Cover Image

Not for Sale

View/Hide Left Panel
Click for next page ( 19

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement

Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 18
18 A Guidebook for Using American Community Survey Data for Transportation Planning The effects on the published data of implementing rules to ensure compliance with this law are discussed in Section 4.3 of this guidebook. 2.1.4 Data Product Dissemination The many ACS data products are made available to users via the Census Bureau's website "American FactFinder" page. American FactFinder provides access to data and products related to the Census Bureau's Decennial census, ACS, Economic census, Annual economic surveys, and Population estimates program. The process of locating and obtaining ACS data from American FactFinder is described in the next section of this guidebook. 2.2 Additional Information Sources on ACS Implementation The key source of ACS operations and implementation information is, of course, the Census Bureau. The website includes the Operations Plan, as well as other documents that summarize different aspects of ACS implementation. The Census website also includes several archived documents that provide a useful perspec- tive on how ACS was conceived and how it has developed. In addition to providing a historical record, these documents provide insights into how the survey may evolve over time. For these purposes, we recommend the following documents described in the remainder of this section. United States Government Accountability Office, "American Community Survey: Key Unre- solved Issues."8 The Government Accountability Office report on ACS noted the following as "key unresolved issues": 1. Introduction of a new concept of residence: "Sufficient research has not been conducted to make the final set of rules for the `current residence' used for ACS." 2. Uncertainty about the new methodology for deriving independent controls for population and housing characteristics: "The Census Bureau has not developed a methodology for using the Intercensal Population Estimates (ICPE) program for the full ACS to derive controls consistent with the ACS residence concept and ACS reference period, or at the same level of geography used for the Census 2000 Long Form." 3. Lack of guidance for users on the characteristics of multiyear averages for small geographic areas: "Because of statistical properties of multiyear averages and users' unfamiliarity with them . . . it is critical for Census Bureau to provide users with guidance on topics such as reliability of multiyear averages for areas with rapidly changing populations, reliability of trends calculated from annual changes in multiyear averages, and the use of multiple esti- mates from ACS data for geographic areas with populations greater than 20,000." 8 GAO-05-82, October 2004. See

OCR for page 18
American Community Survey 19 4. Operational procedures, such as questionnaire design, and adjustment to dollar-denominated values, and to the consistency between ACS and Census 2000 data. Alternatives to improve small geographic area data: An alternative to provide more reliable small area data is to additionally fund a larger sample for 2009-2011, and provide a replacement for the Long Form one year earlier. Barry Edmonston and Charles Schultze, editors, "Modernizing the U.S. Census."9 This study provides a review of the traditional U.S. census; considers ways to improve cover- age, reduce differential undercount, and limit enumeration; examines needs for small area data during intercensal years, and explores the use of sampling methods. It recommends ways to improve initial response rates for both Short and Long Forms, examination and testing of ques- tions on race and ethnicity, use of continuous measurement and other methods to obtain small area data, reduce costs, and suggests a new design for the census questionnaire. Several recom- mendations are related to the improvement of MAF/Tiger, and development of intercensal esti- mates for small areas. Alternatives to the decennial census--such as use of administrative records, a national regis- ter, and a rolling census--also are presented. The panel evaluated the uses of Long Form data to arrive at the conclusion that ". . . in addi- tion to data to satisfy constitutional requirements, there are essential public needs for small area data and data on small population groups of the type and breadth now collected in the decen- nial census." In the panel's judgment The Long Form is not responsible for the decline in response rates or increase in costs in the previous censuses, Dropping the Long Form would not have a very large effect on response rates, and The inclusion of the Long Form questionnaire for a large sample of households is a cost- effective way of obtaining highly valuable information. Daniel L. Cork, Michael L. Cohen, and Benjamin F. King, editors, Panel on Research on Future Census Methods, National Research Council, "Reengineering the 2010 Census: Risks and Challenges."10 This study examined the Census Bureau's current plans for a reengineered Census 2010 with MAF/TIGER enhancements, American Community Survey, and Early Integrated Planning as its core concepts. The panel strongly supported the major aims of the Census Bureau's emerging plan for 2010, while noting that considerable challenges must be overcome for the innovations to be successful. Specifically, the panel noted that the Census Bureau should Develop a sound evidentiary base for its 2010 census plan. Identify, articulate, and quantify risks in the census process (especially the impact of reduced funding on the quality of ACS estimates for small area data). The panel espe- cially noted the need for a clear and early decision on ACS and contingency plans for the traditional Long Form if full ACS funding were not forthcoming. Other ACS issues that concerned the panel included the collection of Group Quarters data, the risk of a 9 Barry Edmonston and Charles Schultze, eds., "Modernizing the U.S. Census." Washington, D.C., National Academies Press, 1995. See 10 Daniel L. Cork, Michael L. Cohen, and Benjamin F. King, editors, Panel on Research on Future Census Meth- ods, National Research Council, "Reengineering the 2010 Census: Risks and Challenges." Washington, D.C., National Academies Press, 2004. See

