ACS Sample Design Documentation

tp67.pdf

The American Community Survey

ACS Sample Design Documentation

OMB: 0607-0810

⚠️ Notice: This form may be outdated. More recent filings and information on OMB 0607-0810 can be found here:

Document [pdf]

Download: pdf | pdf

Design and Methodology
American Community Survey

Issued May 2006

TP67

Technical Paper 67 - UNEDITED VERSION

U.S. Department of Commerce
Economics and Statistics Administration
U.S. CENSUS BUREAU

UNEDITED VERSION

Design and Methodology
American Community Survey

U.S. Department of Commerce
Carlos M. Gutierrez,
Secretary
David A. Sampson,
Deputy Secretary
Economics and Statistics
Administration
Kathleen B. Cooper,
Under Secretary for Economic
Aﬀairs
U.S. CENSUS BUREAU
Charles Louis Kincannon,
Director

Issued May 2006

TP67

SUGGESTED CITATION
U.S. CENSUS BUREAU
Design and Methodology
American Community Survey
U.S. Government Printing Oﬃce,
Washington, DC,
2006.

Economics and Statistics
Administration
Kathleen B. Cooper,
Under Secretary
for Economic Aﬀairs

U.S. CENSUS BUREAU
Charles Louis Kincannon,
Director
Hermann Habermann,
Deputy Director and
Chief Operating Oﬃcer
Preston Jay Waite,
Associate Director
for Decennial Census
Teresa Angueira,
Assistant Director
for ACS and Decennial Census

ACKNOWLEDGMENTS
This technical paper was written under the direction of Lawrence S. McGinn, Chief, American
Community Survey Office; Lisa M. Blumerman, Deputy Chief, American Community Survey Office; and
Deborah H. Griffin, Special Assistant to the Assistant Director for American Community Survey and
Decennial Census. Management and coordination of this paper were under the supervision of Agnes S.
Kee, Chief, Project Coordination Staff, assisted by Brian D. Gregory and Marc S. Meyer.
The American Community Survey program is under the direction of Preston Jay Waite, Associate
Director for Decennial Census, and Teresa Angueira, Assistant Director for American Community
Survey and Decennial Census.
The report has been produced through the efforts of a number of individuals directly responsible for the
design and implementation of the American Community Survey, including Sharon K. Boyer, Maryann
M. Chapin, Thomas M. Coughlin, Barbara N. Diskin, Donald P. Fischer, Wendy Davis Hicks,
Douglas W. Hillmer, David L. Hubble, Susan P. Love, Alfredo Navarro, Joan B. Peacock, David
A. Raglin, Nicholas M. Spanos, and Lynn Weidman.
Other individuals who contributed to the review of this report include Mark E. Asiala, Aileen D.
Bennett, Claudette E. Bennett, Celia G. Boertlein, Scott Boggess, Fern J. Bradshaw, Kenneth R.
Bryson, Edwin R. Byerly, Patrick J. Cantwell, Cheryl V. Chambers, Thomas J. Chesnut,
Rosemarie C. Cowan, Arthur R. Cresce, Jr., Thu Q. Dang, Jennifer Cheeseman Day, Jorge H.
del Pinal, Scott P. Fifield, David E. Galdi, Nancy M. Gordon, Marjorie Hanson, Steven P. Hefter,
Adrien A. Hnat, Todd R. Hughes, Karen Humes, Timothy C. Jones, Debra L. U. Klein, Robert A.
Kominski, Robert A. LaMacchia, Gail A. Leithauser, John F. McKay, Louisa F. Miller, Karen M.
Mills, Brian Monaghan, Charles T. Nelson, Alan K. Peterson, Karen A. Piskurich, Sherry B.
Pollock, Joseph C. Powers, J. Gregory Robinson, Brian J. Scott, Linda M. Showalter, Carrie
Simon, Rajendra P. Singh, Denise I. Smith, Renee E. Spraggins, Michael D. Starsinic, John G.
Stiller, Arumugam Sutha, Brian Swanhart, Michael T. Thieme, Nestor Tolstikhin, Nancy K.
Torrieri, Jonah P. Turner, Janice Valdisera, Signe I. Wetrogran, Charles Whittington, Jeanne
Woodward, and Tommy Wright.
Catherine M. Raymond, Wanda K. Cevis, and Heather M. Lilley of the Administrative and Customer
Services Division, Walter C. Odom, Chief, provided publications and printing management, graphics
design and composition, and editorial review for print and electronic media. General direction and
production management were provided by James R. Clark, Assistant Division Chief, and Susan L.
Rappa, Chief, Publications Services Branch.

UNEDITED VERSION
Table of Contents
Chapter 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
Chapter 2. Program History
2.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
2.2 Stakeholders and Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-8
2.3 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
Chapter 3. Frame Development
3.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
3.2 Master Address File Basics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
3.3 Master Address File Development and Updating for Housing Units in the
United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
3.4 Master Address File Development and Updating for Housing Units in Puerto Rico . . . . 3-6
3.5 Master Address File Development and Updating for Special Places and Group Quarters in
the United States and Puerto Rico . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
3.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-8
Chapter 4. Design of the Sample and Sample Selection
4.1
4.2
4.3
4.4
4.5
4.6

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1
Housing Unit Sampling – Initial Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1
Housing Unit Sampling – Computer-Assisted Personal Interviewing Sub-Sample . . . . . 4-7
Group Quarters Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8
Remote Alaska Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-11
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-11

Chapter 5. Content Development Process
5.1
5.2
5.3
5.4
5.5
5.6

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
History of Content Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2003 – 2007 Content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Content Policy and Content Change Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2006 Content Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5-1
5-1
5-2
5-5
5-8
5-8

Chapter 6. Survey Rules, Concepts, and Definitions
6.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1

6.2
6.3
6.4
6.5
6.6

Interview Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
Residence Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
Structure of the Housing Unit Questionnaire . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3
Structure of the Group Quarters Questionnaires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12

Chapter 7. Data Collection and Capture for Housing Units
7.1
7.2
7.3
7.4
7.5

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1
Mail Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-4
Telephone Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-9
Personal Visit Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-11
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-13

Chapter 8. Data Collection and Capture for Group Quarters
8.1
8.2
8.3
8.4
8.5
8.6

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1
Group Quarters-Level Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2
Person-Level Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-5
Check In and Data Capture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8
Special Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-10
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-11

Chapter 9. Language Assistance Program
9.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9.2 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9.3 Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9.4 Mail Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9.5 Telephone and Personal Visit Follow-Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9.6 Group Quarters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9.7 Research and Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9.8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

9-1
9-2
9-2
9-2
9-3
9-4
9-4
9-5

Chapter 10. Data Preparation and Processing for Housing Units and Group Quarters
10.1
10.2
10.3
10.4

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Data Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Data Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

10-1
10-2
10-5
10-8

Chapter 11. Weighting and Estimation
11.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-1
11.2 2005 ACS Weighting – Probability of Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-2
11.3 2005 ACS Weighting – Non-Interview Adjustment . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-3

11.4
11.5
11.6
11.7

2005 ACS Weighting – Housing Unit and Population Controls . . . . . . . . . . . . . . . . . . 11-5
Weighting for Single-Year Estimates of Total Population . . . . . . . . . . . . . . . . . . . . . . 11-9
Multi-Year Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-9
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-10

Chapter 12. Variance Estimation
12.1
12.2
12.3
12.4
12.5
12.6
12.7

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ACS Housing Unit and Person Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Special Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2006 ACS and Beyond . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Generalized Variance Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Confidence Intervals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

12-1
12-1
12-2
12-3
12-3
12-4
12-6

Chapter 13. Preparation and Review of Data Products
13.1
13.2
13.3
13.4
13.5
13.6
13.7
13.8
13.9

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-1
Geography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-2
Defining the Data Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-5
Description of the Data Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-5
Public Use Microdata Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-8
Generation of Data Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-8
Data Review and Acceptance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-11
Custom Data Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-12
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-12

Chapter 14. Data Dissemination and User Assistance
14.1
14.2
14.3
14.4
14.5

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-1
Schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-1
Presentation of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-2
Education and Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-17
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-20

Chapter 15. Improving Data Quality by Reducing Non-Sampling Error
15.1
15.2
15.3
15.4
15.5
15.6

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Coverage Error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Non-Response Error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Measurement Error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Processing Error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

15-1
15-1
15-3
15-6
15-7
15-8

Figures
Figure 2-1
Figure 5-1
Figure 7-1
Figure 7-2
Figure 10-1
Figure 14-1
Figure 14-2
Figure 14-3
Figure 14-4
Figure 14-5
Figure 14-6
Figure 14-7
Figure 14-8
Figure 14-9
Figure 14-10
Figure 14-11
Figure 14-12
Figure 14-13
Figure 14-14
Figure 14-15
Figure 15-1
Figure 15-2
Figure 15-3

Test, C2SS, and 2005 Expansion Counties, American Community Survey,
1996 to Present . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-7
Example of Two ACS Questions Modified for the PRCS . . . . . . . . . . . . . . . . . 5-5
ACS Data Collection Consists of Three Overlapping Phases . . . . . . . . . . . . . . 7-2
Distribution of ACS Interviews and Non-Interviews . . . . . . . . . . . . . . . . . . . . 7-3
American Community Survey (ACS) Data Preparation and Processing . . . . . 10-2
Data Release Schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-2
The AFF Data Sets Page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-3
American FactFinder Base Table Presentation . . . . . . . . . . . . . . . . . . . . . . . . 14-4
American FactFinder Data Profile Presentation . . . . . . . . . . . . . . . . . . . . . . . 14-5
American FactFinder Narrative Profile Presentation . . . . . . . . . . . . . . . . . . . . 14-6
American FactFinder Ranking Chart Presentation . . . . . . . . . . . . . . . . . . . . . 14-7
American FactFinder Ranking Table Presentation . . . . . . . . . . . . . . . . . . . . . 14-8
American FactFinder Statistical Significance Presentation . . . . . . . . . . . . . . . 14-9
American FactFinder Thematic Maps Presentation . . . . . . . . . . . . . . . . . . . . 14-10
American FactFinder Subject Table Presentation . . . . . . . . . . . . . . . . . . . . . 14-11
Sample Select Population Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-12
AFF PUMS Main Page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-13
ACS Main Page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-14
Presentation of the ACS Quality Measures Main Page . . . . . . . . . . . . . . . . . 14-15
Main Page of 2004 FTP Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-17
ACS Quality Measures Web Site – Coverage Rates . . . . . . . . . . . . . . . . . . . . 15-3
ACS Quality Measures Web Site - Survey Response Rates . . . . . . . . . . . . . . 15-5
ACS Quality Measures Web Site - Item Allocation Rates . . . . . . . . . . . . . . . 15-6

Tables
Table 4-1
Table 4-2
Table 4-3
Table 5-1
Table 13-1

Sampling Rate Categories for the 2005 ACS . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-3
Initial and Final Overall Sampling Rates for the 2005 ACS . . . . . . . . . . . . . . . 4-5
CAPI Sub-Sampling Rates for the 2005 ACS . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8
ACS Topics Listed by Type of Characteristic and Question Number . . . . . . . . 5-4
Major Geographic Summary Levels To Be Published in Single-Year,
Three-Year, and Five-Year Data Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-4

Appendix A. Content Development
Appendix A.1 Changes to the American Community Survey (ACS) Questionnaires
(Housing Questions) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-1
Appendix A.2 Changes to the American Community Survey (ACS) Questionnaires
(Person Questions) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-18

Appendix B. Housing Unit
Appendix B.1
Appendix B.2
Appendix B.3
Appendix B.4
Appendix B.5
Appendix B.6
Appendix B.7
Appendix B.8
Appendix B.9
Appendix B.10
Appendix B.11
Appendix B.12
Appendix B.13
Appendix B.14
Appendix B.15
Appendix B.16

Pre-Notice Letter (U.S. Version) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1
Pre-Notice Letter (Puerto Rico Version) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-2
Outgoing Envelope (U.S. Version) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-3
Outgoing Envelope (Puerto Rico Version) . . . . . . . . . . . . . . . . . . . . . . . . . . . B-4
Initial Mail Package Cover Letter (U.S. Version) . . . . . . . . . . . . . . . . . . . . . . B-5
Initial Mail Package Cover Letter (Puerto Rico Version) . . . . . . . . . . . . . . . . B-6
ACS Questionnaire (U.S. Version) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-7
ACS Questionnaire (Puerto Rico Version) . . . . . . . . . . . . . . . . . . . . . . . . . . B-19
Guide to the ACS (U.S. Version) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-43
Guide to the ACS (Puerto Rico Version) . . . . . . . . . . . . . . . . . . . . . . . . . . . B-57
Frequently Asked Questions (FAQ) Brochure (U.S. Version) . . . . . . . . . . . B-71
Frequently Asked Questions (FAQ) Brochure (Puerto Rico Version) . . . . . B-73
Reminder Postcard (U.S. Version) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-75
Reminder Postcard (Puerto Rico Version) . . . . . . . . . . . . . . . . . . . . . . . . . . B-77
Replacement Mail Package Cover Letter (U.S. Version) . . . . . . . . . . . . . . . B-79
Replacement Mail Package Cover Letter (Puerto Rico Version) . . . . . . . . . B-80

Appendix C. Group Quarters
Appendix C.1
Appendix C.2
Appendix C.3
Appendix C.4
Appendix C.5
Appendix C.6
Appendix C.7
Appendix C.8
Appendix C.9
Appendix C.10
Appendix C.11
Appendix C.12
Appendix C.13
Appendix C.14

2006 American Community Survey Group Quarters Type Codes and
Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-1
American Community Survey Group Quarters Introductory Letters . . . . C-11
American Community Survey Group Quarters Frequently Asked Questions
Brochure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-23
Puerto Rico Community Survey Group Quarters Frequently Asked Questions
Brochure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-25
Remote Alaska American Community Survey Group Quarters Frequently
Asked Questions Brochure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-27
American Community Survey Group Quarters Letter for State and Local
Correctional Facilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-29
American Community Survey Group Quarters Field Representative’s
Information Card Booklet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-32
American Community Survey Group Quarters Listing Sheet . . . . . . . . . . . C-51
American Community Survey Group Quarters Sample Resident Introductory
Letter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-52
2006 American Community Survey Group Quarters Questionnaire . . . . . . C-53
2006 Puerto Rico Community Survey Group Quarters Questionnaire . . . . C-65
American Community Survey Group Quarters Instruction Guide . . . . . . . C-77
American Community Survey Group Quarters Question and
Answer Brochure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-102
American Community Survey and Puerto Rico Community Survey
Questionnaire Return Envelope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-108

Appendix D. Data Processing Recoded Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

D-1

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GLOSSARY-1

Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ACRONYMS-1

Chapter 1.
Introduction
The American Community Survey (ACS) is a new survey conducted by the U.S. Census Bureau.
This survey uses a series of monthly samples to produce annually updated data for the same
small areas (census tracts and block groups) as the decennial census long-form sample formerly
surveyed. Initially, five years of samples are required to produce these small-area data. Once
the Census Bureau has collected five years of data, new small-area data are produced annually.
The Census Bureau will also produce three-year and single-year data products for larger
geographic areas. Like the decennial census, the ACS will include people living in both housing
units (HUs) and group quarters (GQ) facilities.
This document describes the basic design of the ACS and details the full set of methods and
procedures that are currently in place. The ACS is conducted in the United States and in Puerto
Rico. In Puerto Rico the survey is called the Puerto Rico Community Survey (PRCS). For ease
of discussion, throughout this document the term ACS is used to represent both the survey that is
conducted in the United States and in Puerto Rico. The purpose of this paper is to provide data
users and other interested individuals with documentation of the final methods planned for the
ACS. Future updates of this paper will reflect revisions to existing methods, descriptions of new
methods, and details on multi-year estimation. Citations are provided that cover critical
background documentation for those with an interest in greater detail.
The paper is organized in a series of 15 chapters. Each chapter includes a brief overview
followed by detailed documentation and a list of references. A set of appendices provide
facsimiles of forms and letters. A glossary is included that defines terms and acronyms used
throughout the paper.
Chapter 2 summarizes the history and evolution of the ACS from the earliest origins to the
development of a survey prototype, national testing, and ultimately to its current implementation.
Chapters 3 and 4 cover topics associated with the ACS sample. Chapter 3 describes the survey
frame including the methods for updating that frame over time. Chapter 4 documents the ACS
sample design and includes an explanation of how the ACS samples are selected.
Chapters 5 and 6 detail the content covered by the ACS and define several basic concepts that
are critical to the ACS. Chapter 5 provides information on the content development process for
the ACS, focusing on current content and highlighting the ACS policy for changes to existing
content. Chapter 6 explains the interview and residence rules used in ACS data collection and
includes definitions of the concepts covered in the survey.
Chapters 7, 8, and 9 cover data collection and data capture methods and procedures. Chapter 7
focuses on the methods used to collect data from people who live in HUs while Chapter 8

1-1

focuses on methods used to collect data from people who live in GQ facilities. Chapter 9
discusses the ACS language assistance program, a critical support for data collection.
Chapters 10, 11, and 12 describe the data processing, weighting and estimation, and variance
estimation methods. Chapter 10 includes a discussion of data preparation activities (such as
coding) that are required to produce files for data processing activities such as editing and
imputation. Chapter 11 is a technical discussion of the process to produce survey weights.
Chapter 12 details the methods used to produce variance estimates.
Chapters 13 and 14 cover the definition, production, and dissemination of ACS data products.
Chapter 13 explains the process followed to produce, review, and release ACS data. Chapter 14
explains how these products can be accessed and provides examples of each type of data
product.
Chapter 15 documents the methods used in the ACS to control for non-sampling error. Also
included are examples of measures of quality that are produced annually to accompany each data
release.

