Dallas Region
A. Dallas County South, LCO #3035
B. El Paso, LCO #3036

REGIONAL CENSUS CENTER

Staff from both sides of the Census Monitoring Board visited the Dallas Regional Census Center (RCC) and two local census offices in the region - Dallas County South and El Paso, Texas February 28 - 29, 2000. Senior management officials of the RCC briefed the staff for approximately four hours. Conducting the briefing were the Dallas Regional Director, Deputy Regional Director, Area Regional Census Managers, Regional Recruiters, the Regional Media Specialist, and the Regional Partnership Coordinator.

The Dallas Regional Census Center is responsible for conducting the decennial in three states - Texas, Louisiana and Mississippi. The combined population of these three states is estimated at 24 million people living in approximately 11,455,000 housing units. There are 7 American Indian Reservations and over 6,000 governmental units. The region contains twenty-five cities with over 100,000 people and vast stretches of rural counties. To assist in counting this area of over 350,000 square miles, the Bureau opened 46 Local Census Offices. Thirty-two are in Texas, 9 are in Louisiana, and 5 are in Mississippi.

Dallas Region Enumeration Strategies

STRATEGY	PERCENT OF TOTAL
Mailout/Mailback	75.9%
Update/Leave and Urban Update/Leave	23.2%
Update/Enumerate	0.8%
List/Enumerate	0.1%

Due to the diverse geography of the region, the Dallas RCC will employ five different enumeration strategies: mailout/mailback, update/leave, urban update leave, update/enumerate, and list/enumerate. Mailout/Mailback describes the method where the Bureau mails out a census form and expects a form back in the mail. Update/leave is when the Bureau updates its address list while delivering questionnaires in rural areas and expects the form back in the mail. Urban update/leave is the same operation as update/leave but in urban areas. Update/enumerate is used in targeted areas with historically or potentially high non-response rates by updating the address list while conducting door-to-door enumeration. List/enumerate is used in remote and sparsely populated areas to create an address list and conduct door-to-door enumeration.

Hard-To-Enumerate Planning

RCC Managers used the Planning Database (PDB) developed by Census Headquarters to help identify hard-to-enumerate areas, areas of potentially low mail return rates, placement of Questionnaire Assistance Centers (QACs) and/or Be Counted sites, areas that need special outreach and promotion activities and where additional recruitment activities are needed in the Dallas Region. The PDB evaluates and arranges 1990 census data and mail response rates to generate a Hard to Enumerate (HTE) score for each census tract in the country. Furthermore, the RCC worked with the Texas State Data Center and the Attorney General's Office to pre-identify HTE areas and shared the findings with local government officials and planners. In order to raise awareness early in HTE areas, meetings between census officials and from state and local governments took place in the fall of 1999. In these meetings, the PDB was updated with changes from the last 10 years, some tracts were added and some were removed from the HTE list.

The Regional Director sent a memo in April 1999 asking each Early [Opening] Local Census Office (ELCO) to "check" the PDB to determine its HTE areas and develop an action plan for the areas by July 1, 1999. The plans were, and are, subject to change. According to the April memo, "[the Bureau] cannot guarantee that all aspects of a plan will be implemented. There are obstacles, such as budgetary constraints, that may be limiting factors."

As LCOs opened, the new PDB spreadsheets that listed HTE tracts were shared with the office. The spreadsheet included suggested strategies to count the pre-identified hard to count areas. Such strategies include pair/team enumeration, blitz enumeration, bilingual enumeration, guides or local facilitators, Be Counted site location, QAC location, and use of update/enumerate and urban update/leave delivery strategies. Relying on local knowledge, the LCOs were able to locate the pre-identified HTE tracts and decide the appropriate enumeration tool to use in that particular area. LCOs also further refined and continue to update the PDB to reflect local changes.

Colonias

The Dallas RCC has taken proactive steps in one of the most undercounted areas of the country in the 1990 census - colonias. Colonias are small, rural, unincorporated communities primarily along the Texas/Mexico border. These communities often lack basic services such as water, paved streets, and sewage systems.

According to the Texas Water Development Board, in 1996, there were a total of 1,512 colonias. The highest concentration of colonias is found in Hidalgo, Cameron, Starr and El Paso Counties.

