Reengineering the 2010 Census : Risks and Challenges

by Cork, Daniel L.; Cohen, Michael L.; King, Benjamin F.

• ISBN13: 9780309091893
• ISBN10: 0309091896
• Format: Paperback
• Copyright: 2004-03-30
• Publisher: Natl Academy Pr

Summary

At the request of the U.S. Census Bureau, the National Research Councilrs"s Committee on National Statistics established the Panel on Research on Future Census Methods to review the early planning process for the 2010 census. This new report documents the panelrs"s strong support for the major aims of the Census Bureaurs"s emerging plan for 2010. At the same time, it notes the considerable challenges that must be overcome if the bureaurs"s innovations are to be successful. The panel agrees with the Census Bureau that implementation of the American Community Survey and, with it, the separation of the long form from the census process are excellent concepts. Moreover, it concurs that the critically important Master Address File and TIGER geographic systems are in dire need of comprehensive updating and that new technologies have the potential to improve the accuracy of the count. The report identifies the risks and rewards of these and other components of the Census Bureaurs"s plan. The report emphasizes the need for the bureau to link its research and evaluation efforts much more closely to operational planning and the importance of funding for a comprehensive and rigorous testing program before 2010.

Table of Contents

Acronyms and Abbreviations	p. xix
Executive Summary	p. 1
Background and General Planning	p. 15
The Panel on Research on Future Census Methods	p. 17
Charge and Operations of the Panel	p. 17
Previous Reports of the Panel	p. 20
Overview of this Report	p. 20
The General Plan for the 2010 Census	p. 23
Basic Steps in the Decennial Census Process	p. 24
Preparation	p. 24
Basic Steps in Conducting the Decennial Census	p. 25
Organization of the Census Bureau	p. 26
Taking the Count	p. 28
Data Processing	p. 29
Research	p. 31
Planning and Conducting the 2000 Census	p. 32
The "Three-Legged Stool" Approach to the 2010 Census	p. 37
Relation of the 2010 Plan to Basic Census Processes	p. 39
Relation of the 2010 Plan to the 2000 Census	p. 40
Planning Milestones of the 2010 Census	p. 41
Status of the 2010 Census Plan	p. 42
Planned Testing and Development Cycle for the 2010 Decennial Census, Assuming a Short-Form-Only Census	p. 43
Reengineering The 2010 Census: A Process At Risk	p. 44
Census Bureau Listing of Perceived Risks in 2010 Census Planning	p. 47
Specific Risk Areas	p. 48
Mitigating the Risks	p. 53
Issues of Census Design	p. 55
Modernizing Geographic Resources	p. 57
Development and Current State of the MAF and TIGER	p. 59
The Master Address File	p. 59
Results of LUCA Working Group Study	p. 64
The TIGER Database	p. 66
The MAF/TIGER Enhancements Program	p. 71
Objective One: Address/Street Location Accuracy	p. 72
Objective Two: Modern Processing Environment	p. 75
Objective Three: Geographic Partnerships	p. 76
Objective Four: Community Address Updating System	p. 76
Objective Five: Evaluation and Quality Metrics	p. 77
Update on Enhancements Program Progress	p. 78
Assessment of Geographic Modernization Efforts	p. 79
Locational Accuracy of TIGER	p. 79
Balance of the MAF/TIGER Enhancements Program	p. 83
Weakness: Enhancing the MAF	p. 84
Current Plans for MAF Updates for 2010	p. 84
Block Canvassing	p. 87
Conclusions	p. 88
Recommendations	p. 90
Plan MAF Improvements Independent of MAF/TIGER Enhancements	p. 90
Coordinate Responsibility for the MAF	p. 92
Improve Research on the Delivery Sequence File	p. 93
Define the Role of the Community Address Updating System	p. 93
Plan Local Geographic Partnerships and Implement Early	p. 94
Justify the Complete Block Canvass	p. 97
Exploit 2000 MAF Data, and Redesign MAF for Evaluation in 2010	p. 99
