some exceptions detailed in the following report, and subject to their relatively limited scope, our observations indicate the planning for census operations and community relations is generally proceeding well in the regional offices and local census offices we visited.
Participation in the census is one of the most basic civic responsibilities. The results of the 2000 Census will, for the next ten years, affect every aspect of American life. The results will guide the allocation of political representation at the federal, state and local levels, determine funding for schools, hospitals, roads, bridges, and utilities, and guide vital decisions to address some of our most pressing social problems. In the private sector, businesses will rely on these numbers to decide where to invest, and where to build stores, factories and bank branches and housing.
This census has seen its share of political controversy. But for all the debate, we join unanimously to urge every resident of the United States to complete and return the census forms they have received and to cooperate with the Census Bureau's enumerators when they begin their efforts to find those who have been missed.
CENSUS OPERATIONS UPDATE
Following is a brief summary of major census operations. As with any undertaking of such massive scale, there have been operational problems to date and there likely will be more before the Census Bureau concludes its field operations later this year.
It should be noted that the largest and most vital operations, the Mailout and return of census questionnaires by mail and the field effort to follow up on non-responding households, are either still underway or yet to begin. These observations about census operations must, therefore, be considered preliminary.
Remote Alaska Enumeration
Enumeration efforts began on January 19, 2000 with the Remote Alaska Enumeration. This operation begins well in advance of Census Day in order to count residents of remote villages before the change of seasons prompts many to leave their communities to hunt, fish, or engage in other employment. For details on the Monitoring Board's observations of this operation, see Section IV of this report.
Advance Notice Letter
On March 6, 2000, the Bureau mailed advance notice letters announcing that Census 2000 questionnaires would be arriving soon. The letter was sent to approximately 112 million households nationwide. In addition to announcing the arrival of the census forms, the letter announced the availability of census forms in Spanish, Korean, Japanese, Vietnamese, Chinese or Tagalog. By checking off a selection on the letter and returning via an enclosed envelope, households could receive a form in one of these six languages.
A significant printing error on the letters was discovered after they had been shipped to U.S. Postal Service facilities for delivery. The contractor hired by the Government Printing Office (GPO) to print the letters inadvertently inserted an extra digit as the first character in the street address line of the address. The Bureau's quality control procedures failed to detect the error.
The Postal Service provided the Bureau with assurances that the letters would be successfully delivered despite the error, and the Monitoring Board would like to thank the Postal Service for their efforts to ensure this important operation was carried out.
There are indications the advance letter has had a positive effect. As of March 20, 2000 roughly 1.5 million non-English forms had been requested via return of the advance letter. In addition, on March 21, 2000, the Bureau Director reported the results of the Inter-Survey (a survey funded by private donations but coordinated with the Bureau) which found that the number of respondents who understood that the census is primarily a Mailback operation climbed from 58 percent prior to the letter's delivery to 84 percent afterward.
The Update/Leave enumeration efforts (in which enumerators hand-deliver questionnaires in rural areas while simultaneously updating the address lists) began on March 3, 2000 and were scheduled to end on March 30, shortly after this report went to press. List/Enumerate operations, used in remote and sparsely populated areas where enumerators will directly enumerate residents rather than delivering questionnaires for Mailback, began on March 13, 2000 and are scheduled to continue until May 1, 2000.
On March 13, 2000, the Bureau mailed out roughly 98 million forms to households as part of its Mailout/Mailback effort. The Bureau had previously verified that the printing error on the advance letters was not duplicated on the forms. As of March 21, 2000, just under 15 million households had returned their forms, and the Bureau's four Data Capture Centers (DCCs) had processed 7.3 million of them.
Urban Update/Enumerate Operations
Urban Update/Enumerate operations were scheduled to begin on March 20, 2000, and are scheduled to be completed by May 30, 2000. The Bureau intends to use the Urban Update/Enumerate strategy in selected areas with historically high or potentially high undercounts. Enumerators will update the census address list and directly interview residents of selected areas' households, rather than dropping off questionnaires and awaiting a reply by mail.
Recruiting efforts aimed at filling the more than 535,000 staff positions the Bureau will need at peak operations were on track at the national level. On March 21, 2000 the Bureau Director announced that the Bureau had recruited an applicant pool of 2.2 million qualified individuals. This is roughly 91 percent of the national recruiting goal, and was reached just under one month prior to the April 19, 2000 recruiting deadline. The Bureau's 520 Local Census Offices (LCOs) were open and operating. For details on the Monitoring Board's findings on recruiting and LCO operations in the Dallas and New York census regions, see Sections II and III of this report.