1-2

Chapter 2.
Program History
2.1 OVERVIEW
Continuous measurement has long been viewed as a possible alternative method for collecting
detailed information on the characteristics of population and housing, but it was not considered a
practical alternative for the decennial census long form until 15 years ago. At that time, federal,
state, and local government, as well as private sector demands for current, nationally consistent
data led policymakers in government to consider the feasibility of collecting social and economic
data continuously throughout the decade, instead of only once every 10 years. The benefits of
current data, along with the anticipated benefits in cost savings, planning, improved census
coverage, and more efficient operations led the Census Bureau to plan to implement continuous
measurement in 2000, which was later renamed the American Community Survey (ACS). The
need to understand the nuances of sample design, survey methods, and data products under the
new program delayed implementation until after Census 2000. After additional testing, outreach
to stakeholders, and an ongoing process of interaction with key data users, especially those in the
statistical and demographic communities, the Census Bureau expanded the ACS to full sample
size for housing units (HUs) in 2005 and for group quarters (GQ) facilities in 2006.
The history of the ACS can be divided into four distinct stages. The design and early proposals
stage, 1990 to 1993, occurred when the concept of continuous measurement was first proposed.
The development stage, between 1994 and 1999, occurred when the Census Bureau tested early
prototypes of continuous measurement in a small number of sites. The demonstration stage,
between 2000 and 2004, occurred when the Census Bureau carried out large-scale, nationwide
surveys and produced reports for the country, states, and large geographic areas. The full
implementation stage began in January 2005 with an annual HU sample of about three million
addresses throughout the United States and about 36,000 addresses in Puerto Rico.
Design Origins and Early Proposals
In 1981, Leslie Kish introduced the concept of a rolling sample design in the context of the
decennial census (Kish, 1981). During the time that Kish was conducting his research, the
Census Bureau also recognized the need for more frequently updated data. In 1985, Congress
authorized a mid-decade census, but the funds were never appropriated. In the early 1990s,
Congress expressed renewed interest in an alternative to the once-a-decade census. The Census
Bureau began developing continuous measurement methods in the mid-1990s based on Kish’s
research.
The Census Bureau developed a research proposal for a continuous measurement alternative to
the collection of detailed sample data (Alexander, 1993g). Charles Alexander, Jr. developed

2-1

three prototypes for continuous measurement (Alexander, 1993i). Based on staff assessments of
operational and technical feasibility, policy issues, cost, and benefits (Alexander, 1994d), the
Census Bureau selected one prototype for further development.
Designers made several decisions in prototype development. They knew that in order to be cost
efficient, the survey would need to use a mailout mode of data collection. It was also decided
that the survey would use both telephone and personal visit follow-up modes. Additional
decisions were made with regard to the prototype’s geographic makeup, sampling rates, and the
use of population controls.
With the objective of producing five-year cumulations for small areas at the same level of
sampling reliability as the long form census sample, a monthly sample size of 500,000 HUs was
suggested (Alexander, 1993i), but this sample size drove costs out of an acceptable range. It was
determined that a monthly sample size of 250,000 would generate an acceptable level of
reliability when potential improvements in non-sampling error were considered. While great
progress had been made, there were still many outstanding issues to be resolved.
Development
Development moved to the next level with the establishment of a permanent Continuous
Measurement Staff in 1994. This staff continued the development of the survey prototype and
several design elements were identified as the foundation of this survey.
•
•
•

•
•

Data would be collected continuously throughout the year using independent monthly
samples.
Three modes of data collection would be used: mailout, telephone non-response followup, and personal visit non-response follow-up.
The survey reference date for establishing HU occupancy status and for many
characteristics would be the day the data were collected. Some data items would refer to
a longer reference period (for example, “last week,” “past 12 months”).
The survey’s estimates would be controlled to intercensal population and housing
estimates.
All estimates would be produced by aggregating data collected in the monthly surveys
over a period of time, such as the calendar year.

Documentation of early development took several forms. A group of 20 reports, known as the
Continuous Measurement Series (Alexander, C., 1994; Alexander, Charles H, 1992, 1993a –
1993i, 1994a – 1994e, and 1995a – 1995b; Alexander and Wetrogan, 1994; and Cresce, 1993),
were critical to the development of these design elements. These reports began in 1993 and
documented the research that led to the final prototype design. Plans for continuous
measurement were formally introduced to the statistical establishment at the American Statistical
Association’s (ASA’s) Joint Statistical Meetings in 1995. Love, Dalzell, and Alexander (1995)
outlined the assumptions that had to be met for such a survey to be successful, while Dawson,
Sebold, Love, and Weidman (1995) reported on some early feasibility studies of collecting

2-2

survey information by telephone. Possible modifications that were being considered and
evaluated with continuous measurement data were also discussed at that first ASA presentation
(Weidman, Alexander, Diffendal, and Love, 1995).
Operational testing of the ACS began in November 1995 in four test sites: Rockland County,
NY; Brevard County, FL; Multnomah County, OR; and Fulton County, PA. Testing was
expanded in November 1996 to include areas with a variety of geographic and demographic
characteristics. These areas included Harris County, TX; Fort Bend County, TX; Douglas
County, NE; Franklin County, OH; and Otero County, NM. This testing was undertaken to
validate methods and procedures, as well as to develop cost models for future implementation.
The tests resulted in revisions to the prototype design and identified additional areas for research.
Further research occurred in numerous areas, including: small area estimation (Chand and
Alexander, 1996), estimation methods (Alexander, Dahl, and Weidman, 1997), non-response
follow-up (Salvo and Lobo, 1997), weighting in ACS tests (Dahl, 1998), item non-response
(Tersine, 1998), response rates (Love and Diffendal, 1998), and the quality of rural data (Kalton,
Levine, Waksberg, and Helmick, 1998).
Operational testing continued and in 1998 three counties were added: Kershaw County, SC;
Richland County, SC; and Broward County, FL. The two counties in South Carolina were
included only in 1998 to produce data to compare with the 1998 Census Dress Rehearsal results.
Broward County was substituted for Brevard County. In 1999, testing expanded to 36 counties
in 26 states (U.S. Census Bureau, 2004e). The sites were not selected to be representative of the
country, but to represent different combinations of county population size, difficulty of
enumeration, and 1990-1995 population growth. The selection also incorporated geographic
diversity and sought to include areas representing different characteristics, such as racial and
ethnic diversity, seasonal populations, migrant workers, American Indian reservations, changing
economic conditions, and predominant occupation or industry types. Additionally, the Census
Bureau attempted to select sites with active data users who could participate in evaluating and
improving the ACS program. Based on the results of the operational tests, revisions were made
to the prototype and additional areas for research were identified.
In addition to testing methods for data collection and processing for HU populations, tests of
methods for the enumeration of people living in GQ facilities were held in 1999 and 2001.
These tests focused on the methodology for visiting GQ facilities, selecting resident samples,
and conducting resident interviews. The tests selected GQ facilities in all 36 test counties and
used procedures developed in the prototyping stage. Results of the tests led to modification of
sampling techniques and revisions to data collection methods.
While the main objective of the testing in the development phase was to determine the viability
of the methodologies utilized, usable data were generated. Data tables and profiles were
produced and released in 1999. These products provided data on demographic, social,
economic, and housing topics in both tabular and narrative format. Additionally, public use
microdata sample (PUMS) files were generated for a limited number of locations from 1996
through 1999. PUMS files show data for a sample of all of the survey’s HUs, with information

2-3

on the housing and population characteristics of each selected unit. All identifying information
is removed and other disclosure avoidance techniques are used to ensure confidentiality.
Demonstration
In 2000, a large scale demonstration was undertaken to assure Congress and other data users that
the ACS was capable of producing the demographic, social, economic, and housing data
previously attained from the census long-form sample.
The demonstration stage of the ACS was initially called the Census 2000 Supplementary Survey
(C2SS). The primary goal of the C2SS was to provide critical assessments of feasibility, quality,
and comparability with Census 2000, thus demonstrating the ability to fully implement the ACS.
Although ACS methods had been successful in the test sites, national implementation needed to
be demonstrated. Additional goals included refining procedures, improving the understanding of
the cost structure, improving cost projections, exploring data quality issues, and assuring users of
the reliability of ACS data and its ability to meet their needs.
The C2SS was conducted in 1,239 counties, of which 36 were ACS test counties and 1,203 were
new counties. It is important to note that only the 36 ACS test counties used the proposed ACS
sample design. The remaining counties used a primary sampling unit stratified design similar to
the Current Population Survey. The annual sample size increased from 165,000 HUs in 1999 to
800,000 HUs in 2000. The test sites remained in the sample throughout the C2SS and were
sampled at higher rates than the C2SS counties through 2001. This increased sample allowed
three-year period estimates from the ACS in these counties to be comparable to the planned fiveyear period estimates of a fully implemented ACS and to data from Census 2000.
Supplementary surveys were repeated from 2001 through 2004 using the same methodology.
A series of 11 reports was issued during the demonstration stage that analyzed various aspects of
the program. There were two general types of reports, methodology and data quality and
comparability. The methodology reports reviewed operational feasibility of the ACS. The data
quality and comparability reports compared C2SS data with the data from Census 2000,
including comparisons of three years of ACS test site data with Census 2000 data for the same
areas.
Report 1 (U.S. Census Bureau, 2001) found that the C2SS was operationally successful, planned
tasks were completed on time and within budget, and that the data collected met certain basic
Census Bureau quality standards. However, the report also noted that certain areas needed
improvement. Specifically, telephone questionnaire assistance (TQA) and failed-edit follow-up
(FEFU) operations were not sufficiently staffed to handle the large increase in workloads, due to
coinciding with the decennial census. This evaluation further noted that the ACS will improve
planning and simplify the 2010 decennial census design. The report notes, implementing the
ACS, supported by an accurate Master Address File (MAF) and Topologically Integrated
Geographic Encoding and Referencing (TIGER) database, holds promise to improve decennial
census coverage. Report 6 (U.S. Census Bureau, 2004c) was a follow-up evaluation on

2-4

feasibility utilizing data from 2001 and 2002. The evaluation concluded that the ACS was well
managed, achieving the desired response rates, and quality control procedures were functioning
as intended.
Report 2 (U.S. Census Bureau, 2002) concluded that the ACS would provide a reasonable
alternative to the decennial long form sample and added that the timeliness of the ACS data
gives it advantages over the long form. This evaluation did note that while ACS methodology
was sound, improving it needed to be an ongoing activity.
A series of reports compared national, state, and limited sub-state single-year period estimates
from the C2SS and Census 2000. Reports 4 and 10 (U.S. Census Bureau, 2004a and 2004g)
noted differences; however, the overall conclusion was that this research supported the proposal
to eliminate the decennial long form and move forward with plans for the ACS. Report 5 (U.S.
Census Bureau, 2004b) compared economic characteristics and concluded that estimates from
both methodologies were essentially the same. Report 9 (U.S. Census Bureau, 2004f) compared
social characteristics and noted that estimates from both methods were generally consistent, with
the exceptions of disability and ancestry. The report included the suggestion to complete further
research on these and other issues.
A set of multi-year period estimates (1999-2001) from the ACS test sites was created in order to
help demonstrate the usability, and reliability of ACS estimates at the county and tract
geographic levels. Results are found in Report 7 (U.S. Census Bureau, 2004d) and Report 8
(U.S. Census Bureau, 2004e). These comparisons with Census 2000 sample data helped further
confirm the comparability of the ACS and the Census 2000 long form estimates. Also identified
were additional areas requiring further research, such as variance reduction in sub-county
estimates.
In addition to the C2SS, a voluntary methods test was conducted during the demonstration phase,
at the request of Congress. The test was conducted between March and June of 2003 and was
designed to examine the impact, if any, that a change from mandatory to voluntary methods
would have on mail response, survey quality, and costs. Report 3 (U.S. Census Bureau, 2003b)
and Report 11 (U.S. Census Bureau, 2004h) examined the results of the test. The major impacts
of instituting voluntary methods included reductions in response rates across all three modes of
data collection (with the largest drop in traditionally low response areas); reductions in the
reliability of estimates; and increases in cost of more than $59 million annually.
As in the development phase, the demonstration phase yielded valuable demographic, social,
economic, and housing data. In addition to the data tables and PUMS files generated during the
development phase, ranking tables and charts were created to give a visual view of comparative
data on key characteristics across states, counties, and places. Detailed tables were produced to
provide data on all topics and geographic areas with populations of at least 250,000. Profiles
were made available comparing estimates between specific years for geographic regions with
more than one million in population.

2-5

Full Implementation
In 2003, with full implementation of the ACS approaching, the American Community Survey
Office (ACSO) came under the direction of the Associate Director for the Decennial Census.
While the Census Bureau’s original plan was to fully implement the ACS in 2003, budget
restrictions pushed back full HU implementation of the ACS in the United States and Puerto
Rico to January 2005. The GQ component of the ACS was fully implemented in January 2006.
Under full implementation, the ACS expanded from 1,240 counties in the C2SS and ACS test
sites, covering 37.1 percent of the geography of United States, to all 3,141 counties in all 50
states and the District of Columbia and to all 78 municipios in Puerto Rico, as shown in
Figure 2-1.
The annual ACS sample increased from approximately 800,000 addresses in the demonstration
phase to about three million addresses in full implementation. Workloads for all ACS operations
increased by more than 300 percent. Monthly mailout from the National Processing Center
(NPC) went from approximately 67,000 addresses per month to approximately 250,000
addresses per month. Telephone non-response follow-up workloads conducted from three
telephone call centers expanded from 25,000 calls per month to approximately 85,000. In
addition, about 3,500 field representatives (FRs) across the country conducted follow-up visits at
approximately 40,000 addresses each month, up from 1,200 FRs conducting follow-up visits at
11,000 addresses each month in 2004.
About 36,000 addresses in Puerto Rico are sampled every year. The same three modes of data
collection are used in Puerto Rico. Beginning in 2006, the ACS sampled 2.5 percent of the
population living in GQ facilities. This includes about 20,000 GQ facilities and approximately
195,000 people in GQ facilities in the United States and Puerto Rico.
With full implementation beginning in 2005, population and housing profiles for 2005 will first
become available in the summer of 2006 and every year thereafter for specific geographic areas
with populations of 65,000 or more. Three-year period estimates will be available in 2008 for
specific areas with populations of 20,000 or more, and five-year period estimates will be
available in 2010 for areas down to the smallest block groups, census tracts, small towns, and
rural areas. Beginning in 2010, and every year thereafter, the nation will have a five-year period
estimate available as an alternative to the decennial census long-form sample, a community
information resource that shows change over time, even for neighborhoods and rural areas. See
Chapters 13 and 14 for additional information on data products and their dissemination.

