Enumerating the Colonias: A Case for Better Communication and Greater Reliance on Local Knowledge
Communities along the Texas-Mexico border that are home primarily to immigrant workers from Mexico and their families − collectively and commonly known as "colonias," Spanish for "neighborhoods" − present one of the most difficult enumeration environments for the U.S. Census Bureau. The bureau describes these areas as "generally unincorporated and low-income residential subdivisions, lacking basic infrastructure and services," noting that while the settlements have existed for decades, poorer residents from border cities such as El Paso and Brownsville began moving to the colonias in greater numbers in the 1980s and 1990s, drawn by the lower cost of land and housing.32 According to the state of Texas, there are roughly 2,300 colonias in Texas that are home to a predominantly Hispanic population of 400,000, the majority of whom were born in the United States.33
The Census Bureau conducted a qualitative evaluation of 2000 census operations in the colonias as part of its broad Census 2000 Testing, Experimentation, and Evaluation (TXE) Program to assess the completed decennial count and inform planning for the 2010 census.34 Researchers identified several major barriers to achieving a complete count in these border communities but found that these challenges varied in degree from settlement to settlement. Those barriers included:
- Irregular housing and addressing;
- Limited English proficiency and limited formal education;
- Concerns regarding confidentiality of census responses; and
- Complex households and households with mobile or transitional members.
Taken as a whole, this range of characteristics clearly placed the colonias among geographic areas historically considered hard-to-count, and the Census Bureau designated the settlements as "Update Enumerate" (U/E) areas − places where it employs modified census procedures to enumerate the population.
Because the decennial census is an address-based survey, designed to count people at a specific geographic location (such as an individual housing unit or group facility) as of Census Day (April 1), the prevalence of irregular housing and addressing in a community makes the widely used mail-out/mail-back method of gathering census responses highly problematic. Consequently, the Census Bureau developed alternative methods for delivering census forms and gathering information from residents in areas less conducive to traditional mail operations. The bureau uses the U/E method primarily in remote locations with irregular housing (for example, missing or inconsistent house numbers, lack of identifiable street names, hidden housing units, or unusually subdivided housing units) or inconsistent postal delivery, and with high rates of mobility. In U/E areas, residents do not receive a standard census questionnaire, either by mail or hand-delivered. Instead, during a designated time period, census enumerators visit every home in a community, confirming each unit's address and exact location on census maps and collecting responses from residents to the same questions asked on the printed census form. For the 2010 census, the U/E operation ran from March 22 through May 29, 2010.
In addition to the colonias, the Census Bureau designated most American Indian reservations, Alaska Native villages, migrant worker communities, and remote seasonal resorts as U/E areas.
The Census Bureau concluded after the 2000 count that U/E procedures were particularly effective in the colonias, saying in its evaluation that, "census enumerators were able to successfully negotiate the obstacles presented by irregular housing" using the alternate method.35 Census researchers also credited the limited use of "cultural facilitators" − local residents hired on an as-needed basis to help official census takers make contact with hard-to-identify housing units and reluctant households − and promotoras – local advocates hired to encourage participation – as well as paid advertisements in Spanish language media, with helping to overcome significant barriers in the colonias in 2000.
Census Bureau staff made several recommendations to improve enumeration of the colonias in 2010 based on their evaluation of 2000 census operations.36 Foremost among their suggestions was an expanded use of cultural facilitators and promotoras who are intimately familiar with the population and culture of the colonias, to work alongside census enumerators. (The Census Bureau used facilitators on a limited basis in 2000 in El Paso County colonias.) They also recommended testing the use of Spanish language census forms and enumerator data-gathering forms, as well as revised training methods and materials that focus more on field and on-the-job training than classroom instruction.