COUNTY	NUMBER OF COLONIAS
Cameron	111
El Paso	151
Hidalgo	867
Maverick	44
Starr	124
Webb	43

The Census Bureau is expending extra attention to these traditionally undercounted communities. In 1990, colonias were not considered census designated places. As a result, data from colonias were combined with data from adjacent non-colonia areas. The former Attorney General of Texas, Dan Morales, and the Dallas Regional Census Center worked together with colonia leaders, local governments, planning agencies and Councils of Governments to ensure that in the 2000 census, colonias would be listed as census designated places. This designation allows the Bureau to compile data specifically for colonia areas.

The Bureau plans to hire 900 bilingual enumerators and colonia cultural facilitators in Starr, Cameron, Hidalgo and El Paso counties. Recruiting and testing for enumerators in the colonias is in Spanish. Cultural facilitators will not be required to take the enumerator test and will be paid separately from the PAMS/ADAMS payroll system.

To conduct the actual enumeration in 70 percent of the colonias, the Dallas RCC will employ an update/enumerate questionnaire delivery strategy in El Paso, Hidalgo, Cameron, and Starr Counties. Beginning in late March and continuing through May, a local census enumerator accompanied by a cultural facilitator will visit each colonia household and assist residents in completing an English-language census questionnaire. Enumerators and cultural facilitators will employ Spanish language guides during this operation, but Spanish questionnaires will not be used.

The Bureau is also working with Texas A&M's colonia program. Texas A&M is employing the colonias' promotoras - local community leaders - to conduct home visits with small groups to build awareness about the census and provide any assistance residents might need. An update/enumerate strategy will also be used on all 7 of the American Indian Reservations located within the region.

In extremely rural areas which lack residential mail delivery, census enumerators will list and map housing units and enumerate occupants on the same visit. Bilingual enumerators will be used for this operation.

In the remaining 30 percent of the colonia areas, the Dallas RCC will use the update/leave method. The Bureau will drop off an English language questionnaire along with a Spanish language questionnaire assistance guides. Regional staff reported the decision to deliver English language forms to all areas, including monolingual Spanish areas such as some colonias, was a decision made early in the decennial process by the Bureau's national headquarters. Questionnaire Assistance Centers are in place with paid and volunteer staff.

Congressional Members' Position

The Congressional Members of the Monitoring Board are skeptical of the Census Bureau's assertion that limited and targeted distribution of Spanish language census forms was impossible for Census 2000. We believe the Bureau has made a mistake that may reduce response rates in the linguistically isolated, hard-to-count Hispanic communities.

This is not a belief solely held by the Congressional Members of the Monitoring Board. The Board has been informed throughout the past two years by many stakeholders that there are a limited number of hard-to-count communities throughout the United States, especially in the Southwest and Texas, that are completely linguistically isolated. The Bureau's Planning Database and local offices corroborate this fact. These communities, we have been told on numerous occasions, need to get a their census forms in Spanish. Benito Juarez, an immigrant and community leader working in the immigrant community in Texas, told the Monitoring Board "we believe that instead of sending this form in English to Spanish speakers, they should send a form in Spanish. So in that way people will be able to understand better what it is about and feel more confident in filling out and participating in the census."

The Bureau has consistently resisted targeted distribution of census questionnaires in these communities. The Bureau has held steadfast to their plan, requiring residents in these linguistically isolated communities to receive a letter written in English to then request a Spanish language questionnaire. However, senior Bureau officials indicated Regional Census Centers have the discretion to deliver notices in Spanish in these neighborhoods. Despite frequent staff visits to colonias and surrounding areas, the Congressional Members have been unable to confirm delivery of Spanish-language notices.

Moreover, in the colonias of South Texas - areas the Bureau knows to be monolingual Spanish - the Bureau is delivering English language questionnaires in update/enumerate and list/enumerate areas. This is particularly unsettling because, unlike mailout/mailback areas where the Postal Service delivers the questionnaire through regular mail service, these are areas where a Census 2000 employee will hand-deliver a questionnaire. We believe this attitude toward the hard-to-count, linguistically isolated communities should change. The Census Bureau and the Regional Census Centers, in particular the Dallas and Los Angeles Regional Census Centers, should deliver Spanish language questionnaires in these linguistically isolated communities, including the colonias.