OCR for page 18
20 A Guidebook for Using American Community Survey Data for Transportation Planning voluntary versus mandatory response, interaction with intercensal population estimates and the demographic analysis programs, and the use of sequential hot-deck imputation for the treatment of individual non-response. Develop a comprehensive plan for updating and improving the MAF. The panel notes that each of the tasks related to modernization of TIGER carries considerable risk-- especially the timeliness of realignment of TIGER geographic features to be consistent with GPS coordinates and the conversion of MAF/TIGER from its current homegrown format to a modern object-oriented computing environment. Work with the postal service in assessing the quality of the Delivery Sequence File. Analyze the Community Address Updating System; and justify plans to implement a complete block canvas. Constance F. Citro, Daniel L. Cork, and Janet L. Norwood, editors, Panel to Review the 2000 Census, National Research Council, "The 2000 Census: Counting under Adversity."11 The panel's overall conclusion was that "Census 2000 experienced both major successes and significant problems." The successes pointed out in the report are the completeness of demo- graphic coverage and the quality of basic demographic data. Census 2000 saw a halt to the decline in the mail response rates, and operations were conducted in a timely manner. Net undercounts were lower in Census 2000 than in the 1990 Census. The problems cited included errors in the MAF, a large number of duplicates, problems with some Long Form items such as employment and income, and inaccuracies in the enumeration of Group Quarters population. The panel found that "census counts at the block level, whether adjusted or unadjusted, are subject to high levels of error, and hence should be used only when aggregated to larger geogra- phies." The lack of agreement until 1999 on the basic design hampered planning and increased costs for Census 2000. The panel recommended that the Census Bureau, the administration, and Con- gress should agree on the basic design for Census 2010 and the ACS by 2006. In its assessment of Census 2000 operations, the panel found "limited pieces of evidence to suggest some problems in the imputation of whole persons." An administrative records experi- ment conducted in five counties showed that 41 percent of imputed census households were larger in size than linked administrative households, while 27 percent were smaller. "Missing data rates for some Long Form items were high in many cases; in some cases, higher than the comparable rates in 1990. The Census Bureau relied on imputation of these items on procedures that it used for many censuses with little evaluation of their appropriateness or effectiveness." Also determined was that "The Census Bureau should conduct experiments to test the relative costs of more imputation versus more follow-up before deciding whether to continue the 2000 strategy in 2010." For the household population, missing data item rates were high (10 percent or more) for over one-half of the Long Form items, and very high (20 percent or more) for over one sixth of the Long Form items. Given these high rates of imputation, the panel recommended that the Cen- sus Bureau develop procedures to quantify and report variability of 2000 Long Form estimates, further study the effects of imputation, and conduct research on improving imputation meth- ods for ACS (or the 2010 Census if it includes a Long Form). 11Constance F. Citro, Daniel L. Cork, and Janet L. Norwood, Editors, Panel to Review the 2000 Census, National Research Council, "The 2000 Census: Counting under Adversity." Washington, D.C., National Academies Press, 2004. See

OCR for page 18
American Community Survey 21 With respect to the MAF, the panel recommended that the Census Bureau develop procedures to accurately identify housing units within multi-unit structures, redesign the Local Update of Census Addresses (LUCA) program to benefit participating state and local governments, and plan evaluations of MAF well in advance of the 2010 Census. The panel also recommended the development of an improved Accuracy and Coverage Evaluation (ACE) program for the 2010 Census.