2-6

Figure 2-1. Test, C2SS, and 2005 Expansion Counties, American Community Survey, 1996 to Present

2-7

2.2 STAKEHOLDERS AND CONTRIBUTORS
Consultations with stakeholders began early in the ACS development process with the goal of
gaining feedback on the overall approach, as well as identifying potential pitfalls and obstacles.
Stakeholders include data users, federal agencies, and others with an interest in the survey. A
wide range of contacts was included in developmental activities, including federal, state and
local governments, advisory committees, professional organizations, and data users at many
levels. These groups provided their insights and expertise to the staff charged with developing
the ACS.
The Census Bureau established special-purpose advisory panels with the Committee on National
Statistics of the National Academies of Science (NAS) to identify issues of relevance in the
design of the survey. ACS staff undertook meetings, presentations, and other activities in
support of carrying out the ACS in American Indian and Alaska Native areas. These activities
included meeting with tribal officials and tribal liaisons, attendance at the National Conference
of American Indians, and continued interactions with the Advisory Committee for the American
Indian and Alaska Native Populations. A Rural Data Users Conference was held in May 1998 to
discuss issues of concern relating to small areas and populations. Numerous presentations were
made at annual meetings of the American Statistical Association (ASA) and other professional
associations.
Data users were also given opportunities to learn more about the ACS. A series of community
workshops were held during the development phase of the ACS. From March 1996 to
November 1999, 31 town hall-style meetings occurred throughout the country. More than 600
community members attended the meetings. A series of three regional outreach meetings, in
Dallas, TX; Grand Rapids, MI; and Seattle, WA, was held in mid-2004, with an overall
attendance of more than 200 individuals representing data users, academicians, the media, and
local governments.
Meetings with the Decennial Census Advisory Committee, the Census Advisory Committee of
Professional Associations, and the Race and Ethnic Advisory Committees provided opportunities
for ACS staff to discuss methods and to receive specific advice on methods and procedures to
improve the quality of the survey and the value of ACS data for all users. The Census Bureau’s
Field Division Partnership, Data Services Staff, and regional directors have played a prominent
role in communicating the message of the ACS to a variety of stakeholders. These groups
provide valuable input to the decision making process. Further, the ACS staff regularly briefs
several oversight groups including the Office of Management and Budget (OMB), the
Government Accountability Office (GAO), and the Inspector General of the Department of
Commerce (DOC). The Census Bureau has regularly briefed Congress on multiple aspects of
the ACS. Congressional briefings have been conducted with staff of the:
•
•

Committee on Government Reform, Subcommittee on Technology, Information Policy,
Intergovernmental Relations and the Census (House)
Committee on Governmental Affairs (Senate)

2-8

•
•
•
•
•
•
•

Committee on Appropriations (House)
Committee on Appropriations, Subcommittee on Justice, State, Commerce, and Related
Agencies (House)
Committee on the Budget (House)
Finance Committee (Senate)
Committee on Government Reform, Subcommittee on the Census (House)
Congressional Rural Caucus
Republican Conference (House)

These briefings began during the early stages of the ACS and continue to occur on a regular
basis.
In September 2000, the “Symposium on the ACS: Data Collectors and Disseminators” took
place. This symposium focused on the data uses and needs of the private sector. Public
comments were requested in a Federal Register Notice of May 14, 2004 regarding ACS data
products. As a result of this notice, comments were received, discussed, and ultimately led to
several changes. Continuing comments will be requested through the ACS Alert, a periodic
electronic newsletter. Changes based on stakeholder input were important in shaping the design
of the ACS and in the development of ACS. Stakeholder interactions continue to influence the
future form of the ACS, including questionnaire content and design.
Knowing that census long-form data were used by numerous government agencies, ACS staff
began a series of information meetings with the key federal agencies that are primary data users.
These meetings resulted in the formation of two independent groups. The Interagency
Committee for the ACS was formed in 2002 to discuss data collected by the ACS. The ACS
Federal Agency Information Program (FAIP), started in 2003, was developed to ensure that
federal agencies having a current or potential use for data from the ACS will have the assistance
they need in using the data. In addition, workshops are conducted by ACS staff members to
assist federal agencies in making the transition to ACS data.
Efforts are being made to share the Census Bureau’s experiences from the development and
implementation of the ACS internationally. Presentations have been made and working papers
have been shared with the United Nations Economic Commission for Europe (UNECE) working
sessions. Presentations have also been made at the International Seminar on Alternative
Methods for Housing and Population Censuses in Mexico and at a Workshop on Alternative
Approaches to Population and Housing Censuses in Brazil. A paper titled “Continuous
Measurement of Quality in the American Community Survey” was presented for the European
Conference on Quality and Methodology in Official Statistics conducted in Germany. A
presentation was also made to the International Statistical Institute’s Biennial Session in
Australia.

2-9

2.3 REFERENCES
Alexander, C. (1994), “An Idea for Using the Continuous Measurement (CM) Sample as the
CPS Frame,” Internal Census Bureau Reports CM-18, U.S. Census Bureau, Washington, DC,
1994.
Alexander, C., Dahl, S., and Weidman, L. (1997), “Making Estimates from the American
Community Survey,” Presented to the Annual Meeting of the American Statistical Association,
1997.
Alexander, Charles H. (1992), “An Initial Review of Possible Continuous Measurement
Designs,” Internal Census Bureau Reports CM-2, U.S. Census Bureau, Washington, DC, 1992.
Alexander, Charles H. (1993a), “A Continuous Measurement Alternative for the U.S. Census,”
Internal Census Bureau Reports CM-10, U.S. Census Bureau, Washington, DC, 1993.
Alexander, Charles H. (1993b), “Determination of Sample Size for the Intercensal Long Form
Survey Prototype,” Internal Census Bureau Reports CM-8, U.S. Census Bureau, Washington,
DC, 1993.
Alexander, Charles H. (1993c), “Including Current Household Surveys in a ‘Cumulated Rolling
Sample’ Design,” Internal Census Bureau Reports CM-5, U.S. Census Bureau, Washington, DC,
1993.
Alexander, Charles H. (1993d), “Overview of Continuous Measurement for the Technical
Committee,” Internal Census Bureau Reports CM-4, U.S. Census Bureau, Washington, DC,
1993.
Alexander, Charles H. (1993e), “Overview of Research on the ‘Continuous Measurement’
Alternative for the U.S. Census,” Internal Census Bureau Reports CM-11, U.S. Census Bureau,
Washington, DC, 1993.
Alexander, Charles H. (1993f), “Preliminary Conclusions About Content Needs for Continuous
Measurement,” Internal Census Bureau Reports CM-6, U.S. Census Bureau, Washington, DC,
1993.
Alexander, Charles H. (1993g), “Proposed Technical Research to Select a Continuous
Measurement Prototype,” Internal Census Bureau Reports CM-3, U.S. Census Bureau,
Washington, DC, 1993.
Alexander, Charles H. (1993h), “A Prototype Design for Continuous Measurement,” Internal
Census Bureau Reports CM-7, U.S. Census Bureau, Washington, DC, 1993.

2-10

Alexander, Charles H. (1993i), “Three General Prototypes for a Continuous Measurement
System,” Internal Census Bureau Reports CM-1, U.S. Census Bureau, Washington, DC, 1993.
Alexander, Charles H. (1994a), “Further Exploration of Issues Raised at the CNSTAT
Requirements Panel Meeting,” Internal Census Bureau Reports CM-13, U.S. Census Bureau,
Washington, DC, 1994.
Alexander, Charles H. (1994b), “Plans for Work on the Continuous Measurement Approach to
Collecting Census Content,” Internal Census Bureau Reports CM-16, U.S. Census Bureau,
Washington, DC, 1994.
Alexander, Charles H. (1994c), “Progress on the Continuous Measurement Prototype,” Internal
Census Bureau Reports CM-12, U.S. Census Bureau, Washington, DC, 1994.
Alexander, Charles H. (1994d), “A Prototype Continuous Measurement System for the U.S.
Census of Population and Housing,” Internal Census Bureau Reports CM-17, U.S. Census
Bureau, Washington, DC, 1994.
Alexander, Charles H. (1994e), “Research Tasks for the Continuous Measurement Development
Staff,” Internal Census Bureau Reports CM-15, U.S. Census Bureau, Washington, DC, 1994.
Alexander, Charles H. (1995a), “Continuous Measurement and the Statistical System,” Internal
Census Bureau Reports CM-20, U.S. Census Bureau, Washington, DC, 1995.
Alexander, Charles H. (1995b), “Some Ideas for Integrating the Continuous Measurement
System into the Nation’s System of Household Surveys,” Internal Census Bureau Reports CM19, U.S. Census Bureau, Washington, DC, 1995.
Alexander, Charles H., and Wetrogan, Signe I. (1994), “Small Area Estimation with Continuous
Measurement: What We Have and What We Want,” Internal Census Bureau Reports CM-14,
U.S. Census Bureau, Washington, DC, 1994.
American Community Survey Data Products, 69 Fed.Reg. 26806 (May 14, 2004).
Chambers, Cheryl V. (2005), “The American Community Survey - Communications Activities:
Past, Present and Future,” Paper for the record and in partial response to GAO request, U.S.
Census Bureau, Washington, DC, 2005.
Chand, N. and Alexander, C. (1996), “Small Area Estimation with Administrative Records and
Continuous Measurement,” Presented at the Annual Meeting of the American Statistical
Association, 1996.

2-11

Cresce, Art (1993), “‘Final’ Version of JAD Report and Data Tables from Content and Data
Quality Work Team,” Internal Census Bureau Reports CM-9, U.S. Census Bureau, Washington,
DC, 1993.
Dahl, S. (1998), “Weighting the 1996 and 1997 American Community Surveys,” Presented at
the American Community Survey Symposium, 1998.
Dawson, Kenneth, Sebold, Janice, Love, Susan, and Weidman, Lynn (1995), “Collecting Census
Long Form Data Over the Telephone: Operational Results of the 1995 CM CATI Test,”
Presented at 1996 Annual Meeting of the American Statistical Association, 1995.
Griffin, D. (2004), “Continuous Measurement of Quality in the American Community Survey,”
Presented at the European Conference on Quality and Methodology in Official Statistics, 2004.
Griffin, D. (2005), “Making the Transition from a Decennial Census Long Form to the American
Community Survey,” Presented at the 2005 ISI Meetings, Sydney, Australia, 2005.
Griffin, Deborah H. (2004), “The American Community Survey: Developing a Continuous
Measurement Application in the United States,” Powerpoint presentation presented at Workshop
on Alternative Approaches to Population and Housing Censuses, Instituto Brasileiro de
Geografia e Estatistica, October 2004.
Herriot, Roger A., Bateman, David V., and McCarthy, William F. (1989), “The Decade Census
Program - A New Approach for Meeting the Nation’s Needs for Sub-National Data,” American
Statistical Association, Proceedings of the Section on Social Statistics, 1989.
Kalton, G., Levine, D., Waksberg, J., and Helmick, J. (1998), “The American Community
Survey: The Quality of Rural Data, Report on a Conference,” Prepared by Westat, June 29,
1998.
Kish, Leslie (1981), “Using Cumulated Rolling Samples to Integrate Census and Survey
Operations of the Census Bureau: An Analysis, Review, and Response,” Washington, DC, U.S.
Government Printing Office, 1981.
Love, S. and Diffendal, G. (1998), “The 1996 American Community Survey Monthly Response
Rates, by Mode,” Presented to the American Community Survey Symposium, 1998.
Love, Susan, Dalzell, Donald, and Alexander, Charles (1995), “Constructing a Major Survey Operational Plans and Issues for Continuous Measurement,” Presented at 1995 Annual Meeting
of American Statistical Association, 1995.
Salvo, J. and Lobo, J. (1997), “The American Community Survey: Non-Response Follow-Up in
the Rockland County Test Site,” Presented to the Annual Meeting of the American Statistical
Association, 1997.

2-12

Tersine, A. (1998), “Item Non-Response: 1996 American Community Survey,” Presented to the
American Community Survey Symposium, 1998.
U.S. Census Bureau (1992), “2000 Census Research and Development Alternative Design
Program,” Internal Census Bureau report of the Year 2000 Research and Development Staff,
Washington, DC, 1992.
U.S. Census Bureau (2001), “Meeting 21st Century Demographic Data Needs--Implementing the
American Community Survey: July 2001, Report 1: Demonstrating Operational Feasibility,”
Washington, DC, July 2001.
U.S. Census Bureau (2002), “Meeting 21st Century Demographic Data Needs--Implementing the
American Community Survey: May 2002, Report 2: Demonstrating Survey Quality,”
Washington, DC, May 2002.
U.S. Census Bureau (2003a), “American Community Survey Operations Plan,” Washington,
DC, March 2003.
U.S. Census Bureau (2003b), “Meeting 21st Century Demographic Data Needs--Implementing
the American Community Survey: Report 3: Testing the Use of Voluntary Methods,”
Washington, DC, December 2003.
U.S. Census Bureau (2004a), “Meeting 21st Century Demographic Data Needs--Implementing
the American Community Survey: Report 4: Comparing General Demographic and Housing
Characteristics With Census 2000,” Washington, DC, May 2004.
U.S. Census Bureau (2004b), “Meeting 21st Century Demographic Data Needs--Implementing
the American Community Survey: Report 5: Comparing Economic Characteristics With Census
2000,” Washington, DC, May 2004.
U.S. Census Bureau (2004c), “Meeting 21st Century Demographic Data Needs--Implementing
the American Community Survey: Report 6: The 2001-2002 Operational Feasibility Report of
the American Community Survey,” Washington, DC, May 2004.
U.S. Census Bureau (2004d), “Meeting 21st Century Demographic Data Needs--Implementing
the American Community Survey: Report 7: Comparing Quality Measures: Comparing the
American Community Survey's Three-year Averages and Census 2000's Long Form Sample
Estimates,” Washington, DC, June 2004.
U.S. Census Bureau (2004e), “Meeting 21st Century Demographic Data Needs--Implementing
the American Community Survey: Report 8: Comparison of the ACS 3-year Average and the
Census 2000 Sample for a Sample of Counties and Tracts,” Washington, DC, June 2004.

2-13

U.S. Census Bureau (2004f), “Meeting 21st Century Demographic Data Needs--Implementing
the American Community Survey: Report 9: Comparing Social Characteristics with Census
2000,” Washington, DC, June 2004.
U.S. Census Bureau (2004g), “Meeting 21st Century Demographic Data Needs--Implementing
the American Community Survey: Report 10: Comparing Selected Physical and Financial
Housing Characteristics with Census 2000,” Washington, DC, July 2004.
U.S. Census Bureau (2004h), “Meeting 21st Century Demographic Data Needs--Implementing
the American Community Survey: Report 11: Testing Voluntary Methods - Additional Results,”
Washington, DC, December 2004.
Weidman, L., Alexander, C., Diffendal, G., and Love, S. (1995), “Estimation Issues for the
Continuous Measurement Survey,” Presented at the Annual Meeting of the American Statistical
Association, 1995.

2-14

Chapter 3.
Frame Development
3.1 OVERVIEW
The U.S. Census Bureau maintains a national Master Address File (MAF) that is used as the
source of addresses for the American Community Survey (ACS), other Census Bureau
demographic surveys’ sampling frames, and decennial censuses. The MAF is the Census
Bureau’s official inventory of known living quarters (housing units (HUs) and group quarters
(GQ) facilities) and selected non-residential units (public, private, and commercial) in the United
States and Puerto Rico. The MAF contains mailing and location address information, geocodes,
and other attribute information about each living quarters.
The MAF is linked to the Topological Integrated Geographic Encoding and Referencing
(TIGER) files. TIGER is a database with a digital representation of all census-required map
features and related attributes. TIGER provides a resource for the production of maps, data
tabulation, and the automated assignment of addresses to geographic locations in a process
known as geocoding.
For Census 2000, the initial MAF was created using multiple sources that included the 1990
Address Control File, the U.S. Postal Service’s (USPS’s) delivery sequence file (DSF), field
listing operations, and addresses supplied by local governments through partnership operations.
The MAF, as it existed at the conclusion of Census 2000, is used as the initial base frame for the
ACS. The Census Bureau continues to update the MAF using the DSF and various automated,
clerical, and field operations such as the Community Address Updating System (CAUS).
This chapter contains four sections that provide detailed information on the development of lists
from which the ACS sample units are selected. The first section provides basic information
about the MAF and its development. The next two sections describe the MAF development and
update activities for HUs in the United States and Puerto Rico. The fourth section describes the
MAF development and update activities for ACS GQ facilities.