Presidential Members' Position

The Presidential Members of the Board agree that distribution of non-English forms to residents who wish to receive them is an excellent goal. We note that the Census Bureau explored the possibility of distributing non-English forms for Census 2000, but were unable to overcome the operational difficulties involved.

We believe the subject is worthy of further study for possible inclusion in the 2010 census. However, at this late date, we do not believe it is feasible for the Census Bureau to alter its questionnaire delivery strategies for Census 2000. Update/Leave operations are well underway and the mailout of forms has already occurred.

Steps have been taken in the field to reach non-English speaking residents. In particular, we note the efforts of the Dallas Regional Census Center and the LCOs we examined in the Dallas Region to reach the residents of the colonias along the Texas/Mexico border. In those areas, a Spanish language assistance guide will be hand delivered with each questionnaire.

Questionnaire delivery operations that involve door-to-door delivery by hand, such as Update/Leave, may provide an excellent opportunity for enumerators to deliver non-English forms to residents in selected neighborhoods.

However, there are serious challenges to the idea of incorporating non-English forms in the mailout/mailback operation. In order to mail non-English forms to a given neighborhood, the Bureau must effectively assume that all residents of that neighborhood in fact speak the targeted language. This may be a difficult assumption to justify; it is doubtful that any neighborhood in the nation is entirely monolingual in a language other than English. In addition, the fact that the census occurs only once per decade may make precise targeting difficult, if not impossible.

We do not conclude that these barriers are insurmountable. However, additional study is needed before such a major change to census operations is undertaken. Given that questionnaire delivery operations are already underway, such a major change to field procedures might well do more harm than good.

Recruiting and Staffing

In line with the Bureau's national recruiting plan to recruit five people for every census position, the Dallas Region plans to recruit over 276,415 people to fill 55,105 positions. As of February 18, 2000, the region as a whole had reached 84.5 percent of its recruiting goal - well above the national benchmark of 70 percent. The RCC was at 74.3 percent of its recruiting goal in Louisiana, 83.6 percent in Mississippi, and 84.5 percent in Texas.

OPERATION	TIME	POSITIONS	TOTAL RECRUITING GOAL
Update / Leave	March	7,617	38,975
Update / Enumerate	March	942	4,710
Non-Response Follow-Up	April 27 through June	46,546	232,7307

At the time of interview, the major operation underway nationally was recruiting. The Dallas RCC had 500 full-time recruiting positions. In addition, management encouraged all staff to act as recruiters. Staff attribute recruiting success to aggressive strategies, including mailing 2.4 million "Census Jobs" post cards to zip codes where recruiting was low, providing recruiting materials and testing in Spanish, and offering competitive pay.

The Bureau's policy of recruiting five times as many people as are needed and training twice as many people to fill temporary positions is meant to offset turnover. In order to recruit such a volume of applicants, the Bureau is testing applicants months before they are needed. While this method allows the Bureau a large number of people from which to draw, it also creates some frustration among early applicants because the Dallas Regional Office has not implemented a method of notifying applicants whether they will be offered a job. Radio, television, and print solicitations for additional applicants may have confused or discouraged those who had already applied, but were not notified of their status.

The Monitoring Board recommends that, for the 2010 Census, the Bureau establish a process of routine follow-up with job applicants. After testing, applicants should be notified as to whether they passed, and when the Bureau plans to begin filling positions.

The Dallas Regional Office printed a pamphlet informing applicants they would not hear from the Bureau until mid-April (when most census jobs are offered.) These pamphlets were distributed at testing sites. While this is a positive informational tool, a routine follow-up notice would prevent some applicant frustration. Nationally, the Bureau expects to recruit 2.5 million people to fill census jobs - a gesture of professional courtesy could prevent some public relations problems. (The New York City Region has been able to mitigate some of these criticisms by sending a post card to those who do not pass the enumerator test. Since applicants can re-test repeatedly, the notice encourages re-testing.)

The Bureau had a national goal to hire 4,000 welfare recipients by September 1999 with more to be hired this year. Each LCO developed Welfare-to-Work recruiting goals. At the time of interview the Dallas Region had hired 434 welfare recipients.

As of March 15, all states in the region had applied for federal waivers that would allow income from temporary census employment to be exempt from the public assistance income eligibility caps. That is, people on public assistance could work for the census without the risk of losing benefits.