Enumerators fluent in the language of the community are essential to a successful count in linguistically isolated communities. The Board will continue to monitor the effect of English proficiency tests - required of all enumerators in the lower 49 states and the District of Columbia - on hiring in these areas. We are encouraged by the reports from local offices detailed in this report, which indicated the English proficiency test had not been a barrier to recruiting a suitable workforce.
Congressional Members' Position
Initial reporting of the Bureau's recruiting progress has focused on national, not local, recruiting goals. The Congressional Members of the Monitoring Board believe that neighborhood hiring levels - not national goals - are the appropriate measure of recruiting success. Reporting on the national or aggregate level can mask the issue that there may be neighborhoods, particularly hard-to-count neighborhoods, where indigenous recruiting and hiring is not successful.
Based on conversations and interviews with local leaders across the country, the Congressional Members are still concerned that the Bureau's national plans to recruit and hire locally may not be realized in some hard-to-enumerate areas. We will continue our efforts to monitor local hiring in the field, through observation and interview with both local census offices and local partners.
The Congressional Members agree that the best enumerator is a person who looks like and sounds like the person answering the door. Neighborhood residents have the knowledge of their own communities - something altogether different than verbatim training and operational directive. The Congressional Members believe the best way to reduce the undercount in hard-to-count neighborhoods is to ensure local residents are hired to take the census in their own neighborhoods.
For example, enumeration in Miami's Little Haiti depends on hiring Haitian residents of Little Haiti; and enumeration in nearby Little Havana depends on employing Cuban residents of Little Havana.
In 1990, differential pay scales were first used to meet this challenge. The differential pay scales meant that people living in urban neighborhoods could effectively be hired. This was an important first step that the Bureau has continued in 2000.
The public statements of the Secretary of Commerce and the Director of the Bureau of the Census share this view. Secretary of Commerce William M. Daley stated, "the ideal census taker for us is a person who lives in a neighborhood. It is someone who knows the territory and knows the families in the area in which they live…it is someone who knows where the children are and how many children there may be in the building."
The Director of the Census Bureau publicly echoed this commitment, saying, "our goal is to have a pool of local people who are familiar with their communities and committed to a successful count in their own neighborhoods." While these statements are encouraging, the Board does not have sufficient information to affirm that stated goals are being implemented in the neighborhoods and on the streets in the hard-to-count communities.
We recognize the Bureau is fully capable of incorporating cultural sensitivity and adaptability in their hiring practices. The current enumeration operations in Remote Alaska illustrate this, and should be used as a model for other communities throughout America that are the hardest-to-count: colonias, vast sections of public housing, and remote rural areas such as the Mississippi Delta. For example, in the Native Alaska communities of Remote Alaska, the Bureau has waived the hiring test requirement and solicits the input of Tribal Leaders when they hire village residents as enumerators.
Presidential Members' Position
The Presidential Members join the Congressional Members in reaffirming the importance of local recruiting and hiring to a successful enumeration. However, we note that the Census Bureau has long incorporated local recruiting and hiring as an integral part of its planning for Census 2000 - a commitment which pre-dates the establishment of the Monitoring Board and one which was documented in the Board's joint report on April 1, 1999.
The importance of local recruiting, hiring and assignment has been repeatedly stressed by the Secretary of Commerce, the Director of the Census Bureau, and by all the regional and local census office officials with whom we spoke in the process of researching this report.
Each LCO visited during the preparation of this report had made the recruiting of local residents a priority. Their dedicated efforts to ensure that recruiting reaches into hard to enumerate neighborhoods are documented in this report. In particular, our visit to the New York Northwest LCO confirmed that the LCO staff was aware of the challenges they were facing in recruiting sufficient staff to enumerate the Mexican American community in their jurisdiction. We saw no lack of commitment by the LCO to ensuring those recruiting challenges are met.
While we agree that measuring the success of these efforts will be an important evaluation, and will continue to examine the Bureau's progress, we find no basis for questioning the Census Bureau's commitment to this recruiting strategy.