Census Director Robert Groves, who took office in July 2009 well after final preparations for the 2010 count were in motion, appears to have recognized the unique challenges the Census Bureau faced in the border region. In February 2010, the director visited the San Carlos colonia in Laredo at the invitation of Rep. Henry Cuellar, D. Texas. An Associated Press article described San Carlos as a community with "unpaved roads" and "ramshackle houses," home to many undocumented immigrants and Spanish-only speakers, with a high incidence of poverty.37 Dr. Groves' mission was to reassure anxious residents about the strict confidentiality of census responses. Rep. Cuellar emphasized that the census did not ask about immigration status and highlighted the potential loss of needed federal resources and political representation in areas with a census undercount.
Knowing the stakes for the communities they serve, many nonprofit organizations in the Rio Grande Valley eagerly promoted the census and acted as much-needed "trusted voices" among residents who had reason to fear or mistrust federal authorities, especially at a time of heightened anti-immigrant sentiment in many areas. Community-based and regional organizations conducting census outreach campaigns included Proyecto Libertad, Proyecto Azteca, Lupe, Southwest Workers' Union, Project ARISE, and the Frontera Asset Building Network. The Equal Voice for America's Families network partnered with the Rio Grande Guardian newspaper to promote the count.
These and other organizations received grants from various regional and national funders for census outreach in hard-to-count communities along the U.S.-Mexico border; some of the larger networks made subgrants to smaller nonprofits throughout the Valley. The Center for Housing and Urban Development at Texas A&M was one of several programs sponsoring local workers, or promotoras, to conduct outreach in the colonias in advance of the start of the census.
Community organizers sponsored neighborhood meetings, distributed census materials, and canvassed blocks in the colonias, serving voluntarily as visible, trusted advocates for the Census Bureau in some of the nation's most difficult to enumerate areas. The Frontera Asset Building Network38 educated its vast network of community partners through workshops, an online information clearinghouse, and town hall meetings.
Activists also engaged schools and Spanish-language television and radio outlets. Parents of school children received calls with a taped public service announcement promoting census participation. Local media featured numerous articles on the census, highlighting the political and fiscal benefits that would flow to the region based on an accurate census count, particularly for schools and emergency services. As census field operations unfolded, the newspaper documented efforts in Spanish-language media (especially Univision and Telemundo) to promote participation in the Valley and reassure people that their responses would remain confidential. Most notably, local outreach messages highlighted the presumed census plan for the colonias — that residents would receive forms in the mail in mid-March, fill them out in the privacy of their own homes, and mail them back postage-free.
Enthusiasm for the upcoming census, however, turned first to dismay, then frustration, and finally anger, when it became clear that a significant lapse in communication between the Census Bureau and its community partners had led advocates to believe that the bureau would enumerate the colonias using the most common methods: mailing or hand-delivering questionnaires to households and asking residents to mail them back.39 Instead, as most Americans began receiving forms at their homes in early and mid-March, community leaders in the Valley discovered that households in the colonias had not received theirs. Working through funders, the advocates learned from Leadership Conference Education Fund staff that the less-well-known U/E operation would be used in the colonias, starting in late March and continuing through May.