The RCC is monitoring the recruitment process closely. The Recruiting Managers at the RCC receive weekly updates on available test sites and monitor closely the number of welfare-to-work test sites and applicants. The Managers also watch each LCO progress towards their recruiting goal and make personal visits to LCOs experiencing recruiting difficulty. By maintaining weekly updates, the Recruiting Managers are able to send a biweekly bulletin to the region's recruiters, Fast-Breaking Recruiting News Briefs.

At the time of interview, the Dallas RCC had tested more than 3,000 applicants in Spanish. Staff reported more than 90 percent of applicants scored the requisite 70 or higher on the census test.

Training

The bulk of the Bureau's hiring - enumerators to conduct Non-Response Follow-Up (NRFU) - will begin in mid-April. At the time of interview, the RCC had hired 230 Managers, 22 Regional Technicians and 55 Partnership Specialists. Each LCO had 5 managers: the Local Census Office Manager, the Assistant Manager for Field Operations, the Assistant Manager for Administration, the Assistant Manager for Recruiting and the Automation Technician. Regional technicians are responsible for a subsection of the region and the Partnership Specialists are responsible primarily to the Regional Office.

Each manager went through extensive training for 4-5 days in October and December of 1999. The RCC provides job-specific training and just-in-time training as a refresher before major operations, such as special places enumeration and the implementation of QAC/Be Counted sites. These sessions started in December of 1999 and will continue through July 10, 2000. Just-in-time training will also take place before the update/leave, update/enumerate, list/enumerate, non-response follow-up operations.

LCO Managers receive media training, including instruction in maintaining confidentiality of census records. This training proved helpful in the El Paso office when an FBI agent sought enumerator credentials, thus creating a potential public relations problem. The LCO manager followed Bureau policy of maintaining strict confidentiality and did not share any census information with the FBI agent - even when threatened with a potential warrant.

Partnership

For the 2000 Census, the Bureau reports relying heavily on its Partnership Specialists and local partners to improve the response rate, reduce the differential undercount and increase the public awareness of the census and census jobs.

The Dallas Region has 48 Partnership Specialists - 33 in Texas, 6 in Mississippi and 9 in Louisiana. Partnership Specialists are based in local census offices but report to the Regional Office. In Texas and Mississippi, two local census offices have two partnership specialists working in their areas while the rest of the local census offices have one. In some cases, the partnership specialist is responsible for a vast geographic area. For example, the Dallas County South LCO's partnership specialist covers the southern half of Dallas County, while the El Paso LCO's specialist is responsible for 10 counties.

All partnership activities are coordinated out of the Dallas RCC through the Partnership/Recruiting Assistant Regional Census Manager (ARCM) and two Partnership Coordinators, one for Texas and one for Louisiana and Mississippi. Each state in the region has sponsored a web site to promote the census.

In Texas, the Dallas RCC was able to coordinate a joint letter sent by the Director of the Dallas Regional Office and the Governor in January to 500 Complete Count Committees in Texas thanking them for their participation and announcing the State's own promotional efforts. The State also has six ombudspersons working in the colonias promoting the census. The State Attorney General's office hired a Census 2000 Outreach Coordinator, sent Spanish language radio ads to every Spanish-language radio station in Texas, and sent census reminders in drivers' license notice mailers.

The Bureau held 10 informational and promotional conferences in Texas and Mississippi. Regional staff reported Louisiana has not been as aggressive at promoting the census as the other two states in the region. However, the Governor did three public service announcements about the census and discusses the census during his weekly radio broadcast.

The Regional Office has an African American Initiative, an American Indian Tribal Program, a Religious Initiative, supports Complete Count Committees, encourages Census in the Schools and participates in the national road tour. The RCC also has solicited support from regional business to supplement census promotion. For example, Coca-Cola in Louisiana and Mississippi will place sleeves on two-liter bottles promoting the census and including a coupon for a local baseball game. A Texas statewide grocery chain, HEB, will place the census tagline and the census logo on grocery bags in Spanish and English. The Midland Reporter-Telegram printed and distributed a census coloring book in English and Spanish.

The Partnership Program was able to develop supplemental RCC Spanish-language fact sheets in addition to those provided by headquarters. These included fliers listing QAC locations and census awareness posters. Also, partnership staff arranged translation of the enumerator test into Spanish. These materials are generated at the RCC and then sent to the LCOs.