3.2 MASTER ADDRESS FILE BASICS
The MAF is the Census Bureau’s official inventory of known HUs in the United States and
Puerto Rico. Each HU is represented by a separate MAF record that contains various
information about the mailing address, the location address, the description of the unit, and
various geographic codes. The geographic codes contained in the MAF identify a variety of
areas including states, counties, county subdivisions, places, American Indian Reservations,
Alaska Native Villages, Hawaiian Homelands, census tracts, block groups, and blocks. Two
important geographic code sets within the MAF are the Census 2000 tabulation geography set,
which is based on the January 1, 2000 legal boundaries, and the current geography set, which is
3-1

based on the January 1 legal boundaries of the most recent year. For example, MAF extracts
received in July 2006 reflect the legal boundaries as of January 1, 2006. The geographic codes
associated with each MAF record are assigned by the TIGER database. Because each MAF
record contains these various geographic codes, it is possible to sort the MAF in different
geographic hierarchies. Sorting by state, county, tract, and block is generally used for ACS
operations. Each MAF record also contains attribute information including some or all of the
following: current unit status (a code indicating whether an address is valid, demolished,
nonexistent, under construction, and so on), residential status (a code identifying each MAF
record as either residential or nonresidential), latitude and longitude coordinates, and information
indicating the operation(s) that added/updated the record.
The MAF contains both city-style and non-city-style mailing addresses. A city-style address is
one that uses a structure number and street name format; for example, 201 Main Street,
Anytown, MA 01977. A non-city-style mailing address is one that uses either a rural route and
box number format, a post office (PO) box format, or a general delivery format. Examples of
these types of addresses are RR 2, Box 13, Anytown, MA 01977; PO Box 123, Anytown, MA
01977; and T. Smith, General Delivery, Anytown, MA 01977.
In the United States, the city-style addresses are most prevalent in urban and suburban areas and
at the end of Census 2000 accounted for 94.4 percent of all residential addresses in the MAF.
Most city-style addresses represent both the mailing address and the location address of the unit.
In some situations, a city-style address may be used only as a location address, such as addresses
used for emergency-response assistance. For example, the location address could be 77 West
Street, but the mailing address is PO Box 123.
The non-city-style mailing addresses are prevalent in rural areas of the United States and
represent approximately 2.5 percent of all residential addresses in the MAF. The non-city-style
address provides a mailing address but it does not provide specific information about the
location of the unit. Locating a rural route and box number address in the field can be difficult
because the mail box is not always adjacent to the unit. Locating the unit associated with a PO
box address or a general delivery address is very difficult. In order to easily locate non-city-style
addresses in the field, the MAF often contains a location description of the unit and its latitude
and longitude coordinates. The presence of this information in the MAF makes field follow-up
operations possible.
The MAF also contains some address records that are classified as incomplete because they lack
a complete city-style or a complete non-city-style address. Records in this category are often
just a description of the unit’s location and usually its latitude and longitude. This incomplete
category accounts for the remaining 3.1 percent of the United States addresses in the MAF.
The MAF, as an inventory of living quarters and some non-residential units, is a dynamic entity.
It contains millions of addresses that reflect ongoing additions, deletions, and changes. The
MAF contains current addresses, as well as addresses determined to no longer exist. Users of
the MAF, such as the ACS, define the set of usable addresses for their program. These addresses
are provided in what are known as MAF extracts. The ACS MAF extracts are generated based
3-2

on requirements that define the frame from which the sample is selected. In general, the ACS
MAF extracts contain several classes of units including: HUs that existed during Census 2000,
post census adds from the DSF, adds from the demographic area address listing (DAAL), count
question resolution (CQR) adds and reinstatements, and Census 2000 deletes that persist on the
DSF. As the ACS learns about its sample addresses and learns more about the coverage and
content of the MAF, the filter is enhanced. Filtering rules change and as such, the ACS frame
changes. For details on the ACS requirements for MAF extracts, see Dean (2005a). For more
information on the ACS sample selection, see Chapter 4.
For details on the MAF, including its content and structure, see U.S. Census Bureau (2000b).
For a description of data collection for these different kinds of addresses, see Chapter 7. For
details on the MAF, its coverage, and the implications of extract rules on the ACS frame, see
Shapiro and Waksberg (1999) and Johns (2005).

3.3 MASTER ADDRESS FILE DEVELOPMENT AND UPDATING FOR HOUSING
UNITS IN THE UNITED STATES
MAF Development
For the 1990 Decennial Census and earlier censuses, address lists were compiled from several
sources (commercial vendors, field listings, and others) and assembled. Before 1990, the
address lists were not maintained or updated after the Census was completed. Following the
1990 Census, the Census Bureau decided to develop and maintain a master address list in order
to support both the decennial census and other Census Bureau survey programs and to avoid the
need to rebuild the address list just prior to each census.
The MAF was created by merging city-style addresses from the 1990 address control file, field
listing operations, the USPS’s DSF, and addresses supplied by local governments through
partnership operations like the Local Update of Census Addresses (LUCA) and other Census
2000 activities such as the Be Counted Campaign. For areas that contained predominantly noncity-style addresses, the Census Bureau obtained information from various Census 2000 field
operations (i.e., address listing, list/enumerate, and non-response follow-up), map update
operations, and feedback from states and local governments. At the conclusion of Census 2000,
the MAF contained a complete inventory of known HUs nationwide. For details on the address
list development for Census 2000, see U.S. Census Bureau (2000a).
MAF Improvement Activities and Operations
MAF maintenance is an ongoing, complex task. New HUs are constantly being built, older units
are demolished, and the institution of new emergency-response addressing schemes render many
addresses, that have been used for decades, obsolete. Maintenance of the MAF occurs through a
coordinated combination of automated, clerical, and field operations designed to improve
existing MAF records and keep up with the nation’s changing housing stock and associated
addresses. With the completion of Census 2000, the Census Bureau implemented several short3-3

term, one-time operations to improve the quality of the MAF. These operations included CQR,
automated address unduplication, MAF/TIGER reconciliation, and address corrections from
rural directories. For the most part, these operations were implemented to improve the addresses
recognized in Census 2000 and their associated characteristics. Many other operations to
improve the MAF are ongoing. Some are designed to deal with errors remaining from Census
2000, while others are to keep pace with post-Census 2000 address development. In the
remainder of this section, we will discuss several ongoing operations including DSF updates,
MAF geocoding office resolution (MAFGOR), ACS non-response follow-up updates, and
DAAL updates, of which the CAUS is a part.
Delivery Sequence File. The DSF is the USPS’s master list of all delivery-point addresses
served by postal carriers. The file contains specific data coded for each record, a standardized
address and ZIP code, and codes that indicate how the address is served by mail delivery (for
example, carrier route and the sequential order in which the address is serviced on that route).
The DSF record for a particular address also includes a code for delivery type that indicates
whether the address is business or residential. After Census 2000, the DSF has been the primary
source of new city-style MAF addresses. For areas in which a significant number of mail
deliveries are accomplished using non-city-style addresses, DSF addresses are not used to update
the MAF because those addresses lack geocodes and might provide different (and unmatchable)
address representations for HUs whose addresses already exist on the MAF. New versions of the
DSF are shared with the Census Bureau twice a year, and updates or “refreshes” to the MAF are
made at those times. For additional details on the MAF update process via the DSF, see Hilts
(2005).
MAFGOR. MAFGOR is an ongoing clerical operation in all Census Bureau regional offices.
In this operation, geographic clerks examine address clusters representing addresses that do not
geocode to the TIGER database. Reference materials available commercially, from local
governments, and on the Internet are used to add or correct features, feature names, and/or
address ranges in the TIGER database. This process increases the Census Bureau’s ability to
assign block geocodes to DSF addresses.
Address Updates from ACS Non-Response Follow-Up. Field Representatives (FRs) can
obtain address corrections for each HU visited during the personal visit non-response follow-up
phase for the ACS. The MAF is updated to reflect these corrections. For additional details on
the MAF update process for ACS address corrections collected at time of interview, see Riley
(2005).
Demographic Area Address Listing. DAAL is a combination of operations, systems, and
procedures associated with coverage improvement operations, address list development, and
automated listing for the CAUS and the demographic household surveys. The objectives of the
DAAL are to: update the inventory of HUs and feature information for specified areas of the
country to improve the completeness and coverage of the MAF in preparation for the ACS
sample selection (see the section on the CAUS, below), and to create and update the inventory of
HUs and GQ facilities in identified areas for surveys such as the Current Population Survey

3-4

(CPS), the National Health Interview Survey (NHIS), and the National Crime Victimization
Survey (NCVS).
In a listing operation, such as DAAL, a defined land area is traveled in a systematic manner as
an FR records the location and address of every structure where a person lives or could live.
Listings for DAAL are conducted using software known as the automated listing and mapping
instrument (ALMI), on laptop computers. The ALMI uses extracts from the current MAF and
TIGER databases as inputs. Functionality in the ALMI allows users to edit, add, delete, and
verify addresses; edit, add, and delete streets and other map features; view a list of addresses
associated with a selected level of geography; and view and denote the location of HUs on the
electronic map. For information once collected by paper and pencil, the ALMI allows
standardization of the data collected through edits and defined data entry fields, standardization
of field procedures, efficiencies in data transfer, and timely reflection of the address and feature
updates in MAF and TIGER. For details on DAAL, see Perrone (2005).
Community Address Updating System. As a program designed specifically to address
coverage concerns for the ACS, the CAUS program is described below in additional detail. The
Census Bureau recognized that the DSF, as the primary source of ACS frame updates, was not
adequate to deal with changes in predominantly rural areas of the nation where city-style
addresses are generally not used for mail delivery. The CAUS program, an automated field data
collection operation, is designed to provide a rural counterpart to the update of city-style
addresses received from the DSF. Without the CAUS program, coverage of the ACS would be
negatively impacted by addresses that exist but do not appear on the DSF; non-city-style
addresses on the DSF that do not appear on the MAF; and addresses on the DSF that also appear
on the MAF but are erroneously excluded from the ACS frame or are erroneously included in the
ACS frame.
Implemented in September 2003, the CAUS program focuses its efforts on census blocks with
high concentrations of non-city-style addresses and suspected growth in the HU inventory. Of
the approximately 8.2 million blocks nationwide, the CAUS universe is defined as the
approximately 750,000 blocks where DSF updates are not used to provide adequate coverage.
CAUS blocks are selected using a model-based selection method that transforms information
gained from previous field data collection efforts and administrative records into knowledge for
predicting where the CAUS work is needed. The blocks are ranked according to the model and
the number of selected blocks is subject to budget and operational considerations of the program.
Through DAAL operations, the CAUS captures address information for new HUs not currently
found in the MAF, as well as updates for existing MAF HUs. Past results indicate that the
model-based selection is more effective than a random selection of blocks. For details on the
CAUS program and its block selection methodology, see Dean (2005b).
All of these MAF improvement activities and operations (DSF, MAFGOR, and so on) contribute
to the overall update of the MAF. In general, quality assurance and quality control checks are
implemented for operations that update the MAF. Continual evaluation of the MAF and its
updating are planned and will be cited in future releases of this report.

3-5

Planning for the 2010 Census is underway. Field operations to support the 2010 Census will be
implemented to enable HU and GQ updates, additions, and deletions to be identified, collected,
and used to update the MAF. The Census Bureau will begin its 2010 Census operations in 2008.
The operations will include several nationwide field canvassing and enumeration operations, and
will obtain address data through a cooperative effort with tribal, county, and local governments
to enhance the MAF. The MAF extracts used by the ACS for sample selection will be improved
by these operations. ACS and Census 2010 planners are working closely to assess the impact of
the decennial operations on the ACS.

3.4 MASTER ADDRESS FILE DEVELOPMENT AND UPDATING FOR HOUSING
UNITS IN PUERTO RICO
The Census Bureau created an initial MAF for Puerto Rico through field listing operations and
other Census 2000 activities. This initial MAF did not include mailing addresses because
Census 2000 used an update/leave methodology through which a census questionnaire was
delivered by an enumerator to each living quarters. The MAF update activities that took place in
2002 through 2004 were focused on developing mailing addresses, updating the address
information, and improving the coverage through yearly updates. These improvements will
benefit both the ACS in Puerto Rico and the 2010 Census.
Puerto Rico MAF Development
As with the MAF development in the United States, MAF development in Puerto Rico used the
Census 2000 operations as its foundation. The Census 2000 operations in Puerto Rico included
address listing, update/leave, the LUCA, and the Be Counted Campaign. For details on the
Census 2000 for Puerto Rico, see U.S. Census Bureau (2004b)
The Census 2000 procedures and processing systems were designed to capture, process, transfer,
and store information for the conventional three-line mailing address. The mailing addresses in
Puerto Rico generally incorporate the urbanization name (neighborhood equivalent) in the
address, which creates a four-line address. Use of the urbanization name eliminates a problem
created when street names are repeated in adjacent communities. In some instances, the
urbanization name is used in lieu of the street name.
The differences between the standard three-line address format and the four-line address format
used in Puerto Rico created problems during the early MAF building stages. The resulting file
structure for the Puerto Rico MAF was the same used in other states, and it did not contain the
additional fields required to handle the more complex Puerto Rico mailing address. These
processing problems did not adversely impact Census 2000 operations because the methodology
used to conduct the census was update/leave, and this procedure is designed for an enumerator to
deliver a census questionnaire instead of mailing it. However, in the ACS, where mailout is
planned as the primary means of collecting data, the three-line address format did create a
problem. Due to these limitations, it is not possible to calculate the percentage of city-style
addresses, non-city-style mailing addresses, and incomplete addresses in Puerto Rico.
3-6

Puerto Rico MAF Improvement Activities and Operations
The MAF for Puerto Rico as it existed at the conclusion of Census 2000 required significant
work before it could be used by the ACS. As a result, the Census Bureau had to revise the
address information in the Puerto Rico MAF. This effort included the creation of a new address
record layout and the splitting of address information into the various fields required to construct
a mailing address using Puerto Rico addressing conventions.
The Census Bureau also contracted for the list of addresses on the Puerto Rico MAF to be
updated. Approximately 64,000 new HUs were added to the MAF since Census 2000 with each
address geocoded to a municipio, tract, and block. The Census Bureau also worked with the
USPS customer files for Puerto Rico and attempted to extract information on new HUs from the
USPS DSF. Matching the USPS file to the existing MAF was only partially successful because
of inconsistent naming conventions, missing information in the MAF, and the existence of
different housing numbering schemes (USPS versus local schemes).
The data collection activities in Puerto Rico began in November 2004. FRs in Puerto Rico are
now capturing address corrections for each HU visited during the personal visit non-response
follow-up phase of the ACS. These corrections will be incorporated into the current Puerto Rico
MAF. In addition, the Census Bureau is pursuing options for the ongoing collection of address
updates in Puerto Rico. This may include operations comparable to those that exist in the United
States, such as DSF updates, MAFGOR, and the CAUS. Future versions of this document will
include decisions relating to these operations and MAF development and updating in Puerto
Rico.

3.5 MASTER ADDRESS FILE DEVELOPMENT AND UPDATING FOR SPECIAL
PLACES AND GROUP QUARTERS IN THE UNITED STATES AND PUERTO RICO
Group Quarters MAF Development
In preparation for Census 2000, the Census Bureau developed an inventory of special places
(SPs) and GQ facilities. SPs are places such as prisons, hotels, migrant farm camps, and
universities. GQ facilities are contained within SPs and include college and university
dormitories and hospital/prison wards. The SP/GQ inventory was developed using data from
internal and external administrative lists and numerous Census 2000 operations, such as address
listing, block canvassing, and the SP/GQ facility questionnaire operation used to collect/update
address and location information for SPs and associated GQ facilities. Responses to the
questionnaire identified GQ facilities and any HUs associated with the SP. Similar to the HU
MAF development process, local and tribal governments had an opportunity to review the SP
address list and add any SPs that were missing. In August 2000, after the enumeration of GQ
facilities, address and identification information for GQ facilities was incorporated into the
MAF.

3-7

Group Quarters MAF Improvement Activities and Operations
As with the HU side of the MAF, maintenance of the GQ universe is an ongoing and complex
task. Listed in the earlier section on MAF Improvement Activities and Operations for HUs are
short term, one-time operations that also update GQ information. Additionally, the Census
Bureau completed a GQ geocoding correction operation that corrected errors (mostly census
block geocodes) associated with college dormitories in the MAF and TIGER.
On an ongoing basis, information on new GQ facilities and updated address information for
existing GQ facilities are collected by listing operations, such as DAAL which includes the
CAUS in rural areas. This information is used to update the MAF. Additionally, it is likely that
DSF updates in city-style address areas are providing the Census Bureau with new GQ
addresses; however, the DSF does not yet identify such an address as a GQ facilities.
A process to supplement the activities above was developed to create an updated GQ universe by
which to select the ACS sample. The ACS GQ universe for 2006 was constructed from merging
the updated SP/GQ inventory file, extracts from the MAF, and a file of GQ facilities that were
closed on April 1, 2000 (but which may be open if visited at another time of year). To
supplement the ACS GQ universe, the Census Bureau obtained a file of federal prisons and
detention centers, and a file from the Department of Defense (DOD) containing military bases
and vessels, and conducted Internet-based research to identify new state prisons and state prisons
which had closed.
Additionally, ACS FRs may collect updated address and geographic location information. The
ACS will use the updates collected to provide more accurate information for subsequent visits to
a facility, as well as to update the ACS GQ universe for subsequent years.
As mentioned in the MAF Improvement Activities and Operations section on HUs, planning for
Census 2010 is underway. In addition to the major decennial operations that will collect and
provide updates for GQ facilities, ACS and Census 2010 planners are evaluating the feasibility
of a repeatable operation to extract information on new GQ facilities from administrative sources
including data provided by members of the Federal-State Cooperative Program for Population
Estimates. This approach, if deemed successful, will likely provide a cost-effective mechanism
for updating the GQ universe for the ACS during the intercensal years.