The reaction from local advocates was swift and clear: they believed that the Census Bureau's failure to fully explain its enumeration plans for the colonias — in fact, its very selection of an alternative counting method — signaled a lack of respect for residents and a poor understanding of life in the colonias. Mike Seifert, a well-known community leader with the Equal Voice for America's Families network, concluded in an op-ed that using the U/E method was "demeaning," writing that the Census Bureau believed "colonia residents are incapable of filling out a census form."40
As word of the alternative census method planned for the colonias spread quickly among grassroots organizations, elected officials, and journalists, whatever trust and confidence community leaders had in the Census Bureau vanished virtually overnight. Knowledgeable service providers who worked daily with residents within the colonias found it difficult to believe that enumerators could visit all homes in the many neighborhoods comprising these vast settlements; they cited additional barriers such as heightened fear of strangers and the prevalence of large dogs. Another discouraged advocate told The Education Fund, "[It] sounds almost like … ‘we don't count'"
More specifically, civic leaders believed the Census Bureau had misled them with respect to enumeration methods, especially since some homes in the colonias had traditional postal addresses and received mail regularly through the U.S. Postal Service. (In fact, the Census Bureau did conduct either mail-out/mail-back or Update/Leave operations in some Texas colonias.) Members of Congress and local elected officials representing the colonias told media outlets that they, too, were unaware that the Census Bureau would count residents using a different method. Community leader Armando Garza, development director for the nonprofit Proyecto Azteca, told The Monitor newspaper in McAllen, Texas, "Had we known in November [2009] that 95 percent of the colonias would not be receiving a mailed form, our strategy would have been completely different."41 Their own consternation aside, civic leaders also feared that their credibility as trusted advocates for these struggling but hopeful communities was at risk, having painstakingly laid the groundwork for a conventional counting operation that would not occur. In a letter to Census Director Robert Groves, Hidalgo County Judge Rene Ramirez said, "[W]e have been telling all residents to expect a census form in the mail on or before April 1. This latest news hurts our credibility with the public, especially among those hard-to-count areas."42
It was The Education Fund's 2010 Census Campaign collaborative — with the National Association of Latino Elected and Appointed Officials (NALEO) Educational Fund taking the lead — that first brought the communications breakdown regarding enumeration of the colonias to the attention of Dallas Census Regional Office officials. Senior Census staff at both the regional and headquarters level seemed surprised to learn of the misunderstanding, especially since the bureau had employed the same method in the colonias for several decades; they believed they had taken steps throughout the period of census preparations to adequately explain the U/E method to community partners and elected officials alike. The regional office issued a press release on April 1 (Census Day), describing what it called "special efforts" to enumerate the colonias and calling the U/E method "the most accurate field operation."
Regional officials explained to Education Fund staff that there was no advertising about the U/E operation because too many other areas of the media market covering the Texas colonias, as well as some portions of the colonias themselves, were covered either by the mail-out/mail-back or Update/Leave methods. Spanish-language media, however, disputed this concern, noting the presence of radio and print outlets that focus almost exclusively on the colonias. Confusion also reigned about the role of Questionnaire Assistance Centers and Be Counted sites in the Rio Grande Valley; these Census Bureau-operated support activities were designed to assist people in communities where residents were supposed to receive census forms at their homes, but they were not prepared to help people who lived in U/E areas.
As the extent of the operational misunderstanding and subsequent breach of confidence became clear, Rep. Rubén Hinojosa, D. Texas, held a meeting with leading community nonprofits at his Edinburg district office to discuss their concerns about the accuracy of the census in the valley. The organizations and the congressman issued several recommendations, including a request that the Census Bureau establish a task force to monitor the remainder of census operations in the colonias and other hard-to-count communities in the valley. Civic leaders also highlighted their ongoing concerns about the bureau's ability to hire sufficient numbers of bilingual enumerators familiar with the unique terrain of the colonias.43 In fact, local leaders had been anxious about recruitment and hiring for the colonias enumeration for several months; Rep. Hinojosa said he had asked Dallas Regional Director Gabriel Sanchez for weekly hiring reports starting in mid-February. In mid-March, Census headquarters staff tried to reassure the congressman that the agency was meeting recruitment goals and hiring targets in the valley. By the beginning of April, however, as the U/E field work commenced, regional Census staff acknowledged to the Rio Grande Guardian that recruitment of qualified applicants familiar with local communities was lagging in Hidalgo County, even as recruitment efforts were successful for the colonias enumeration in Laredo and other border communities.44
Meanwhile, Dallas regional census officials hastily arranged meetings with congressional offices and business leaders in the valley, as well as with community organizations. They also addressed concerns about enumeration plans and the status of census operations generally through interviews with local media, including Spanish language radio.45 Based on these meetings and subsequent conference calls, many of which The Education Fund collaborative team helped organize, the bureau agreed to extend the Census Road Tour in the valley and the colonias through May 8; to consider additional advertising to promote the U/E operation; to provide maps of U/E assignment areas to community service organizations; and to hire more cultural facilitators to assist enumerators in gaining the trust of residents during door-to-door interviews. The Education Fund and NALEO urged senior census headquarters officials to authorize more paid advertising targeting the colonias immediately in order to help bridge the communications gap that now threatened the success of the U/E operation in these areas. Census staff moved quickly to accommodate this request, creating a new advertisement to explain the U/E process and moving up the start date of Nonresponse Follow-up advertising on local media to let all households know that census enumerators would soon be on the ground throughout the region, regardless of the counting method used. Community leaders also requested more detailed information on the enumeration schedule for each colonia, which could inform their own continuing outreach efforts and give their staff time to prepare residents for an enumerator visit during the U/E period.
