In this report:
- Acknowledgements
- Executive Summary
- Introduction
- Part I - Demographic Trends in Katrina-Affected Areas and Their Impact on the 2010 Census
- Part II - Census Procedures and Operational Challenges to Getting an Accurate Count in the Aftermath of a Catastrophe
- Part III - Operational and Policy Recommendations for a More Accurate 2010 Census in the Gulf Coast
- Appendix A
Addressing the Challenges of an Address-Based System
The Census Bureau deserves credit for recognizing the need to tailor counting strategies to the unique circumstances facing the still-recovering and transitioning areas hardest hit by Katrina. A significant example of this was the decision two years ago by the Dallas Regional Census Director to designate large swaths of Hurricane-impacted coastal Louisiana and Mississippi counties as Update/Leave, instead of Mail-Out/Mail-Back, areas.
The decision to treat hard-hit areas as Update/Leave rather than Mail-Out/Mail-Back areas will have a significant impact on the enumeration process. For most of the nation, the 2010 Census will start with receipt of an "advance letter" in early March, telling households that they will receive a census form in the mail shortly. Questionnaires are mailed the third week in March, followed by a thank you/reminder postcard a week later. Areas that have recovered more quickly from the 2005 storms, especially with regard to permanent housing, will continue to fall within the "Mail-Out/Mail-Back" universe.
The Update/Leave method, normally reserved for sparsely populated rural areas where homes are not easily distinguished by unique mailing addresses, will allow census takers to walk the streets/roads and update address lists a final time, as they deliver census questionnaires in person. In Update/ Leave areas, the census will start somewhat earlier than in Mail-Out/Mail-Back areas, with households receiving the advance notification letter in mid-February.
The Dallas Regional Census Office determined that portions of Hancock, Harrison, and Jackson Counties in Mississippi – where entire communities were not simply damaged but destroyed – had not recovered sufficiently from the storms to allow easy implementation of traditional mail-out/mail-back procedures. As in the southernmost Louisiana counties, the Census Bureau has designated several cities, including Waveland, Bay St. Louis, and Pass Christian, as well as the northern suburbs of Biloxi and some other communities, as Update/Leave areas.
That decision reflected a central challenging reality to counting the Gulf Coast region in 2010. The census is an address-based process, and the area's stock of residential housing is in extraordinary flux, with residential blight, rapidly changing housing stock, and proliferation of potentially "hidden" housing units, such as accessory apartments, posing a far greater challenge than in more stable communities with higher rates of home ownership, few unoccupied properties, and a predominance of single-family homes. Hurricanes Katrina and Rita left in their wake several factors that create enormous challenges for an address-based census design, most notably widespread destruction of homes and group living facilities, and a displaced population struggling to return home and establish stable households against significant obstacles.
Ironically, as noted in the preceding section, even encouraging demographic trends -- such as the fast pace of population growth and a steady increase in the number of households receiving mail – present difficult circumstances for the Census Bureau, which must finalize an initial address list for the enumeration months in advance of the start of the census. (The initial address canvass was completed by the Census Bureau earlier this summer.)
The New Orleans Index authors concluded that construction and rehabilitation of rental units and new homes in recovering neighborhoods "may also be sparking moves from one neighborhood to another" as better housing opportunities become available. Yet, high rates of mobility are a characteristic associated with HTC areas, according to Census Bureau analyses of mail response rates and coverage evaluation in previous censuses.23
Even though the number of households receiving mail in the New Orleans metro area has reached about 90 percent of its pre-Katrina number, the proportion in Orleans Parish is only about three-quarters. In addition, the relationship of mail receptacles to housing units in recovering communities could be too uncertain to ensure that every unique housing unit receives a census form by mail.24
Therefore, in 14 of the hardest-hit counties – all of Orleans, Plaquemines, and St. Bernard Parishes, as well as portions of eight other Louisiana parishes, and Hancock, Harrison, and Jackson counties in Mississippi – the Census Bureau will use the Update/Leave method,. Throughout the month of March, census workers will personally visit all addresses in Update/Leave communities, dropping off census forms and verifying the existence and location of each housing unit. The enumerators also can add new or hidden housing units they might find that are not already included on the master address list, an important additional opportunity to capture rebuilt and renovated homes that might not have been habitable, or simply were missed, during this summer's address canvassing.
It is too early to assess the results of the address canvassing operation, but the Commerce Department's Office of Inspector General (OIG) raised concerns early in the process about the thoroughness of address listing, including in rural areas. According to a May 2009 OIG Flash Report, its random observations of address canvassing revealed failures to knock on doors, as required, to ascertain the true status of a structure as a single or multi-unit dwelling; as well as troubling omissions in rural areas, where census workers "completely skipped roads in rural areas when they assumed no houses existed on the road."25 The report observes that rough roads, tree cover, and other conditions make the operation especially difficult in rural areas, but that the work is "essential to accurately locate rural living quarters."
While the OIG's report does not specifically refer to communities along the Gulf Coast, the rural characteristics prevalent in coastal counties in Louisiana, Mississippi, and Alabama may have made these areas vulnerable to less thorough address canvassing, especially if environmental damage from Hurricane Katrina increased the isolation of some rural housing.
Southern Echo, a leadership development and training organization focused on Mississippi's rural Black population, told LCCREF last spring that it was not confident local census offices were hiring address listers who lived in and were familiar with the areas to which they were assigned, particularly in rural communities where homes are far apart and strangers can easily get lost. Advertising for address lister positions was not evident, Southern Echo observed, until the local census office essentially had filled all of the jobs, and the on-line registration process was not accessible to many prospective applicants in poor communities. Failure to tap local knowledge in historically HTC areas – such as poor, predominantly Black, rural swaths of southern Mississippi – can adversely affect the quality of the address lists used in next spring's enumeration. Gaps in the final address lists could increase the effort required to locate rural homes when the enumeration starts and put these communities at risk of an undercount.
Next Section: Residence Rules and Displaced Residents
23. U.S. Census Bureau, Tract-Level Planning Database With Census 2000 Data <2010.census.gov/2010census/pdf/TractLevelCensus2000Apr_2_09.pdf - 2009-05-27>.
24. New Orleans Index, August 2009, pg. 8.
25. U.S. Department of Commerce, Office of Inspector General, OIG Flash Report, Census 2010: Observations and Address Listers' Reports Provide Serious Indications That Important Address Canvassing Procedures Are Not Being Followed, OIG-19636-01, May 2009.