DALLAS COUNTY SOUTH, 3035

Congressional and Presidential Monitoring Board staff met with the management of the Dallas County South local census office (LCO) for approximately two hours during the morning of February 23, 2000. The LCO manager, assistant managers of administration, field operations, and recruiting were present, as well as the office's automation technician. Also present were the Assistant Regional Census Manager (ARCM), the Partnership Specialist and the Media Specialist responsible for the office. Representatives of the Bureau of the Census Decennial Communications Division and the Department of Commerce Economic Affairs Office were in attendance.

Dallas County South includes the southern third of Dallas County, including six small cities south of the city of Dallas. The population is a mix of Anglo, African American, and Hispanic residents, and includes substantial communities of Mexican immigrants. Office staff noted a relatively small Asian/Pacific Islander population in Pleasant Grove. The LCO area has several trailer parks, and approximately 8,000 housing units that were scheduled to receive questionnaires through update/leave operations in March. Of 107 census tracts in the area, 75 were identified as hard-to-enumerate (HTE) by the Bureau's Planning Database (PDB), using demographic data and response rates from the 1990 Census.

At the time of interview, Dallas County South appeared extremely well-organized, and thoroughly prepared for the upcoming enumeration. Staff cited an emphasis on early and thorough training, and were pleased with the Region's approach, new for this decennial, of training management as a team. The emphasis on teamwork and training continued at the LCO, with the management team meeting every morning. Partnership specialists were included in these meetings.

Perhaps most significantly, the office has the benefit of an LCO manager with more than 30 years of experience at the Bureau, including work on three decennials. Actively recruited out of retirement by the Dallas Regional Director, the office manager clearly has the respect of the Regional Office, and of his staff. Operating with considerable autonomy, the manager has employed his experience (including an extensive network of Bureau contacts) and thorough knowledge of the area to train a motivated, organized team of managers, and position Dallas County South for a successful enumeration.

Hard-to-Enumerate Planning

The LCO Manager demonstrated little familiarity with the Bureau's Planning Database (PDB), and the large maps that papered many of the office walls were not generated by the Dallas Region's geography division. Instead, store-bought maps in several work stations and common areas were covered with neatly-labeled, hand-drawn recruiting and coverage districts. Hard-to-count neighborhoods were identified by name, testing sites, Questionnaire Assistance Centers (QACs) and Be Counted sites were carefully pinned into place. Other than to reference mail response rates, Dallas County South appeared to have little need of the PDB. A detailed strategy to enumerate the area, the product of years of census experience combined with local knowledge, was clearly integrated into office operations.

Questionnaire Assistance Centers (QACs) and Be Counted Sites

LCO staff had selected 20 QACs and 70 Be Counted sites. Partnership specialists and local partners had input in site selection. Be Counted forms were already at the LCO.

Recruiting and Staffing

Dallas County South had surpassed recruiting goals at the time of interview. Bureau records show 6,048 qualified applicants as of February 24, 2000: 124 percent of the goal.

At the time of interview, 95 people were on the payroll. For non-response follow-up (NRFU), the most labor-intensive census operation, Dallas County South anticipates hiring five field operation supervisors (FOS), 40 crew leaders, and 900 enumerators. The LCO manager estimates three or four of every 10 enumerators will be fluent in Spanish.

Recruiting had been carefully planned by census tract, and the LCO area split into eight recruiting districts. Each district included between 10 and 20 tracts. Ten assistants oversaw recruiting: one to each district, with two "floaters."

The Assistant Manager for Recruiting maintained a database of recruiting sources. Many also provided space for testing and training. Sources included churches, recreation centers and schools, among several others. The office also directed special efforts to partner with and recruit local firefighters. According to the LCO manager, emergency workers often make outstanding enumerators, due to their knowledge of the area, community involvement, and schedules providing several days off-duty.

There were 67 test sites and 40 training sites established. The partnership specialist assisted in locating space for test sites. Staff reported that the total time for testing, including filling out employment forms and taking the half-hour timed test, may last between an hour and 90 minutes. Crew leader training is five days straight, on evenings and Saturdays.

Partnership

Staff reported consistent and helpful coordination with the partnership specialist assigned to the LCO, who assisted in planning and locating test sites and QACs. Three active area Complete Count Committees (CCCs) were cited: Dallas, Grand Prairie and DeSoto.