Foundations supporting community-based census outreach offered encouragement to their grantees and provided a forum for groups to share information and ideas to salvage their promotional activities. Despite their disappointment and frustration, local advocates mobilized quickly to revise their outreach messages and materials and to inform their constituencies about the U/E process. New messages included descriptions of U/E household visits, with an emphasis on data confidentiality and reminders that all people living in a household on April 1 should be reported. The Education Fund prepared a fact sheet on the U/E operation, outlining the exact responsibilities of enumerators going door-to-door and suggesting messages that community advocates could use to ease fears among residents and encourage their cooperation with census takers.
While these initial steps helped to ensure that census advocates understood the enumeration plans for the colonias, it was clear that it would not be easy to restore the confidence of local leaders in the Census Bureau or in the likelihood of an accurate census in the colonias. The NALEO Education Fund (NALEO), a partner in The Education Fund's 2010 census collaborative, offered to serve as a bridge and mediator between Census Bureau officials and community organizations as the count progressed.46 NALEO assigned several staff members to focus almost exclusively on assisting community advocates with their census campaigns, monitoring the progress of the count in the colonias, and ensuring that Census officials addressed community concerns in a timely and thorough manner. At NALEO's suggestion, the Census Bureau assigned a senior level, bilingual staff member from the Seattle Regional Census Office to work exclusively with partner organizations and elected leaders in the valley as the enumeration progressed, helping to answer questions and provide requested information quickly. Also at NALEO's urging, Census Director Groves again visited the valley on April 17, as the mail phase of the national count was winding down and the Census Bureau was preparing for the massive Nonresponse Follow-up operation starting May 1.47 In many Valley communities designated as mail-out/mail-back areas — including Rio Grande City in Starr County, where the director appeared with Rep. Cuellar, Mayor Ruben Villarreal, and other local leaders — participation rates were disturbingly low, foreshadowing the enormity of the challenge to complete the count in all communities, regardless of the type of method used. The director said the bureau had hired additional staff in the region, and he urged residents who had not yet been counted to "react in a favorable way" when an enumerator visits.48
Alarm about census operations in colonias along the entire U.S.-Mexico border extended to Washington, D.C., where the Congressional Hispanic Caucus (CHC) sent a strong letter of concern to Director Groves on April 19. Led by Chairwoman Nydia Velazquez, D. N.Y., the CHC noted that many colonias residents were just learning that they would not receive census questionnaires in the mail, and it questioned the Census Bureau's ability to hire a sufficient number of qualified, local enumerators to enumerate the colonias door-to-door. The caucus made several recommendations to help ensure a more accurate count in these communities, including extending the U/E period beyond May and into early July, if necessary; working collaboratively with local organizations and media to provide advance notice of visits neighborhood by neighborhood; and keeping Be Counted sites open in the colonias and rural border areas through July 10, the scheduled end of Nonresponse Follow-up.49
As the U/E operation continued in the colonias, community leaders continued to express skepticism that the Census Bureau could achieve a comprehensive count of residents through a door-to-door count. Local nonprofits partnered with NALEO to establish a hotline for residents to call if they did not receive a visit from an enumerator or to report other concerns about the enumeration. NALEO continued to act as an intermediary between community advocates (who had formed their own census task force), elected officials, and the Census Bureau, arranging meetings, sharing information and concerns, and suggesting solutions to challenges the Census Bureau encountered in the field. Members of Congress representing the valley urged constituents in all communities to cooperate with enumerators who were going door-to-door.