American Community Survey	p. 103
Background and Current Plans	p. 105
Test Sites and the Census 2000 Supplementary Survey	p. 105
Current ACS Implementation Plans	p. 106
Assessing the ACS	p. 108
Estimation Using the ACS	p. 109
Adequacy of Moving Averages as Point Estimates	p. 109
Comparing ACS/C2SS to the Census Long Form	p. 111
Quality of ACS Estimates	p. 112
Estimating Nonresponse	p. 113
Imputation Rates for Selected Long-Form Items, 2000 Long-Form Sample and Census 2000 Supplemental Survey, by Type of Response, Household Population (weighted)	p. 116
Quality of Imputed Responses	p. 117
Measurement Error	p. 118
Summary and Assessment	p. 119
Benefits and Costs	p. 119
ACS Funding	p. 125
Contingency Planning	p. 127
Topics for Further Research and Design Consideration	p. 128
Group Quarters	p. 128
Voluntary versus Mandatory Response	p. 129
ACS as Both a Census Process and a Federal Survey	p. 130
Revisiting Sampling Strategies	p. 132
Interaction with Intercensal Population Estimates and Demographic Analysis Programs	p. 134
Research on General Estimation Issues	p. 136
Enumeration and Data-Processing Methods	p. 139
Portable Computing Devices	p. 140
Testing PCDs: Pretests and the 2004 Census Test	p. 142
Assessment	p. 144
Challenging Definitions for a Modern Census	p. 149
Housing Units	p. 149
Group Quarters	p. 150
Residence and Residence Rules	p. 153
Wording and Format of Race and Hispanic Origin Questions	p. 153
Hard-to-Count Population Groups: Extremes of Urbanicity	p. 156
Small Multiunit Structures and Immigrant Communities	p. 157
Rural Enumeration	p. 159
Alternative Response Modes and Contact Strategies	p. 160
Response Modes in 2000 and Early 2010 Testing	p. 161
Response Mode Effects	p. 163
Replacement Questionnaires	p. 164
Data-Processing Methodologies: Unduplication and Imputation	p. 165
Technical Infrastructure and Business Process	p. 169
Toward a "Business Process" of the Decennial Census	p. 172
Baseline: Logical Architecture of the 2000 Census	p. 173
Reengineering Exercise	p. 176
After the Pilot: Steps Toward an Architecture	p. 177
Assessment	p. 177
The Need for Institutional Commitment	p. 178
Management "Champions"	p. 179
Establishing a System Architect	p. 180
Cautionary Note: Breadth and Difficulty of Task	p. 181
The Architecture of Crucial Subsystems: The TIGER Redesign	p. 183
Challenges in Transition from Logical to Physical Infrastructure	p. 188
Potential Pitfall: Locking in Physical Infrastructure Too Early	p. 188
Enterprise Architecture as Learning Tool and Guide to Organizational Change	p. 189
Changing Architecture and Methods Simultaneously	p. 190
Improving Software Engineering and Development	p. 191
Coverage Measurement	p. 193
The Shape of Coverage Measurement in 2010	p. 194
Enhancing Demographic Analysis for 2010	p. 196
Enhancing Administrative Records Analysis for 2010	p. 199
Administrative Records Experiment in the 2000 Census	p. 199
Administrative Records for 2010	p. 202
Other Possibilities: Megalist and Reverse Record Check	p. 203
The Census Bureau's Research and Research Design	p. 205
Evaluations	p. 207
Strengthening the Evaluation Program of the 2010 Census	p. 208
Correcting the Disconnect Between Research and Operations	p. 210
Pursuing New Research Directions	p. 212
Exploiting Existing Data Resources	p. 214
Master Trace Sample	p. 215
Census Tests	p. 225
Constraints on Census Testing	p. 226
2003 National Census Test	p. 228
2004 Census Field Test	p. 230
The 2006 Census Test as a Proof of Concept	p. 231
Design Options that Should Be Considered for Testing in 2006	p. 233
Human Factors for Portable Computing Devices	p. 233
Various Cost-Benefit Trade-offs	p. 234
Other Testing Considerations	p. 236
Site Selection	p. 238
Recommendations	p. 239
Bibliography	p. 251
Biographical Sketches of Panel Members and Staff	p. 263
Table of Contents provided by Rittenhouse. All Rights Reserved.