3.6 REFERENCES
Bates, Lawrence M. (2005), “Creating the Group Quarters Universe for the American
Community Survey For Sample Year 2006," Internal U.S. Census Bureau Memorandum from R.
Singh to L. McGinn, Draft, Washington, DC, August 26, 2005.
Dean, Jared (2005a), “Geographic Products Requirements for the American Community Survey
REVISED for July 2005 Delivery,” Internal U.S. Census Bureau Memorandum from R.Singh
and L. McGinn to R. LaMacchia, Draft, Washington, DC, May 25, 2005.
3-8

Dean, Jared (2005b), “Updating the Master Address File: Analysis of Adding Addresses via the
Community Address Updating System,” Washington, DC, 2005.
Hilts, Jeremy (2005), “Software Requirement Specification for Updating the Master Address
File from the U.S. Postal Service’s Delivery Sequence File,” Version 7.0, Washington, DC,
April 18, 2005.
Johns, Christopher (2005), “Final Report for National Estimate of Coverage of the MAF for
2004,” Internal U.S. Census Bureau Memorandum from R. Killion to R. LaMacchia,
Washington, DC, September 29, 2005.
National Research Council (2003), “Planning the 2010 Census: Second Interim Report,” Panel
on Research on Future Census Methods, Daniel L. Cork, Michael L. Cohen, and Benjamin F.
King, eds., Committee on National Statistics, Washington, DC, 2003.
Perrone, Susan (2005), “Final Report for the Assessment of the Demographic Area Address
Listing (DAAL) Program,” Internal U.S. Census Bureau Memorandum from R. Killion to R.
LaMacchia, Washington, DC, November 9, 2005.
Riley, Paul L. (2005), “Software Requirements Specification for MAF Update from the ACSTime of Interview Operation,” Version 1.0, Washington, DC, April 25, 2005.
Shapiro, Gary and Waksberg, Joseph (1999), “Coverage Analysis for the American Community
Survey Memo,” Final Report Submitted by Westat to the U.S. Census Bureau, Washington, DC,
November 1999.
U.S. Census Bureau (2000a), “Census 2000 Operational Plan,” Washington, DC, 2000.
U.S. Census Bureau (2000b), “MAF Basics,” Washington, DC, 2000.
U.S. Census Bureau (2003), “American Community Survey Operations Plan: Release 1: March
2003," Washington, DC, 2003.
U.S. Census Bureau (2004a), “Census 2000 Topic Report No. 8: Address List Development in
Census 2000,” Washington, DC, 2004
U.S. Census Bureau (2004b), “Census 2000 Topic Report No. 14: Puerto Rico,” Washington,
DC, 2004.

3-9

Chapter 4.
Design of the Sample and Sample Selection
4.1 OVERVIEW
The sample for the ACS is selected in each of the 3,141 counties and county equivalents in the
United States, including the District of Columbia, and each of the 78 municipios in Puerto Rico.
Samples of housing unit (HU) addresses and group quarters (GQ) facilities are selected
separately. The first full-implementation sample of HU addresses was selected for use in 2005.
Each year the ACS samples about three million HU addresses in the United States and about
36,000 HU addresses in Puerto Rico. The first full-implementation sample of GQ facilities was
selected for use in 2006 and approximately 2.5 percent of the people in GQ facilities are included
in the ACS annually.
Each HU address in sample is assigned to a month for which it is eligible to receive a mail
questionnaire. Addresses that are not complete enough for mailing purposes are termed
unmailable. An interview for a HU and its residents can be completed during the assigned
month or the two months following it. All addresses mailed a questionnaire for which no
response is received during the assigned month and that have an available telephone number are
sent to the computer-assisted telephone interviewing (CATI) staff to conduct interviews during
the following month. After the CATI month, a sub-sample of the cases for which neither a
completed questionnaire has been received nor a CATI interview completed, and a sub-sample of
unmailable addresses is selected for computer-assisted personal interviewing (CAPI) in the third
month. Each GQ sample is assigned to a month and in most GQ facilities six weeks are allowed
for collection of data by personal interview only. Details on the data collection methods are
provided in Chapters 7 and 8.
This chapter presents details about the selection of the HU address samples, the GQ samples, and
the CAPI sub-samples. Some hard-to-reach areas in Alaska, referred to as Remote Alaska, are
treated differently from the rest of the ACS counties 1 in several of the sampling and data
collection processes. The section on Remote Alaska sampling at the end of this chapter
describes these differences.

4.2 HOUSING UNIT SAMPLING – INITIAL SAMPLE
Two separate sampling operations are carried out at different times to determine the addresses
that will be assigned to each month during a given calendar year -- the main sample is selected in
August/September preceding the sample year and the supplemental sample is selected in
January/February of the sample year. The ACS frame of HU addresses for each of these samples
is derived from the most recently available updated Master Address File (MAF), so the frames
differ between the main and supplemental samples for a given year. The MAF available at the
time of the main sample is referred to as the main sample MAF and the MAF available at the
1

In the remainder of the chapter the term “county” includes county equivalents and municipios.

4-1

time of the supplemental sample is referred to as the supplemental sample MAF. In the main
sample addresses are selected from the full frame and allocated to all 12 months. In the
supplemental sample only addresses new to the frame since the main sampling are selected and
allocated to the months of April through December.
One of the ACS design requirements is that no HU address can be sampled more than once in
five years. To accommodate this restriction, the addresses in the frame are randomly split into
five sub-frames, each containing 20 percent of the frame. Addresses from only one of these subframes are eligible to be in the ACS sample in each year, and a sub-frame can be used only once
every five years. As a result, each main and supplemental sample is conducted in two phases.
The first phase splits the frame into the five sub-frames and the second phase selects addresses to
be included in the ACS from the sub-frame eligible for the sample year.
Sampling Rate Categories
Before the first phase of address sampling can proceed for each year’s main sample, the
sampling rate category to which each address belongs must be determined. This is because the
ACS produces estimates for geographic areas having a wide range of populations, and in order
that the estimates for these areas have similar reliabilities as measured by their variances, areas
with smaller populations must have their HU addresses sampled at higher rates than those with
larger populations. To accomplish this, blocks and their constituent addresses are separated into
five categories with different sampling rates. The addresses in each of these categories are
allocated evenly to the five sub-frames in the first-phase sample to ensure that the variances of
individual estimates do not differ too much across years. Sampling rates for these categories are
calculated for use in the second phase of sampling.
Every block’s sampling rate category is determined based on information about both the census
tract and the sampling entity where it is located. Sampling entities are defined as counties;
places with active and functioning governments; tracts; school districts; American Indian
Area/Alaska Native Area/Hawaiian Home Land (AIANHH); minor civil divisions (MCDs) in 12
states 2 for which they are considered as functioning and active governmental units for ACS
sampling; and, in Hawaii only, census designated places. The sampling rate for each block is
based on:
1. The estimated number of occupied HUs in the smallest sampling entity that contains the
block or, in American Indian/Alaska Native Village Statistical Areas (AIANSAs) 3 , the
estimated number of occupied HUs with an American Indian/Alaskan Native resident;
and
2. The estimated number of occupied HUs in the block’s tract.

The 12 states are Connecticut, Maine, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New
York, Pennsylvania, Rhode Island, Vermont, and Wisconsin.
3
AIANSA is a general term used to describe American Indian and Alaska Native statistical areas. For detailed
technical information on the Census Bureau’s American Indian and Alaska Native Areas Geographic Program for
Census 2000, see Federal Register Notice Vol. 65, No. 121, June 22, 2000.

4-2

Factor 1 is called a block’s measure of size (MOS) as defined by its smallest sampling entity and
is calculated in the following manner. Block-level counts are derived from the main sample
MAF for each block in the nation. This count is converted to an estimated number of occupied
HUs by multiplying it by the proportion of the HUs in the block that was occupied in Census
2000. For each sampling entity the estimated number of occupied HUs is summed across all
blocks in the entity.
For AIANHHs, which can cross state boundaries, the estimated counts are cumulated across
states. If a block is not in an AIANSA, then the smallest of these sums across all sampling
entities in which it is contained is its MOS. If a block is in an AIANSA, then an additional set of
calculations is required. Each block in an AIANSA has its estimated number of occupied HUs
converted to an estimated number of occupied HUs having an American Indian or Alaska Native
householder by multiplying it by the proportion of its population that was American Indian or
Alaska Native in Census 2000. If the sum of these estimates across all blocks in an AIANSA is
non-zero, then it is the MOS for all blocks in the AIANSA. If it is zero, then the estimate of
occupied HUs in the AIANSA is the MOS for all of its blocks. See Hefter (2005a) for additional
details.
Factor 2 is obtained by summing the estimated number of occupied HUs across all blocks in a
tract. Using the two factors, blocks are separated into five sampling rate categories as defined in
Table 4.1 below. These categories are consistent with the categories used in Census 2000.
Table 4.1. Sampling Rate Categories for the 2005 ACS
Sampling Rate Category
Blocks in smallest sampling entities
Blocks in smaller sampling entities
Blocks in small sampling entities
Blocks in large tracts
All other blocks

Measure of Size (MOS)
<200
200≤MOS<800
800≤MOS≤1200
12002000
1200 1200, tract occupied HUs > 2000)
Mailable addresses ≥ 75% and predicted
levels of completed interviews prior to subsampling > 60%
Mailable addresses < 75% and/or predicted
levels of completed interviews prior to subsampling ≤ 60%
All other blocks
(MOS > 1200, tract occupied HUs ≤ 2000)
Mailable addresses ≥ 75% and predicted
levels of completed interviews prior to subsampling > 60%
Mailable addresses < 75% and/or predicted
levels of completed interviews prior to subsampling ≤ 60%

Overall Sampling Rates
United States
Puerto Rico
Initial
Final
Initial and Final
(2)
(3)
(4)
10.0%

10.0%

6.9%

8.1%

3.6%

4.1%

1.7%

2.0%

1.6%

1.7%

2.3%

2.7%

2.1%

2.3%

Final Overall Sampling Rates. As part of maintaining the reliability of estimates across
areas with differing proportions of the initial sample subject to CAPI sub-sampling, these
initial overall rates are reduced for selected tracts in the United States. This reduction is
based on the predicted level of completed interviews prior to CAPI sub-sampling for each
tract.
Predicted rate of completed interviews prior to CAPI sub-sampling = (annual number of
interviews completed by the mail and CATI modes, by the end of their CATI month, for
addresses mailed questionnaires)/(annual estimated number of occupied HUs mailed
questionnaires)
Since these rates were not available for all tracts in the country prior to 2005, projections
were used. For tracts that were included in the 2000-2003 ACS, these projections were based
on production data from those years. In the remaining tracts these rates were projections
based on a model that also uses information from Census 2000.
In the 2005 ACS, the initial overall sampling rate for each block in the United States in one
of the last two sampling rate categories from Table 4.2 (MOS>1200) was multiplied by 0.92
to obtain its final overall sampling rate if its tract had a level of completed interviews prior to
CAPI sub-sampling of at least 60 percent and at least 75 percent of its addresses defined as
4-5

mailable. In the remaining blocks in the United States the final overall sampling rates are the
same as the initial overall sampling rates, so there are seven different final overall sampling
rates in the United States as shown in column (3). A brief description of the relationship
between this reduction and CAPI sub-sampling rates is given in Section 4.3. For full details
see Asiala (2005). If the reduction factor is recalculated in the future, it will be based on
levels of completed interviews prior to CAPI sub-sampling obtained directly from the ACS
for all tracts. This reduction does not occur in Puerto Rico, so all blocks in Puerto Rico have
the same initial and final overall sampling rates.
Second-Phase Sample
The second-phase sample selects a subset of the addresses from the first-phase sample to be
included in the ACS during the sample year. In the main sample, addresses selected for a county
in the first phase are sorted by sampling rate category, new versus previously existing address,
and first-phase order of selection. In the supplemental sample they are sorted by sampling rate
category and first-phase order of selection. A systematic sample of addresses is selected for both
the main and supplemental samples, where each address has a second-phase sampling rate equal
to its final overall sampling rate divided by 20 percent, the first-phase sampling rate for all
addresses.
Month Assignment
Each address selected in the second-phase sample is assigned to a month. The set of all
addresses assigned to a specific month is referred to as the month’s sample. For the main
sample, most of the addresses are sorted by their second-phase order of selection and assigned in
a specified order to the 12 months of the year, starting with a randomly determined month.
However, addresses that have been selected for one of several Census Bureau household
surveys 4 in the same year or in specified months (which vary by survey) of the two surrounding
years are treated differently. Each of these addresses is assigned a month based on its interview
month(s) for the other survey in such a way as to avoid individuals being asked to complete
interviews for both the ACS and one of these surveys within a few months of each other.

These surveys include the Survey of Income and Program Participation, the National Crime Victimization Survey,
the Consumer Expenditures Quarterly and Diary Surveys, the Current Population Survey, and the State Child Health
Insurance Program.

4-6

In the supplemental sample, the addresses are sorted by their second-phase order of selection and
assigned to the months of April through December in a specified order, starting with a randomly
determined month. Since the supplemental sample will have very few addresses that are also in
one of the other household surveys in the specified months, no special month assignment is
performed.

4.3 HOUSING UNIT SAMPLING -- COMPUTER-ASSISTED PERSONAL
INTERVIEWING SUB-SAMPLE
The ACS uses three consecutive modes of data collection—mail, telephone, and personal visit.
(See Chapter 7 for more information.) An interview for a HU and its residents can be completed
during the month to which the address is assigned or the two following months. All addresses
mailed a questionnaire can return a completed questionnaire during this three-month time period.
The mailable addresses for which no response is received during the assigned month and that
have an available telephone number are sent to a telephone follow-up operation to conduct
interviews using CATI methods during the following month. No sampling is done for CATI.
After the CATI month, cases are selected for personal visit follow-up using CAPI methods in the
third month. The universe of CAPI cases for the third month consists of the mailable addresses
for which neither a completed questionnaire has been received nor a CATI interview completed
and all unmailable addresses. A systematic sub-sample of these addresses is selected for CAPI,
using the rates in Table 4.3, after sorting them within county by mailable versus unmailable,
CAPI sub-sampling rate, and geographical order on the MAF. The determination of CAPI subsampling rates is summarized in the following paragraphs. See Hefter (2005d) for details of
CAPI sub-sampling.
During the years prior to 2005, when the ACS was being developed, only two CAPI subsampling rates were used: 1-in-3 for mailable addresses and 2-in-3 for unmailable addresses.
Because the reliability of estimates for HUs and people living in HUs is a function of the number
of interviews completed within the area, these sub-sampling rates were revisited. Because
interviews are attempted for only a fraction of the cases eligible for CAPI and the proportion of
completed CAPI interviews are similar across areas, for two areas with the same number of
addresses in sample, the one with the highest rate of completed interviews prior to CAPI subsampling will tend to have more reliable estimates. To mitigate concerns about reliability
differentials resulting from large differences in the rate of completed interviews prior to CAPI
sub-sampling across areas, tracts in the United States with low predicted levels of completed
interviews prior to CAPI sub-sampling have their CAPI sub-sampling rates adjusted upward
from their default rates in order to potentially increase their numbers of interviews completed
across all modes. To compensate for the resulting increase in interviewing costs, the initial
overall sampling rates for tracts in the United States with the highest rates of completed
interviews prior to CAPI sub-sampling are reduced to obtain their final overall sampling rates.
This process was summarized in Section 4.2. Refer to Asiala (2005) for complete details.
No information was available about the levels of completed interviews prior to CAPI subsampling in Puerto Rico prior to 2005, so the default CAPI sub-sampling rates of 1-in-3 for
mailable addresses and 2-in-3 for unmailable addresses were used initially. Based on early

4-7

results observed during the first months of the ACS in Puerto Rico, the CAPI sub-sampling rate
for mailable addresses in all Puerto Rico tracts was changed to 1-in-2 beginning in June 2005.
Table 4-3. CAPI Sub-Sampling Rates for the 2005 ACS

Address and Tract Characteristics
United States
Unmailable addresses and
addresses in Remote Alaska
Mailable addresses in tracts with predicted levels of completed
interviews prior to CAPI sub-sampling between 0% and 35%
Mailable addresses in tracts with predicted levels of completed
interviews prior to CAPI sub-sampling greater than 35% and less
than 51%
Mailable addresses in other tracts
Puerto Rico
Unmailable addresses
Mailable addresses – June through December
Mailable addresses – January through May