Publicly, regional Census Bureau officials remained confident about the agency's ability to fully count the colonias, pointing to the hiring of more bilingual enumerators and the use of cultural facilitators to help census workers gain the trust of residents. They pleaded for patience and urged people to wait until the end of May for a visit before turning to the toll-free telephone response option.50 By late May, the Dallas regional office told local leaders that they had counted more than 97 percent of Rio Grande Valley households, crediting a positive response from residents and assistance from community organizations. "We are very confident we will have an accurate count in the valley," Hector Moldonado, the Bureau's appointed high-level valley liaison, told the local newspaper.51 Mr. Moldonado also noted that the Bureau was still counting other valley communities through the Nonresponse Follow-up operation and emphasized that continued cooperation in low mail-back areas was essential. While U/E operations officially ended on time in the colonias, Census officials agreed to revisit neighborhoods where residents reported they had not been counted.
* * *
The Education Fund believes the Census Bureau relied too heavily on historical practice and its own research in planning for enumeration of the colonias. The bureau failed to consult regularly with local advocates most familiar with the conditions and population of these areas and to consider thoroughly the use of a different counting method in light of those conditions. For example, local migrant worker advocates told The Education Fund that many farm workers leave their homes in the colonias by mid-April, looking for work in other parts of the country; entire families often migrate north, boarding up their homes or leaving rental units and pulling their children out of school. With the U/E process continuing throughout April and May, enumerators would likely miss many farm workers and their families during their door-to-door visits.
The Census Bureau's failure to adequately explain, well in advance, the alternative counting method planned for the colonias left the agency open to a widespread perception among national and community stakeholders that it viewed the residents of these unique settlements as less capable of participating in the census, and the communities as less important than others. In its April letter to Director Groves, for example, the CHC accused the bureau of being "inflexible in its efforts and methodology in the case of the heavily Hispanic U.S.-Mexico border, especially South Texas." Hispanic lawmakers pointed to operations and methods employed in various other communities across the country, such as mailed replacement questionnaires in low mail-response areas and a special, time-limited counting operation to reach homeless populations. Mike Seifert of the Equal Voice for America's Families network concluded that the decision not to mail census forms to most colonias households suggested that the Census Bureau did not "trust" residents to fill out the form.52 Another advocate for low-income and immigrant populations in the border region told funders that colonias residents were "insulted" by the decision not to mail questionnaires to their homes; they perceived the "different" treatment as an indication of "second class" status in the census process. "Involving the community upfront on this decision would have gone a long way," this advocate observed.
A large number of nonprofits with both regional and local portfolios were engaged in outreach and mobilization in the months leading up to the 2010 Census. The activities appear to have been well suited to conveying information about the importance of census participation and the confidentiality of census responses to people whose experiences and circumstances often made them distrustful and even fearful of government authorities. Through the colonias program at Texas A&M University, community outreach workers (promotoras) included information about the census in their routine outreach to households in the colonias. Program staff member Laura Trevino highlighted the importance of having familiar faces communicate census messages in these communities. "Trust is fundamental," Ms. Trevino told the Valley Morning Star. Colonias residents "trust [the promotoras] because they're not there episodically. They're here on a consistent basis."53
Dozens of nonprofits in the border region, including many in the Rio Grande Valley, used a range of forums to educate residents about the importance and safety of census participation. Activities included partnering with school districts to distribute flyers to families with children; conducting outreach and offering questionnaire assistance at free tax preparation sites (Volunteer Income Tax Assistance program sites), job fairs, and emergency food distribution sites; posting easy-to-access information about the census on the websites of trusted, local organizations; and speaking about the census at church services and distributing promotional materials through church publications. NALEO partnered with the extensive Frontera Building Asset Network (FBAN) to host a training session for advocates in hard-to-count areas who could help promote the census in their communities. FABN also reported that many recipients of its small grant awards used educational and graphic materials from The Education Fund's Civilrights.org 2010 Census website and the NALEO-sponsored ya es hora. HAGASE CONTAR! campaign in their outreach activities.