A partnership was established with Potter's House, an area church with a Sunday congregation between 15,000 and 30,000 members. Every Sunday a video message is played, encouraging the congregation to fill out and return their census forms.

In addition, the LCO manager sent letters to all area cities in September 1999, requesting to meet with the City Council and Chief of Police. During these meetings, the LCO manager described upcoming census operations, what the local officials could expect and when, and whom to call with questions or concerns. "I wanted to put a face with a name," he reported, "To say, 'I'm responsible. If there's a problem, please talk to me."

Facilities and Materials

Dallas County South reported no problems receiving materials. All types of promotional materials and posters were available. Congressional staff noted some materials not seen in other areas of the region.

The PAMS/ADAMS system was up and running with no problems. Staff reported the recently-installed PADE data-entry software was much more user-friendly than entering data directly into the database.

Dallas County South exhibited excellent working conditions, with plenty of open space to accommodate storage of a large volume of forms and materials. They also reported sufficient space to conduct update/leave, special places enumeration, and NRFU in separate areas. (The lessor, unable to lease the remaining space, essentially gave the office several hundred square feet of extra space.) Facilities are centrally located in an area shopping center.

EL PASO, 3036

Congressional and Presidential Monitoring Board staff met with the management of the El Paso local census office (LCO) for approximately two hours during the morning of February 29, 2000. The LCO manager, assistant managers of administration, field operations, and recruiting were present, as well as the office's automation technician. Also present were the Dallas Assistant Regional Census Manager (ARCM), the Partnership Specialist and the Media Specialist responsible for the office. Representatives of the Bureau of the Census Decennial Communications Division and the Department of Commerce Economic Affairs Office were in attendance.

El Paso is responsible for enumerating a vast area on the western edge of the Texas / Mexico border. The area includes 10 counties covering 25,855 square miles. Terrel County, the most distant county in the LCO area, is 300 air miles from the El Paso office. The 1990 census population was 616,713.

This area includes a large and growing immigrant and Spanish-speaking population, including monolingual Spanish communities and colonias (see box). The large Spanish-speaking population creates a demand for in-language recruiting and promotional materials, as well as enumerators fluent in Spanish. In addition, El Paso has recruited and plans to hire cultural facilitators to accompany enumerators in monolingual Spanish areas. The office has developed extensive plans and partnerships to improve enumeration in colonias.

A PROFILE OF THE HARD-TO-COUNT: COLONIAS

The official definition of the colonia, "a residential subdivision lacking essential facilities such as water and wastewater services and paved roads," only begins to describe life in a colonia.

Colonias are found in the states along the US-Mexican border - Arizona, California, New Mexico and Texas. Most of the US colonias are in Texas - over 1200 colonias with approximately 300,000 residents. Almost 90 percent of the colonias in Texas are located in Hidalgo, El Paso, Starr, Cameron, Maverick, Webb and Zavala counties - 251,032 persons who are 90 percent Hispanic and often recent immigrants.

Cameron Park, one of the oldest colonias and featured on national news programs, has become a "model" colonia. Due to the advocacy of the community, the roads in Cameron Park are paved, water is available and there are services available in the colonia. In fact, within the maze of streets there is a community health center, several businesses and a community center.

Yet, Cameron Park is not illustrative of the main. Most colonias are like Del Mar Heights, La Paloma, Blanca Juarez, Sunny Side, Huecho, Montana Vista, Rio Bravo and Basham 23. The residents of these colonias are geographically, culturally and linguistically isolated and struggle to obtain water for their daily needs and even to travel beyond the colonia.

Hard-to-Enumerate Planning

Census and survey coverage is frequently depressed in Latino neighborhoods, particularly in immigrant and monolingual Spanish communities. In addition, El Paso is classified as a "Type D" office, responsible for enumerating largely rural geography. Language and cultural barriers, difficulties of enumerating a large area, as well as under-developed colonias, present formidable challenges to the El Paso enumeration. The Planning Database (PDB), using data from the 1990 census, identifies 50 of the office's 118 census tracts as hard-to-enumerate (HTE). However, much of the area has experienced high growth and immigration in the last decade. In particular, many of the HTE immigrant communities and colonias did not exist ten years ago. As such, 1990 data is not likely to provide an accurate picture of local areas, particularly those at greatest risk of high undercounts.