CAPI
Sub-sampling Rate
66.7%
50.0%
40.0%
33.3%
66.7%
50.0%
33.3%

4.4 GROUP QUARTERS SAMPLING
GQ facilities include such places as college residence halls, residential treatment centers, skilled
nursing facilities, group homes, military barracks, correctional facilities, workers’ dormitories,
and facilities for people experiencing homelessness. Each GQ facility is classified according to
its GQ type. For more information on GQ facilities, see Chapter 8. As noted previously, GQ
facilities were not included in the 2005 ACS but are included in 2006 and will be included in
subsequent years. There is no supplemental GQ sample; therefore, all GQ samples are selected
during the main sampling phase. The sample is selected by state, not county, as is done with the
HU sample. Some locations that were classified in Census 2000 as specific GQ types are not
included in ACS data collection operations. These are domestic violence shelters, soup kitchens,
regularly scheduled mobile food vans, targeted non-sheltered outdoor locations, crews of
commercial maritime vessels, natural disaster shelters, and dangerous encampments. There are
several reasons for their exclusion, depending on GQ type, including concerns about privacy and
the operational feasibility of repeated interviewing for a continuing survey rather than once a
decade for a census. Nevertheless, ACS estimates of the total population are controlled to be
consistent with the intercensal population estimates. The exclusion of these GQ types may result
in a small bias in some ACS estimates.
Two strata are created to sample the GQ facilities. The first stratum includes both small GQ
facilities (having 15 or fewer people according to Census 2000 or updated information) and GQ
facilities listed as closed on Census Day. The second stratum includes large GQ facilities
(having more than 15 people according to Census 2000 or updated information). In the
remainder of this chapter, these strata will be referred to as the small GQ stratum and the large

4-8

GQ stratum. A GQ measure of size (GQMOS) is computed for use in sampling the large GQ
facilities. GQMOS is the population of the GQ facilities divided by 10.
Different sampling procedures are used for these two strata. The small GQ stratum is sampled
like the HU address sample and data are collected for all people in the selected GQ facilities.
Like a HU, a small GQ facility is eligible to be in sample only once in five years. Groups of 10
people are selected for interview from GQ facilities in the large GQ stratum and the number of
groups selected for a large GQ facility is a function of its GQMOS. Unlike HUs, large GQ
facilities are eligible for sampling each year. For details on GQ sampling, see Hefter (2005c).
Small Group Quarters Stratum Sampling
For the small GQ stratum, a two-phase sample of GQ facilities is selected in a method similar to
that used for the HU address sample.
First-Phase Sample. The first-phase sampling procedure for 2006 address assigns all of
the GQ facilities in this stratum to one of five 20 percent sub-frames. The GQ facilities
within each state are sorted by small versus closed on Census Day, new versus previously
existing, GQ type, and geographical order in the frame. Consecutive GQ facilities are
assigned to the five sub-frames in a specified order. One of these sub-frames was
designated as the 2006 first-phase sample and the others were designated for the years
2007-2010. The GQ facilities on the 2006 sub-frame will not be eligible for sampling
again until 2011, as the once-in-five-years restriction of HU addresses also applies to the
small GQ stratum.
Beginning with the 2007 sample, all small GQ facilities that are new since the previous
year’s sampling (there will be no new GQ facilities that were closed on Census Day) will
be sorted by GQ type and geographical order in the frame, and assigned to the five
existing 20 percent sub-frames in a specified order for the first-phase sample. In 2007,
the 2006 sub-frame will be defined as the 2011 sub-frame. The 2007 sub-frame becomes
the 2007 first-phase sample.
Second-Phase Sample. The second-phase sample is a 1-in-8 systematic sample of the
GQ facilities from the first-phase sample sorted by new versus previously existing address
and first-phase order of selection. Regardless of their actual size, all of these small GQ
facilities have the same probability of selection. Every person in the GQ facilities
selected in this second-phase sample is eligible to be interviewed. If the actual number of
people in the GQ facility exceeds 15, even though it was included in the small GQ
stratum, a field sub-sampling operation is performed to reduce the total number of sample
people in the GQ facility to 10.
Large Group Quarters Stratum Sampling
The ultimate sampling units for large GQ facilities are groups of 10 people within a GQ facility.
The number of groups of 10 people that can be sampled from a large GQ facility is equal to its
GQMOS. For example, if a GQ facility had 550 people in Census 2000, then its GQMOS is 55

4-9

and there are 55 groups of 10 people that are eligible for selection in the sample. A two-step
sampling procedure is used to select the people to be included in the large GQ sample. The first
step selects the GQ facilities and the number of groups in sample for each GQ facility. The
second step selects the people within each GQ facility to be included in its sample groups.
First-Step Sample. For the first step, all the large GQ facilities in a state are sorted by
GQ type and geographical order in the frame, and a systematic sample of 1-in-40 groups
is selected. GQ facilities with fewer than 40 groups, therefore, may or may not have one
of its groups selected for the sample; GQ facilities with between 40 and 80 groups will
have at least one group selected; GQ facilities with between 80 and 120 groups will have
at least two groups selected; and the like.
Second-Step Sample. The second step of sampling takes place within each GQ facility
which has at least one group selected in the first step. When a field representative visits a
GQ facility to conduct interviews, an automated listing instrument is used to select the 10
people to be included in each group being interviewed. It computes sampling parameters
based on the observed population at the time of interviewing in order to achieve a group
size of 10. If the observed population is 15 or fewer, then the group size will be at most
10.
For most GQ types, if multiple groups are selected within a GQ facility, their groups of
10 people are assigned to different sample months. Very large GQ facilities with more
than 12 groups selected have multiple groups assigned to some sample months. In these
cases, we attempt to not select the same person more than once in a sample month.
However, there is no attempt to not select a person more than once across sample months
within a year. All GQ facilities in this stratum are eligible for selection every year,
regardless of their sample status in previous years.
Sample Month Assignment
The selected small GQ facilities and groups of 10 people for large GQ facilities are assigned to
months in a manner similar to that used for HUs. All the GQ samples from a state are combined
and sorted by small versus large stratum and second-phase or second-step order of selection.
Consecutive samples are assigned to the 12 months in a pre-determined order, starting with a
randomly determined month, except for correctional facilities and military barracks.
Due to operational and budgeting constraints, the same month is assigned to all sample groups of
10 people within a given correctional facility or military barracks. The sample groups of 10 in
all federal correctional facilities are assigned to September and data collection may take up to 4.5
months, an exception to the six weeks allowed for all other GQ types. For the sample groups of
10 in non-federal correctional facilities -- state prisons, local jails, halfway houses, military
disciplinary barracks, and other correctional institutions -- or military barracks, individual GQ
facilities are assigned to months throughout the year.

4-10

4.5 REMOTE ALASKA SAMPLING
Remote Alaska is a set of rural areas in Alaska that are difficult to access. Due to data collection
related difficulties for field operations during some months, data collection operations in Remote
Alaska differ from what has been described for the rest of the country. In both the main and
supplemental samples, the month assigned for each Remote Alaska HU address is based on
certain types of geographical entities in which it can be contained. All addresses located in each
specific geography of these types are assigned to the same month, either January or September.
The assignment of each specific geographical entity to either January or September is done in
such a way as to balance workloads between these months and to keep groups of cases together
geographically. The addresses for each month are sorted by county and geographical order in the
frame, and a sample of 2-in-3 of them is sent directly to CAPI in the appropriate month. The GQ
sample in Remote Alaska is assigned to January or September in the same manner as are main
sample HU addresses. Up to four months is allowed to complete the HU interviews and GQ data
collection for each assigned month.

4.6 REFERENCES
Asiala, M. (2004), “Specifications for Selecting the ACS 2005 Main HU Sample (ACS-S-40),”
Draft Census Bureau Memorandum to L. McGinn from R. Singh, Washington, DC, August 27,
2004.
Asiala, M. (2005), “American Community Survey Research Report: Differential Sub-Sampling
in the Computer Assisted Personal Interview Sample Selection in Areas of Low Cooperation
Rates,” Census Bureau Memorandum to R. Singh from D. Hubble, Washington, DC, February
15, 2005.
Federal Register Notice (2000), “American Indian and Alaska Native Areas Geographic Program
for Census 2000; Notice,” Department of Commerce, Bureau of the Census, Volume 65, Number
121, Washington, DC, June 22, 2000.
Hefter, S. P. (2005a), “Creating the Governmental Unit Measure of Size (GUMOS) Datasets for
the American Community Survey and the Puerto Rico Community Survey,” 2006 American
Community Survey Sampling Memorandum Series #ACS06-S-1, Draft Census Bureau
Memorandum to L. McGinn from R. P. Singh, Washington, DC, June 30, 2005.
Hefter, S. P. (2005b), “Specifications for Selecting the Main and Supplemental Housing Unit
Address Samples for the American Community Survey,” 2006 American Community Survey
Sampling Memorandum Series #ACS06-S-3, Draft Census Bureau Memorandum to L. McGinn
from R. P. Singh, Washington, DC, June 16, 2005.
Hefter, S. P. (2005c), “Specifications for Selecting the American Community Survey Group
Quarters Sample,” 2006 American Community Survey Sampling Memorandum Series #ACS06S-6, Draft Census Bureau Memorandum to L. McGinn from R. P. Singh, Washington, DC, May
19, 2005.

4-11

Hefter, S. P. (2005d), “American Community Survey: Specifications for Selecting the Computer
Assisted Personal Interview Samples,” 2005 American Community Survey Sampling
Memorandum Series #ACS-S-45, Draft Census Bureau Memorandum to L. McGinn from R. P.
Singh, Washington, DC, May 23, 2005.

4-12

Chapter 5.
Content Development Process
5.1 OVERVIEW
The content for the American Community Survey (ACS) is designed to meet the needs of
federal government agencies. Additionally, these data are a rich source of local area
information, making the data useful to state and local governments, universities, and
private businesses of all sizes. The Census Bureau coordinates the content development
and determination process for the ACS with the Office of Management and Budget
(OMB) through interagency committee meetings based upon the needs of the federal
agencies. The OMB, in consultation with the Census Bureau, establishes new content
determinations for the survey. Approved new content or changes to existing content are
tested according to the ACS content change process, which provides Census Bureau
guidelines for pretesting, field testing, and implementing new content and changes to
existing ACS content. This chapter provides greater detail on the history of content
development for the ACS, the current content of the survey, and the content
determination process and policy.

5.2 HISTORY OF CONTENT DEVELOPMENT
The ACS is part of the 2010 Decennial Census Program and is an alternative method for
collecting the long form sample data collected in the last five censuses. The long form
sample historically collected detailed population and housing characteristics only once a
decade through questions asked of a sample of the population. 1 Beginning in 2005, the
ACS is collecting this detailed information from a sample of the nation’s population on
an ongoing basis, thereby providing more accurate and timely data than was possible
using the once-a-decade census long form. Starting in 2010, the decennial census will no
longer include long form sample data collection, but will include only a short form that
collects basic information for a total count of the nation’s population. 2
Historically, the content of sample questions included on the long form was constrained
by only including questions for which:
1. There was a current federal law that explicitly called for the use of decennial
census data for a particular federal program (mandatory).
1

Sampling began in the 1940 census, when a few long form questions were asked of a small sample of
people. A separate long-form questionnaire was not implemented until 1960.

In addition to counting each person in every household, the “basic” information planned for the Census
2010 short form will include a very small select set of key demographic information needed for voting
rights and other legislative requirements. Currently, data on gender, age, relationship, race, and ethnicity
are planned.

5-1

2. It was unequivocally clear that a federal law (or implementing regulation)
required the use of specific data and the decennial census was the historical or
only source of data (required).
3. The data were necessary for Census Bureau operational needs (programmatic).
Constraining the content of the ACS is just as critical as it had been for the long form
sample in the past. The Census Bureau’s goal is to limit the size of the ACS so that data
quality, measured by survey and item response rates, remains at or above the current
levels.
In order to accomplish this goal, the Census Bureau works closely with the OMB and the
OMB Interagency Committee for the ACS. This Committee (co-chaired by the OMB and
the Census Bureau) was established in July 2002 and includes representatives from more
than 20 federal departments and agencies that use decennial census data. The initial
focus of the committee was to identify legislative justifications for the questions on the
2003 ACS. The federal agencies were asked to examine each question on the ACS and
provide the Census Bureau with justification(s) for their need for the data by subject
matter, the legal authority for the use, the lowest geographic level required, the variables
essential for cross-tabulation, and the frequency with which the data are needed. They
were asked to provide actual statutes and other legislative documentation, with the
relevant sections annotated, and to classify their uses of the ACS questions as
“mandatory,” “required,” or “programmatic” consistent with the constraints of the
traditional long form. See Section 5.4.
In the summer of 2002, the Department of Commerce General Counsel’s Office asked
each federal agency’s General Counsel to examine the justifications submitted for its
respective agency and, if necessary, to revise the information so that it would include the
most up-to-date material necessary to accomplish the statutory mission of its department
in relation to census data. This step was taken to ensure that the highest-ranking legal
officer in each agency validated its stated program requirements and data needs.
Only those subjects classified as either mandatory or required were asked on the 2003
ACS questionnaire, along with two programmatic subjects, fertility and seasonal
residence. The end result was a 2003 through 2007 ACS questionnaire with content
almost identical to the Census 2000 long-form questionnaire content. The ACS
questionnaire was approved by the participating federal agencies and OMB, and
subsequently submitted to and approved by the Congress.

5.3 2003 - 2007 CONTENT
ACS Content
In 2005 and 2006, the ACS included 25 housing and 42 population questions (six basic
questions and 36 detailed questions). The identical set of questions was used in 2003 and
2004 and is scheduled for use in 2007. See Table 5-1 for a complete list of ACS topics.

5-2

Refer to Chapter 6 for definitions of each of these topics and a summary of how they are
organized within the ACS questionnaire. The ACS GQ questionnaire contains all
population questions that are contained in the population column in Figure 5-1, except the
relationship to householder. One housing question, food stamp benefit, is contained on
the ACS GQ questionnaire.

5-3

Table 5-1. ACS Topics Listed by Type of Characteristic and Question Number
Housing

Population

Household Size
H1 Units in Structure
H2 Year Structure Built
H3 Year Householder Moved Into Unit
H4 Acreage
H5 Agricultural Sales
H6 Business on Property
H7 Rooms
H8 Bedrooms
H9 Plumbing Facilities
H10 Kitchen Facilities
H11 Telephone Service Available
H12 Vehicles Available
H13 House Heating Fuel
H14 Cost of Utilities
H15 Food Stamp Benefit
H16 Condominium Status and Fee
H17 Tenure
H18 Monthly Rent
H19 Value of Property
H20 Real Estate Taxes
H21 Insurance for Fire, Hazard, and Flood
H22 Mortgage Status, Payment, Real
Estate Taxes
H23 Second or Junior Mortgage Payment
or Home Equity Loan
H24 Mobile Home Costs
H25 Seasonal Residence

Name
P1 Sex
P2 Age and Date of Birth
P3 Relationship to Householder
P4 Marital Status
P5 Hispanic Origin
P6 Race
P7 Place of Birth
P8 Citizenship
P9 Year of Entry
P10 Type of School and School Enrollment
P11 Educational Attainment
P12 Ancestry
P13 Language Spoken at Home, Ability to Speak English
P14 Residence 1 Year Ago (Migration)
P15 Disability: Sensory, Physical
P16 Disability: Mental, Self-care
P17 Disability: Going out Alone, Ability to Work
P18 Fertility
P19 Grandparents as Caregivers
P20 Veteran Status
P21 Period of Military Service
P22 Years of Military Service
P23 Worked Last Week
P24 Place of Work
P25 Means of Transportation
P26 Private Vehicle Occupancy
P27 Time Leaving Home to Go to Work
P28 Travel Time to Work
P29 Layoff, Temporarily Absent, Informed of Recall or
Return Date
P30 Looking for Work
P31 Available to Work
P32 When Last Worked
P33 Weeks Worked
P34 Usual Hours Worked Per Week
P35 Class of Worker
P36 Employer
P37 Type or Kind of Business
P38 Industry
P39 Occupation
P40 Primary Job Activity
P41 Income in the Past 12 Months (by type of income)
P42 Total Income

5-4

Puerto Rico Community Survey Content
The content for Puerto Rico is identical to the content used in the United States. The
Puerto Rico Community Survey (PRCS) includes six questions that are worded
differently from those on the ACS to accommodate cultural and geographic differences
between the two areas of data collection. See Figure 5-1 for an example of ACS
questions that were modified for the PRCS. (Refer to Appendix A for documentation of
changes that have been made to the ACS questionnaire.)