Partnering with community groups and local institutions is an essential component of an accurate census in all hard-to-count areas. The colonias along the U.S.-Mexico border, and settlements that are home to farm workers in other parts of the country, present especially stark examples of the need to include these partners in planning and decision-making much earlier in the census process. The distinctive conditions and natural wariness of residents in these communities require hands-on assistance from local leaders and service providers who are uniquely positioned to advise the Census Bureau on counting challenges and to serve as "trusted voices" throughout the enumeration. The bureau must take steps to fully educate these indispensible partners about census plans and preparations throughout the decade and offer them reasonable opportunities to inform the planning process through various forums for sharing and exchanging ideas and knowledge.
Previous Section: Mississippi Gulf Coast and Delta | Main Page | Next Section: Recommendations
32. De la Puente, Manuel, and David Stemper, "The Enumeration of Colonias in Census 2000: Perspectives of Ethnographers and Census Enumerators," U.S. Census Bureau, Statistical Research Division, September 22, 2003, p. 1. (http://www.census.gov/pred/www/rpts/J.4.pdf).
33. "Colonias: Frequently Asked Questions," Texas Secretary of State, http://www.sos.state.tx.us/border/colonias/faqs.shtml.
34. See, in general, ibid. This evaluation relied on ethnographic studies and focus groups and covered four colonias spread across California, New Mexico, and Texas; the authors acknowledge that the chosen sites were not intended to be statistically representative of all colonias.
35. Ibid, pp. ii.
36. Ibid, pp. 21-22.
37. Weber, Paul J., "Census chief tries easing immigrant fears in Texas," The Associated Press, February 1, 2010
38. The Annie E. Casey Foundation funded Frontera Asset Building Network's Census 2010 Initiative.
39. The Census Bureau used three primary enumeration methods in 2010; it employed the same methods in 2000. "Mail-out/mail-back" covers most of the country; census forms are mailed to residential addresses, and residents mail them back. "Update/Leave" is used in rural and other communities without traditional city-style addressing; census workers hand-deliver (e.g. leave) census forms to residential addresses and update the address list and maps as they go, to help ensure full coverage and correct geographic location of homes.
40. Seifert, Mike, "Census Bureau's New Policy Demeaning to Colonia Residents," Rio Grande Guardian, April 2, 2010.
41. Jones, Jared, "Census Bureau's counting procedure for colonias frustrates some," The Monitor, April 8, 2010.
42. "Judge concerned about colonia Census count," Valley Morning Star, April 4, 2010.
43. "Border Congressman Calls For a Colonia Census Task Force," Rio Grande Guardian online, April 2, 2010.
44. Ibid.
45. Education Fund staff suggested civic leaders and community organizations in the Valley area to include in meetings with Dallas Regional Census Office officials.
46. The NALEO Educational Fund is a member of the Census Bureau's 2010 Census Advisory Committee.
47. The Census Bureau reported that Director Groves visited the Rio Grande Valley more than any other community in the nation during the 2010 census.
48. "Census director visits Starr County as questionnaire mailback period ends," Monitor (Naples, TX), April 19, 2010.
49. Be Counted sites and Questionnaire Assistance Centers were scheduled to close nationwide on April 19. These sites were not available in U/E areas because enumerators are responsible for answering questions about the census as they go door-to-door.
50. "Nonprofit Groups Fear Census Undercounting in Colonias," KRGV Channel 5 News online, April 27, 2010.
51. "97.4 Percent Census Participation Rate in Valley Under U/E," Rio Grande Guardian, May 25, 2010.
52. Seifert, Mike, "Census Bureau's New Policy Demeaning to Colonia Residents," Rio Grande Guardian, April 2, 2010.
53. "Community outreach groups promote Census in colonias," Valley Morning Star, April 13, 2010.
