The Be Counted Program is designed to provide a means for people to be counted who may not have received a census form or believe they were not included in the census. Be Counted forms are geared to those populations that have historically been hard-to-enumerate, and will be available from March 31 through April 11, 2000.63

The Be Counted form is an unaddressed questionnaire which will be available in six languages and distributed at locations such as city halls, libraries and targeted locations frequented by historically hard-to-enumerate populations, such as certain churches, restaurants or community centers.64 Be Counted questionnaires will be printed in English, Spanish, Chinese, Tagalog (Filipino), Vietnamese and Korean.

The Bureau has already incorporated feedback from the Commerce Department’s Office of the Inspector General (OIG), based on dress rehearsal experience, and plans to limit the scope of the Be Counted Program to "include only sites identified in consultation with local partners in hard-to-enumerate areas."65 Also, the duration of the Be Counted campaign has been abbreviated, from four weeks in the dress rehearsals to eleven days during Census 2000.66

Unduplication: A potential increase in duplicate responses through the use of Be Counted forms was a concern prior to the dress rehearsals. However, extensive duplication did not prove to be a problem, in part due to the minimal response to the Be Counted Program.

Be Counted forms undergo an unduplication process that requires the forms to be geocoded and checked for duplicate responses upon receipt and may require field verification.67 In addition, the Bureau added a question to the form which asks respondents if they have filled out another census form during the past week.68

The dress rehearsal results "indicate that additional error caused by the multiple response operations should be minimal both in overall effect and as a percent of all errors. … In every test including multiple response options that the Census Bureau has conducted to date, the use of Be Counted forms and other multiple response options has been minimal."69

Site Selection: Be Counted site selection was inconsistent during the dress rehearsals. For 2000, the Bureau plans to select distribution sites for Be Counted forms in conjunction with local partners and partnership specialists.

Analysis from the Commerce Department’s Office of the Inspector General (OIG), and firsthand reports to the Board, indicate that communication and coordination with local partners was inconsistent during the dress rehearsals. The OIG reports:

There are a number of reasons for the ineffectiveness of the Be Counted campaign [in South Carolina]. Some sites were targeted too early….Others were unaware of their commitment because the person agreeing to participate had since left the organization….In addition, after the Be Counted campaign began, the regional partnership coordinator recognized that locations contacted in the fall of 1998 should have been contacted again before the start of the Be Counted campaign [in April 1998] to ensure their commitment.70

During the Board’s meeting in Columbia, South Carolina, local census partners testified to their frustration with the Bureau’s internal and external communications, particularly in selecting Be Counted sites. Anita Floyd, Director of the Columbia Women’s Shelter, testified:

The Be Counted sites were not identified early enough; [the Bureau] didn’t follow up on the ones that we suggested, and then the ones that they selected were inappropriate….We just had a very strong impression that we were doing a lot of work that was not going to be effective because it was going nowhere once it actually got to the census staff.71

In Menominee, site selection went smoothly. The locations were well selected, accessible and well stocked – partially due to an inadvertent delivery of more than three times the necessary forms. However, Bureau staff displayed excellent adaptability, expanding the number of sites from eight to twelve: two post offices, two senior citizen centers, two churches, four stores, a health clinic and a community center.72 The staff who worked near the Be Counted displays knew that the forms were for public use for people who believed that they had not filled out the mailed questionnaire.  

The Bureau is right in advocating Be Counted sites to reach traditionally undercounted populations. Although the OIG has questioned the efficacy of QACs and Be Counted sites, the Board concurs with the Bureau that these are important components to an outreach effort targeted to decrease the differential undercount of HTE areas and people with limited English proficiency.

However, the dress rehearsals clearly demonstrate that the Bureau must improve its follow through on the promise to work closely with local partners to target HTE areas for Be Counted sites. As in the case of enumerator recruiting and assignment, the Board recommends a dedicated focus on early identification of HTE areas, using the Planning Database and local partner input. In particular, the Bureau should work with Complete Count Committees to identify areas that need forms in languages other than English, and to ensure appropriate targeting for specific languages.