Figure 5-1. Example of Two ACS Questions Modified for the PRCS

ACS (2005)

PRCS (2005)

5.4 CONTENT POLICY AND CONTENT CHANGE PROCESS
The ACS is designed to produce detailed demographic, housing, social, and economic
data every year. Because it accumulates data over time to obtain sufficient levels of
reliability for small geographic areas, the Census Bureau must minimize content changes.
Consistency must be maintained throughout all ACS data collection operations, including
HUs and GQ facilities both in the United States and in Puerto Rico. Introducing changes
could affect data quality and result in only partial releases of data for a given year if a
question changes significantly or has not been asked for enough years to accumulate
three or five years worth of data.
In 2006, the Census Bureau adopted a more flexible approach to content determinations
for the ACS with the implementation of the ACS Content Policy. This policy states that
the OMB, in consultation with the Census Bureau, will determine any new content for the
ACS. In making content determinations, the OMB, in consultation with the Census
Bureau, will consider issues, such as frequency of data collection; the level of geography
needed to meet the required need; and whether any other source of data would meet the
requestor’s need in lieu of collection on the ACS.

5-5

The Census Bureau recognizes and appreciates the interests of federal partners and
stakeholders in the collection of data on the ACS. The fact that respondents’
participation in the ACS is mandatory requires that the OMB will only approve, and the
Census Bureau will only ask, necessary questions. The OMB’s responsibility under the
Paperwork Reduction Act requires that practical utility of the data be demonstrated and
that respondent burden be kept to a minimum.
The ACS Content Policy is used as a basic guideline for all new question proposals from
federal agencies, the Congress, and the Census Bureau. The ACS Content Change
Process is part of a risk management strategy to ensure that each new or modified
question has been fully tested and will collect quality data without reducing overall
survey response rates.
The ACS Content Policy provides guidance for on-going ACS content development. The
implementation of this policy includes Census Bureau coordination of input from internal
and external groups. The Interagency Committee for the ACS obtains broad input from
all federal agencies. The Census Bureau coordinates the creation of subject area
subcommittee groups that include representatives from the Interagency Committee for the
ACS and the Census Bureau who provide expertise in designing sets of questions and
response categories so the questions will meet the needs of all the agencies. The Census
Bureau’s ACS Content Council members review the subcommittee proposals and provide
comments and internal approval of content changes.
The ACS Content Change Process provides guidance for Census Bureau pretesting,
including a field test, of all new or modified questions prior to incorporating them into the
ACS instruments based on the standards outlined in the Census Bureau’s Standard:
Pretesting Questionnaires and Related Materials for Surveys and Censuses. After new
or modified questions have been fully pretested, data collection of the tested content will
begin at the start of an ACS production year (January). Pretested, new questions will
only be added to the production ACS after OMB approval has been provided to the
Census Bureau.
Content Change Factors
The OMB and the Census Bureau consider several factors when new content is proposed
for the ACS. Additionally, federal agencies must provide the OMB and the Census
Bureau with specific information about the new data collection need(s).
The uses of the data must be determined in order to ascertain the appropriateness of
collecting the new data through a national, mandatory survey, or if another Census
Bureau data-collection operation would better meet the need. ACS data are collected and
tabulated at the tract or block-group level and the response burden to the majority of
respondents that are required to answer a question that is not applicable to them must be
considered. The Census Bureau is bound by law (Title 13, United States Code (U.S.C.),

5-6

Section 9) to ensure that the data produced preserves the confidentiality of individual
respondents. Therefore, it is critical to understand how the collected data will be used.
The principal benefits of the ACS are that it is a mandatory, annual, and national survey.
Federal agencies interested in content changes (new or modifications) must be able to
demonstrate that they require detailed data with the frequency of ACS data collection,
and that failure to obtain the information with ACS’s frequency of data collection and
release would not meet the agency’s needs. Requests for new ACS content will be
assessed with regard to the impact on the requesting federal agency if the data are not
collected through the ACS. Federal agencies requesting new content must also
demonstrate that they have considered legitimate alternate sources for the data, including
both existing and new data collections and administrative records, to meet their needs at
the detail level requested, and why those alternative sources do not meet their needs.
Content Change Requirements
Federal agency or Census Bureau proposals for new content and/or changes to existing
ACS questions due to identified quality issues are subject to the following requirements:
•

ACS content can be added to or revised only once a year due to the annual nature
of the survey and the number of operations that must also be revised to reflect the
change. New content will only be incorporated into the ACS after pretesting,
including a field test, has been completed and OMB has provided final approval
for inclusion of the new content on the ACS.

•

The requesting federal agency will assist with the development of a draft
question(s), work with the Census Bureau and other agencies to develop or revise
the question, and submit the proposal to the OMB and Census Bureau for further
review. In addition to drafting test questions, a plan to pretest new or modified
content, including a field test, must be developed in accordance to the Census
Bureau’s Standard: Pretesting Questionnaires and Related Materials for Surveys
and Censuses.
Pretesting is conducted to detect respondent error and to determine whether or not
a change would increase or decrease respondent’s understanding of what is being
asked. Alternative versions of questions are pretested to try to identify one
version most likely to be answered accurately by respondents. The resulting
question is field-tested.

•

The requesting federal agency must support the cost of pretesting, including a
field test, when proposing new or modified content for the ACS. To meet Census
Bureau operational deadlines, all of the activities above must be carried out early
enough to finalize changes and incorporate them before ACS production
deadlines.

5-7

5.5 2006 CONTENT TEST
In 2004, planning began on a 2006 ACS Content Test that would provide the last
opportunity to field test content changes in the ACS before the 2008 instrument is
finalized. The OMB and the Census Bureau again asked members of the ACS
Interagency Committee to review their legislative authority for their current or proposed
ACS questionnaire content and to identify any questions that needed to be reworded or
reformatted due to data quality issues or changes to the mandatory or required data
collection needs of the requesting federal agency.
The 2006 ACS Content Test is the first opportunity to test revisions to the long-form
sample questions used in Census 2000. The content of the 2006 ACS Content Test
includes new questions on the subjects of marital history, health insurance and coverage,
and veteran’s service-related disability.
The test methodology for the 2006 ACS Content Test is designed to be similar to ACS
data collection in the production phase and incorporates the prenotice letter, initial
mailing package, reminder postcard, and second mailing package if no response is
received after three weeks. A computer-assisted personal interview (CAPI) follow-up is
also planned. To measure response error, a computer-assisted telephone interview
(CATI) content re-interview is conducted. Along with other data quality measures, such
as item non-response rates and measures of distributional changes, simple response
variance and gross difference rates serve as indicators of the quality of the test questions
relative to the current versions of the ACS questions.
The data analysis from the content test and content re-interview will guide the selection
of the version of each question that yields the highest quality data. Census Bureau
analysts, federal experts, and subject matter experts will examine the data to determine
the final content recommendations. The OMB, in consultation with the recommendations
of the Census Bureau, will make final content determinations for the 2008 ACS. These
approvals will occur by January 2007, so that the changes can be reflected and
implemented in the 2008 ACS. Refer to U.S. Census Bureau (2005a) for copies of the
four versions of the 2006 ACS Content Test questionnaires.

5.6 REFERENCES
U.S. Census Bureau (2003a), “American Community Survey Operations Plan,” U.S.
Census Bureau, Washington, DC, 2003.
U.S. Census Bureau (2003b), “Census Bureau Standard: Pretesting Questionnaires and
Related Materials for Surveys and Censuses,” U.S. Census Bureau, Washington, DC,
2003.
U.S. Census Bureau (2005a), “American Community Survey Content Test,” U.S. Census
Bureau, Washington, DC, 2005.

5-8

U.S. Census Bureau (2005b), “Subjects Planned for Inclusion in the 2005 American
Community Survey,” U.S. Census Bureau, Washington, DC, 2005.
U.S. Census Bureau (2006), “U.S. Census Bureau Policy on New Content for the
American Community Survey,” U.S. Census Bureau, Washington, DC, 2006.

5-9

Chapter 6.
Survey Rules, Concepts, and Definitions
6.1 OVERVIEW
Interview and residence rules define the universe, or target population for a survey. These rules
therefore identify the units and people eligible for inclusion in the survey. The American
Community Survey (ACS) interviews the resident population living in both housing units (HUs)
and group quarters (GQ) facilities. The ACS uses residence rules based on the concept of
current residence. Sections 6.2 and 6.3 in this chapter detail the interview and residence rules.
This chapter also includes a discussion of the full set of topics included in the ACS. This
discussion is organized into four sections to parallel the organization of the ACS questionnaire.
Definitions are provided for all address, HU status, and household information, basic
demographic information, detailing housing information, and detailed population information.

6.2 INTERVIEW RULES
The Census Bureau classifies all living quarters as either HUs or GQ facilities. A HU is a house,
an apartment, a group of rooms, or a single room that is either occupied or intended for
occupancy as a separate living quarters. GQ facilities are places where people live or stay that
are normally owned and managed by an entity or organization that provides housing and/or
services for the residents. GQ facilities include correctional facilities and living quarters such as
group homes, healthcare and treatment facilities, and college dormitories. See Chapter 8 for a
full description of GQ facilities.
Interview rules define the scope of data collection by defining the types of places that are
included in the sample frame and the people who are eligible for inclusion in the survey.
Beginning in 2006, the ACS includes HUs and GQ facilities. Only HUs and the people living in
HUs were included in the 2005 ACS. Like the decennial census, the ACS interviews the resident
population without regard to the person’s legal status or citizenship. The ACS excludes people
residing in HUs only if the residence rules (see below) define their current residence to be
somewhere other than the sample address.

6.3 RESIDENCE RULES
Residence rules are the series of rules that define who (if anyone) should be interviewed at a
sample address, and who is considered, for purposes of the survey or census, to be a resident of
the sample address. Residence rules decide the occupancy status of each HU and the people
whose characteristics are to be collected.
6-1

The ACS collects data nearly every day of the year. The survey’s residence rules are applied and
its reference periods are defined as of the date of the interview. For mail returns, this is the date
that the respondent completes the questionnaire, while for telephone and personal visit
interviews it is the date when the interview is conducted.
Residence Rules
Housing Units. The ACS defined the concept of current residence to determine who
should be considered residents of sample HUs. This concept is a modified version of a
de facto rule where a time interval is used to determine residency.1 The basic idea behind
the ACS current residence concept is that everyone who is currently living or staying at a
sample address is considered a resident of that address, except people only staying there
for a short period of time. People who have established residence at the sample unit and
are away from this unit for only a short period of time are also considered to be current
residents. For the purposes of the ACS, the Census Bureau defines this “short period of
time” as two consecutive months, and the ACS current residence rule is often described
as the “two-month rule.” Under this rule, anyone who is living for more than two months
in the sample unit when the unit is interviewed (either by mail, telephone, or personal
visit) is considered a current resident of that sample unit. This means that their expected
length of stay is more than two months, not that they have been staying in the sample unit
for more than two months. For the ACS, the Census Bureau classifies a HU in which no
one is determined to be a current resident, as vacant. In general, people who are away
from the sample unit for two months or less are considered to be current residents, even
though they are not staying there when the interview is conducted, while people who
have been or will be away for more than two months are not considered to be current
residents.
Residency is determined as of the date of the interview. A person who is living or
staying in a sample HU on interview day and whose actual or intended length of stay is
more than two months is considered a current resident of the unit. That person will be
included as a current resident of the unit unless he or she, at the time of interview, has
been or intends to be away from the unit for a period of more than two months. There are
three exceptions to this rule.
•

Children (below college age) who are away at boarding school or summer camp for
more than two months are always considered current residents of their parents’ home.

•

Children who live under joint custody agreements and move between residences are
always considered current residents of the sample unit where they are staying at the
time of the interview.

A de facto rule would include all people who were staying at an address when an interview was conducted,
regardless of the time spent at this address. It would exclude individuals away from a regular residence even if they
were away only for that one day.

6-2

•

People who stay at a residence close to work and return regularly to another residence
to be with their family are always considered current residents of the family
residence.

A person who is staying at a sample HU when the interview is conducted but has no
place where he or she stays for periods of more than two months is also considered to be
a current resident of the sample HU. A person whose length of stay in the sample HU is
only for two months or less and has another place where he or she stays for periods of
more than two months is not a current resident of the unit.
Group Quarters. Residency in GQ facilities is determined by a purely de facto rule.
All people staying in the GQ facility when the roster of residents is made and sampled
are eligible to be selected to be interviewed in the ACS. The GQ sample universe will
include all people residing in the selected GQ facility at the time of interview. Data are
collected for all people sampled regardless of their length of stay in the GQ facility.
Children (below college age) staying at a GQ facility functioning as a summer camp are
not considered to be GQ residents.
Reference Period. As noted earlier, the survey’s reference periods are defined relative
to the date of the interview. The survey questions define the reference periods and
always include the date of the interview. When the question does not specify a time
frame, respondents are told to refer to the situation on the interview day. When the
question mentions a time frame, it refers to an interval that includes the interview day and
covers a period before the interview. For example, a question that asks for information
about the “past 12 months” would be referring to the previous 12 months relative to the
date of the interview.

6.4 STRUCTURE OF THE HOUSING UNIT QUESTIONNAIRE
The ACS questionnaires and survey instruments used to collect data from the HU population are
organized into four sections, each section collecting a specific type of information. Refer to
Appendix B.7 for a copy of the HU questionnaire. The first section verifies basic address
information, determines the occupancy status of the HU, and identifies who should be
interviewed as part of the ACS household. The second section of the questionnaire collects basic
demographic data. The third section collects detailed housing information and the final section
collects detailed population data.
Data are collected by mail, telephone, and in person. Data collection instruments therefore exist
for all three of these modes. The telephone instrument is a computer-assisted telephone
interview (CATI) instrument and the personal interview instrument is a computer-assisted
personal interview (CAPI) instrument. See Chapter 7 for more detailed information on data
collection. This section describes the basic data collection process from a personal visit
perspective but the same basic process is followed in the mail and telephone modes.
6-3

Address, Housing Unit Status, and Household Information
During personal visit follow-up the field representative (FR) first must verify that he or she has
reached the sample address and then must determine if the sample address identifies a HU. If
not, the address is not eligible to be interviewed and is considered out of scope. Out-of-scope
addresses include those determined to be non-existent because, for example, the HU has been
demolished, or because it identifies a business and not a residential unit. Interviewers use the
residence rules to determine whether the sample HU is occupied (at least one person staying in
the unit is a current resident) or vacant (no one qualifies as a current resident of the unit).
Interviewers also apply the residence rules to create a household roster of current occupants to
interview. The name of the household respondent and the telephone number are collected in
case follow-up contact is needed. The terms below are critical to this data collection.
Housing unit (HU). A HU may be a house, an apartment, a mobile home or trailer, a
group of rooms or a single room that is occupied (or, if vacant, intended for occupancy)
as separate living quarters.
Housing unit (HU) status. All sample addresses are assigned a status of either an
occupied, vacant, or temporarily occupied HU, or a status of delete, indicating the
address does not identify a HU. A temporarily occupied unit is a HU where at least one
person is staying, but where no people are determined to be current residents. These
addresses are considered a type of vacant unit. Deleted units are addresses determined to
represent commercial units or HUs that are demolished or non-existent.
Household. A household is defined to be all people (including both related and
unrelated individuals) whose current residence at the time of the ACS interview is the
sample address.
Household roster. The list of all current residents of the sample address and therefore
all people whose characteristics will be collected in the interview according to the ACS
residence rule.
Household respondent. One person may provide data for all members of the household.
The Census Bureau refers to this person as the household respondent. ACS interviewers
try to restrict their household respondents to members who are at least 18 years old, but if
necessary, household members who are 15 and older can be a household respondent. If
no household member can be found to provide the survey information, the interviewer
must code the case as a non-interview.