LCO staff combined local knowledge with the Bureau's PDB to develop special enumeration plans for HTE areas. They partnered extensively with city, county and state governments, and drew on resources of Texas A&M University to collect detailed, recent demographic and geographic information. The result is a systematic plan anticipating and locating specific barriers to enumeration, and listing the LCO enumeration strategies customized to overcome each barrier. This plan was developed in June 1999 (ahead of the national schedule for HTE action plans).

Specific examples of strategies cited during interview included locating a Questionnaire Assistance Center in each HTE tract, employing cultural facilitators in monolingual Spanish neighborhoods, using pair and team enumeration, and establishing aggressive efforts to hire bilingual enumerators.

Monitoring Board staff requested copies of the plan, which would provide greater detail of the Bureau's efforts to improve enumeration in areas at risk of extreme undercounts, and were refused by the representative of the Bureau of the Census Decennial Communications Division.

In the colonias, questionnaires will be hand-delivered to each household by an enumerator during the Update/Enumerate, Update/Leave, or List/Enumerate operation. These three operations comprise approximately 15 percent of the LCO workload. Cultural facilitators will accompany enumerators in colonias.

El Paso management has identified monolingual Spanish communities where it would be appropriate to deliver a Spanish - rather than English - questionnaire. However, the Bureau's national policy is to distribute English questionnaires to all households, unless an in-language questionnaire is specifically requested. As a result, during Update/Leave, scheduled to begin March 3, 2000, El Paso enumerators were to distribute English-language questionnaires, and Spanish Language Assistance Guides. The Update/Leave operation comprises 7 percent of the LCO workload.

During Update/Enumerate, scheduled from mid-March to May, a local census enumerator accompanied by a cultural facilitator will visit each colonia household and assist residents in completing an English-language census questionnaire. Enumerators and cultural facilitators will employ Spanish language guides during this operation, but Spanish questionnaires will not be used.

The Board agrees that delivering non-English questionnaires where appropriate would be of great benefit to a full enumeration. However, members disagree on the feasibility of implementing this policy during the 2000 census. For full discussion of this issue, see discussion beginning on page 25.

Recruiting and Staffing

El Paso had surpassed recruiting goals at the time of interview. Bureau records show 6,048 qualified applicants as of February 24, 2000: 124 percent of the goal. For non-response follow-up (NRFU), the most labor-intensive census operation, El Paso anticipates hiring 600 enumerators.

In addition, El Paso plans to hire local residents as cultural facilitators in colonias to accompany enumerators. Management reported recruiting 200 cultural facilitators at the time of interview. Cultural facilitators will not be required to take the enumerator test and will be paid separately from the PAMS/ADAMS payroll system.

Partnership

El Paso's Partnership Specialist has worked closely with LCO management on most aspects of planning. The formidable outreach campaign includes contact with city, county and state officials, media, academic institutions, businesses, and professional, community-based and religious organizations. Partnership has been especially active in coordinating local partners to develop HTE action plans.

Documentation provided to the Board details more than 1,000 contacts via presentation, site visits, phone, meetings and other channels. Contacts resulted in an estimated 2,350 commitments from local partners, each listed under nearly 30 categories. Types of commitment include endorsements, recruiting assistance, donation of free space, volunteers, and service as Questionnaire Assistance Centers and Be Counted sites.

As of February 1, 2000, El Paso listed nine Complete Count Committees (CCCs) as active or created in the area. Partnerships had been established with the Texas A&M Colonias Program, El Paso County, the local Catholic Diocese, the Displaced Workers Volunteer Group, the Islera Tribe, McDonalds and local radio and television.

El Paso's partnership efforts have succeeded despite some unique logistical challenges. The size of the El Paso LCO area, and the lack of air routes within that area, present major barriers. Partnership and LCO staff conduct "Round Robins": road trips with stops throughout the LCO area. A Round Robin might cover several counties, and several hundred miles, in two days.

Questionnaire Assistance Centers (QACs) and Be Counted Sites

LCO staff selected 41 QACs. Most, 37, are located in El Paso County, where the vast majority of the area's population is concentrated. The partnership specialist and local partners assisted in site selection. QACs were scheduled to open March 8 and will have paid and volunteer staff. Due to the hand-delivery of questionnaires to colonias, no Be Counted sites were planned for those areas. Management reported no problems receiving materials.