Food Stamps, Medicaid, Children's Health Insurance Program (CHIP) and Temporary Aid to Needy Families (TANF) Waivers
Effective hiring at the neighborhood level in economically disadvantaged communities will often mean hiring residents who receive public assistance such as Food Stamps, Medicaid, Children's Health Insurance Program (CHIP) and Temporary Assistance to Needy Families (TANF). Until recently the fear of losing government assistance for taking a temporary census job was a concern for the Census Bureau and for residents in low-income communities.
In 1990, the Census Bureau addressed this concern through a series of cooperative agreements with the Federal agencies that manage the most important government assistance programs. However, the waiver process has proved significantly more complex for 2000. The Welfare Reform Act has transferred much of the responsibility for administering social services to the states.
As a result, for Census 2000, the Census Bureau and partners in Congress have worked to secure waivers at both the national level and in individual states.
On December 29, 1998, the U.S. Department of Housing and Urban Development announced a waiver for recipients of Section 8 housing assistance. Since the U.S. Department of Agriculture announced the availability of a similar waiver for Food Stamp recipients in early February 2000, 46 states have approved the waiver.
Similarly, under a waiver announced by the U.S. Department of Health and Human Services, 25 states have announced waivers for recipients of Temporary Assistance for Needy Families (TANF). Twenty-eight states have announced that they will exclude census earnings from eligibility calculations for Medicaid, and 25 have announced the same policy for CHIP eligibility.
The waiver program remains an important recruiting tool for Census 2000, especially in communities which may be experiencing a shortage of qualified applicants in hard-to-enumerate areas.
Congressional Members' Position
The Congressional Members commend the Bureau for their work in advocating this issue, as well as those Senators who intervened directly with the Secretary of Agriculture to encourage him to reverse the Department's initial rejection of the waiver proposal.
Efforts to form partnerships with state and local governments and community organizations to promote the census continue. As of March 21, 2000, the Bureau reported more than 102,000 partnerships across the country.
Questionnaire Assistance Centers and Be Counted Sites
The 12 Regional Census Centers, and the LCOs in their jurisdictions, had identified a total of 27,000 sites for Questionnaire Assistance Centers (QACs) and more than 21,000 sites at which to place forms and materials for the Be Counted effort. QAC startups began on March 8, 2000. These centers are scheduled to remain open until April 14, 2000, shortly before Non-Response Follow-Up operations begin.
Telephone Questionnaire Assistance
Problems were initially reported with the Telephone Questionnaire Assistance (TQA) system designed to provide assistance to individuals with questions about the census form. (Telephone assistance is available in English, Spanish, Chinese, Japanese, Vietnamese, Korean and Tagalog.) According to Census Bureau Director Kenneth Prewitt, the initial incoming call volume exceeded predictions, which were based on the timing of calls to the 1990 Census telephone assistance system. As a result, calls came in before the system was operating with its full staffing complement of 9,100. Director Prewitt later reported the problem seemed resolved once all the planned operators were in place. The TQA system had logged a total of 1.8 million calls as of March 18, 2000. Roughly 15 percent of these calls were to the Spanish language line.
Special Population, Transient and Group Quarters Enumeration
Bureau operations to enumerate special, and often hard-to-enumerate, populations were scheduled for late March (shortly after this report went to press). On March 27, 2000 the Bureau planned to send enumerators to transitional and emergency shelters to count residents of those facilities. On March 28, enumerators were scheduled to canvass soup kitchens and mobile food vans for those who may be missed by other census operations. On March 29, teams of enumerators were scheduled to count homeless persons living in outdoor areas, and on March 31, persons without a permanent address living at transient locations (including parks, fairgrounds, carnivals and marinas) were to be enumerated. On April 1, 2000, the enumeration of group quarters is scheduled to take place. This operation will count residents of nursing homes, college dormitories, residential treatment facilities and prisons.
Finally, on April 27, 2000, the Bureau will begin a massive Non-Response Follow-Up (NRFU) operation, deploying enumerators across the nation to count those individuals who did not return forms following the mailout/mailback or update/leave operations. The ultimate scope of this operation remains unclear. Much will be determined by the mailback response rate - the greater the number of forms returned by mail, the fewer the households which must be contacted during NRFU.
NRFU field operations are scheduled from April 27 to July 7, 2000. The Monitoring Board will be closely following the progress of this critical operation in the coming months.