6-4

Basic Demographic Information
The basic demographic data of sex, age, relationship, marital status, Hispanic origin, and race are
collected first in the ACS and are considered the most critical data items. They are used in many
of the survey’s tabulations. Age defines the critical paths and skip patterns used in the
instrument/questionnaire. Name is also collected for all household members. One individual in
the household must be identified as a reference person to define relationship within the
household. The section below provides details of the concepts and definitions associated with
the basic demographic data. Additional details on race, relationship, and Hispanic origin
definitions can be found on the ACS website. See
.
Reference person or householder. One person in each household is designated as the
householder. Usually this is the person, or one of the people, in whose name the home is
owned, being bought, or rented and who is listed as “Person 1” on the survey
questionnaire. If there is no such person in the household, any adult household member
15 and older can be designated as the householder. This person is listed first on the form.
Sex. Each household member’s sex is marked as “male” or “female”.
Age and date of birth. The age classification is based on the age of the person in
complete years at the time of interview. Both age and date of birth are used to calculate
each person’s age on the interview day.
Relationship. The questionnaire/instruments ask for each household member’s
relationship to the reference person/householder. Categories include both relatives and
non-relatives.
Martial status. The marital-status question is asked of everyone responding on the mail
form but only of people 15 and older responding through CATI and CAPI interviews.
The response categories are “now married,” “widowed,” “divorced,” “separated,” or
“never married”. Couples who live together (unmarried people, people in common-law
marriages) are allowed to report the marital status they consider the most appropriate.
Hispanic origin. Hispanic origin can be viewed as the heritage, nationality group,
lineage, or country of birth of each person or of each person’s parents or ancestors before
their arrival in the United States. People who identify their origin as Spanish, Hispanic,
or Latino may be of any race. Like the concept of race, Hispanic origin is based on
self-identification.
Race. The concept of race, as defined by the Office of Management and Budget (OMB)
and used by the Census Bureau, reflects self-identification by people according to the
race or races with which they most closely identify. These categories are socio-political

6-5

constructs and should not be interpreted as scientific or anthropological in nature. The
race categories include both racial and national-origin groups.
Detailed Housing Information
The housing section collects data on physical and financial characteristics of housing. The
2005/2006 ACS questionnaire includes 25 detailed housing questions. Selected housing data are
collected for vacant and temporarily occupied HUs. For these units the information is
determined either by observation or from a neighbor or other source. This section details the
concepts associated with some of the housing items. A full description of concepts, definitions,
and terminology used with ACS housing data are available on the ACS website. See
.
Units in structure. All HUs are categorized by the type of structure in which they are
located. A structure is a separate building that either has open spaces on all sides or is
separated from other structures by dividing walls that extend from ground to roof. In
determining the number of units in a structure, all HUs, both occupied and vacant, are
counted. Stores and office space are excluded.
Year structure built. This question determines when the building in which the sample
address is located was first constructed, not when it was remodeled, added to, or
converted. The information is collected for both occupied and vacant HUs. Units that
are under construction are not considered housing units until they meet the HU definition
– that is, when all exterior windows, doors, and final usable floors are in place. This
determines the year it was built. For mobile homes, houseboats, recreational vehicles,
the manufacturer's model year is assumed to be the year the unit was built.
Year householder moved into unit. This question is only collected for occupied HUs,
and these data refer to the year of the latest move by the householder. If the householder
moved back into a HU he or she previously occupied, the year of the last move is to be
reported. If the householder moved from one apartment to another within the same
building, the year the householder moved into the present apartment is reported. The
intent is to establish the year the current occupancy of the unit by the householder began.
The year that the householder moved in is not necessarily the same year other members
of the household moved in, although usually an entire household moves at the same time.
Acreage. This question determines a range of acres on which the house or mobile home
is located. A major purpose for this item is to identify farm units.
Agricultural sales. This item refers to the total amount (before taxes and expenses)
received in the 12 months prior to the interview, from the sale of crops, vegetables, fruits,
nuts, livestock and livestock products, and nursery and forest products, produced on this
property. This item is used to classify HUs as farm or non-farm residences.

6-6

Business on property. A business must be easily recognizable from the outside. It
usually will have a separate outside entrance and the appearance of a business, such as a
grocery store, restaurant, or barbershop. It may be either attached to the house or mobile
home or located elsewhere on the property.
Rooms. The intent of this question is to determine the number of whole rooms in each
HU that are used for living purposes. Living rooms, dining rooms, kitchens, bedrooms,
finished recreation rooms, enclosed porches suitable for year-round use, and lodgers’
rooms are to be included. Excluded are strip or Pullman kitchens, bathrooms, open
porches, balconies, halls or foyers, half rooms, utility rooms, unfinished attics or
basements, or other unfinished space used for storage. A partially divided room is a
separate room only if there is a partition from floor to ceiling, but not if the partition
consists solely of shelves or cabinets.
Bedrooms. Bedrooms include only rooms designed to be used as bedrooms; that is, the
number of rooms that the respondent would list as bedrooms if the house, apartment, or
mobile home were on the market for sale or for rent. Included are all rooms intended to
be used as bedrooms even if they currently are being used for another purpose. A HU
consisting of only one room is classified, by definition, as having no bedroom.
Plumbing facilities. Answers to this question are used to estimate the number of HUs
that do not have complete plumbing facilities. Complete plumbing facilities include: hot
and cold piped water, a flush toilet, and a bathtub or shower. All three facilities must be
located inside the house, apartment, or mobile home, but not necessarily in the same
room. HUs are classified as lacking complete plumbing facilities when any of the three
facilities is not present.
Kitchen facilities. Answers to this question are used to estimate the number of HUs that
do not have complete kitchen facilities. A unit has complete kitchen facilities when it has
all three of the following facilities: a sink with piped water, a range or cook top and
oven, and a refrigerator. All kitchen facilities must be located in the house, apartment, or
mobile home, but they need not be in the same room. A HU having only a microwave or
portable heating equipment such as a hot plate or camping stove is not considered as
having complete kitchen facilities.
Telephone service available. For an occupied unit to have telephone service available,
there must be a telephone in working order and service available in the house, apartment,
or mobile home that allows the respondent to both make and receive calls. Households
whose service has been discontinued for non-payment or other reasons are not considered
to have telephone service available. Beginning in 2004, the instructions that
accompanied the ACS mail questionnaire advised respondents to answer that the house or
apartment has telephone service if cellular telephones are used by household members.

6-7

Vehicles available. These data show the number of passenger cars, vans, and pickup or
panel trucks of one-ton capacity or less kept at home and available for the use of
household members. Vehicles rented or leased for one month or more, company vehicles,
and police and government vehicles are included if kept at home and used for nonbusiness purposes. Dismantled or immobile vehicles are excluded. Vehicles kept at
home but used only for business purposes also are excluded.
House heating fuel. House heating fuel is only collected for occupied HUs. The data
show the type of fuel used most to heat the house, apartment, or mobile home.
Selected monthly owner costs. Selected monthly owner costs are the sum of payments
for mortgages, deeds of trust, contracts to purchase, or similar debts on the property; real
estate taxes: fire, hazard, and flood insurance on the property; utilities (electric, gas,
water, and sewer); and fuels (oil, coal, kerosene, wood, and so on). It also includes,
where appropriate, the monthly condominium fee for condominiums and mobile home
costs.
Food stamp benefit. The Food and Nutrition Service of the U.S. Department of
Agriculture (USDA) administers the Food Stamp Program through state and local welfare
offices. The Food Stamp Program is the major national income-support program to
which all low-income and low-resource households, regardless of household
characteristics, are eligible. The question concerning the receipt of food stamp benefits
estimates the number of households that received the benefit at any time during the
12-month period before the ACS interview.
Tenure. All occupied HUs are divided into two categories – owner-occupied and
renter-occupied. A HU is owner-occupied if the owner or co-owner lives in the unit even
if it is mortgaged or not fully paid for. All occupied HUs that are not owner occupied,
whether they are rented for cash rent or occupied without payment of cash rent, are
classified as renter occupied.
Contract rent. Contract rent is the monthly rent agreed to or contracted for, regardless
of any furnishings, utilities, fees, meals or services that may be included.
Gross rent. Gross rent is the contract rent plus the estimated average monthly cost of
utilities and fuels if these are paid by the renter.
Value of property. The survey estimates of value of property are based on the
respondent's estimate of how much the property (house and lot, mobile home and lot, or
condominium unit) would sell for if it were for sale. The information is collected for
HUs that are owned or being bought, and for vacant HUs that are for sale. If the house or
mobile home is owned or being bought, but the land on which it sits is not, the
respondent is asked to estimate the combined value of the house or mobile home and the

6-8

land. For vacant HUs, value is the price asked for the property. This information is
obtained from real estate agents, property managers, or neighbors.
Mortgage status. Mortgage refers to all forms of debt where the property is pledged as
security for repayment of the debt.
Mortgage payment. This item provides the regular monthly amount required to be paid
the lender for the first mortgage on the property.
Detailed Population Information
Detailed population data are collected for all current household members. Some questions are
limited to a subset of the population based on age or other responses. The ACS includes 36
detailed population questions. The place of birth, residence one year ago (migration), and
citizenship questions used in Puerto Rico differ from those used in the United States. The
definitions in this section refer specifically to the United States. This section describes concepts
and definitions for these items. The ACS website includes a complete inventory of terms,
concepts, and definitions for all population items. See
.
Place of birth. Each person is asked whether he or she was born either in the United
States or outside the United States. Those born in the United States, are then asked to
report the name of the state while people born outside the United States are asked to
report the name of the foreign country, or Puerto Rico, Guam, and so forth.
Citizenship. The responses to this question are used to determine the U.S. citizen and
non-U.S. citizen populations and to determine the native and foreign-born populations.
The foreign-born population includes anyone who is not a U.S. citizen at birth. This
includes people who indicate they are a U.S. citizen by naturalization or are not a U.S.
citizen.
Year of entry. All respondents born outside the United States are asked for the year in
which they came to live in the United States. This includes people born in Puerto Rico
and U.S. Island Areas, people born abroad of an American (U.S. citizen) parent or
parents, and the foreign born.
Type of school and school enrollment. People are classified as enrolled in school if
they have attended a “regular” public or private school or college at any time during the
three months prior to the time of interview. The question includes instructions to
“include only nursery or preschool, kindergarten, elementary school, and schooling
which leads to a high school diploma, or a college degree” as regular school or college.
Data are tabulated for people three years and older.

6-9

Educational attainment. Educational attainment data are tabulated for people 18 years
and older. Respondents are classified according to the highest degree or the highest level
of school completed. The question includes instructions for people currently enrolled in
school to report the level of the previous grade attended or the highest degree received.
Ancestry. Ancestry refers to a person’s ethnic origin or descent, “roots,” or heritage, or
the place of birth of the person, or the person's parents’ ancestors before their arrival in
the United States. Some ethnic identities, such as “Egyptian” or “Polish” can be traced to
geographic areas outside the United States, while other ethnicities such as “Pennsylvania
German” or “Cajun” evolved in the United States.
Language spoken at home. A respondent should mark “Yes” if the person sometimes
or always speaks a language other than English at home but should mark “No” if a
language is spoken only at school or if speaking is limited to a few expressions or slang.
The questionnaire instructs respondents to print the name of the non-English language
spoken at home. If the person speaks more than one language other than English, the
person is to report the language spoken most often or, if they cannot determine the one
spoken most often, the language learned first.
Ability to speak English. The data on ability to speak English represent the person’s
own perception of his or her own ability.
Residence one year ago (migration). Residence one year ago is used in conjunction
with location of current residence to determine the extent of residential mobility of the
population and the resulting redistribution of the population across geographic areas of
the country.
Disability. The Census Bureau defines disability as a long-lasting sensory, physical,
mental, or emotional condition. This condition can make it difficult for a person to do
activities such as walking, climbing stairs, dressing, bathing, learning, or remembering.
It can impede a person from being able to go outside the home alone or to work at a job
or business, and it includes people with severe vision or hearing impairments.
Fertility. This question asks if the person had given birth in the past 12 months.
Grandparents as caregivers. Data are collected on whether a grandchild lives with a
grandparent in the household, whether the grandparent has responsibility for the basic
needs of the grandchild, and the duration of that responsibility.
Veteran status. A “civilian veteran” is a person aged 18 years and older who has served
(even for a short time), but is not now serving, on active duty in the U.S. Army, Navy,
Air Force, Marine Corps, or the Coast Guard, or who served in the U.S. Merchant Marine
during World War II. People who served in the National Guard or military reserves are
classified as veterans only if they were ever called or ordered to active duty, not counting
6-10

the four-six months for initial training or yearly summer camps. All other civilians aged
18 years and older are classified as non-veterans.
Work status. People aged 16 and older who worked one or more weeks are classified
as, “worked in the past 12 months.” All other people aged 16 and older are classified as,
“did not work in the past 12 months.”
Place of work. Data on place of work refer to the location (that is street address,
city/county, state) at which workers carried out their occupational activities during the
reference week.
Means of transportation to work. Means of transportation to work refers to the
principal mode of travel or type of conveyance that the worker usually used to get from
home to work during the reference week.
Time leaving home to go to work. The time of day that the respondent usually left
home to go to work during the reference week.
Travel time to work. The total number of minutes that it usually took the worker to get
from home to work during the reference week.
Labor force status. The series of questions on labor force status is designed to identify,
in this sequence: (1) people who worked at any time during the reference week; (2)
people on temporary layoff who were available for work; (3) people who did not work
during the reference week but who had jobs or businesses from which they were
temporarily absent (excluding layoff); (4) people who did not work during the reference
week, but who were looking for work during the last four weeks and were available for
work during the reference week; and (5) people not in the labor force.
Industry, occupation, class of worker. Information on industry relates to the kind of
business conducted by a person's employing organization, while occupation describes the
kind of work the person does on the job. For employed people, the data refer to the
person's job during the previous week. For those who worked two or more jobs, the data
refer to the job where the person worked the greatest number of hours. For unemployed
people, the data refer to their last job. The information on class of worker refers to the
same job as a respondent's industry and occupation and categorizes people according to
the type of ownership of the employing organization.
Income. ‘‘Total income’’ is the sum of the amounts reported separately for wage or
salary income, net self-employment income, interest, dividends, or net rental or royalty
income or income from estates and trusts, Social Security or railroad retirement income,
Supplemental Security Income (SSI), public assistance or welfare payments, retirement,
survivor, or disability pensions, and all other income. The estimates are inflationadjusted using the Consumer Price Index.
6-11

6.5 STRUCTURE OF THE GROUP QUARTERS QUESTIONNAIRES
The 2006 GQ questionnaire includes all of the population items included on the HU
questionnaire except for relationship. One housing question, food stamp benefit, is asked.
Address information is for the GQ facility itself and is collected as part of the automated GQ
Facility Questionnaire (GQFQ). See Chapter 8 for more information on this automated
instrument. The survey information collected from each person selected to be interviewed in GQ
facilities is entered on a separate questionnaire. The number of questionnaires completed for
each GQ facility is the same as the number of people selected, unless a sample person refuses to
participate.

6.6 REFERENCES
U.S. Census Bureau (2006), “American Community Survey 2004 Subject Definitions,”
Washington, DC, 2006,
.

6-12

Chapter 7.
Data Collection and Capture for Housing Units
7.1 OVERVIEW
A key measure of the success of a data collection effort is reflected in the final response rate.
The American Community Survey (ACS) achieves a high total response rate each year, due in
part to the data collection design, which reflects experience and research in data collection
strategies from the Census Bureau’s decennial census and demographic survey programs.
Success, however, would not be possible without the high level of quality associated with the
implementation of this data collection through the efforts of the interviewing staff in the
telephone centers and the regional offices. This success is also related to the fact that survey
participation is mandatory, meaning that households are required by law to respond to the
survey. Title 13 of the United States Code (U.S.C.) authorizes the Census Bureau to conduct the
ACS, requires households to participate, and requires the Census Bureau to keep confidential all
information collected on the ACS.
The data collection operation for housing units (HUs) consists of three modes: mail, telephone,
and personal visit. For most HUs the first phase of data collection includes a questionnaire
mailed to the sample address for the household to complete and return by mail. If no response
by mail is received, the Census Bureau follows up with computer-assisted telephone
interviewing (CATI) if a telephone number is available for the address. If the Census Bureau is
unable to reach an occupant of the unit using CATI, or if the household refuses to participate, the
address may be selected for computer-assisted personal interviewing (CAPI).
The ACS includes 12 independent samples, with a new one introduced each month. Data
collection for each sample lasts for three months, with mail returns accepted during this entire
period, as shown in Figure 7-1. The three-phase data collection process operates in continuously
overlapping cycles, so that during any given month, three samples are in the mail phase, one is in
the CATI phase, and one is in the CAPI phase.

7-1

Figure 7-1. ACS Data Collection Consists of Three Overlapping Phases
ACS Sample
Month

November 2005

December 2005

January 2006

Month of Data Collection
2005

2006

November

December

January

Mail

Mail
Telephone

Mail
Personal
Visit

Mail

Mail
Telephone

Mail
Personal
Visit

Mail

Mail
Telephone

Mail
Personal
Visit

Mail

Mail
Telephone

Mail
Personal
Visit

Mail

Mail
Telephone

February 2006

March 2006

7-2

February

March

April

Figure 7-2 summarizes the distribution of interviews and non-interviews for the January 2005
through June 2005 samples. Among the ACS sample addresses eligible for interviewing in the
United States, approximately 51 percent were interviewed by mail, nine percent were CATI
interviews, and 38 percent were represented by the CAPI interviews. Two percent were noninterviews (U.S. Census Bureau, 2005a). See Figure 7-2.
Figure 7-2. Distribution of ACS Interviews and Non-Interviews1

Non-Interview
2%
CATI
9%
Mail
51%

CAPI
38%

Source: January -- June 2005 Samples

All results are weighted and estimate the proportion of interviews and non-interviews by mode out of the total
eligible sample addresses in the January through June 2005 samples.

7-3

7.2 MAIL PHASE
Using the mail is the least expensive method of data collection, and the success of the program
depends on high levels of response from this collection method. Sample addresses are reviewed
to determine if the available address information is sufficient for mailing. The requirement for a
“mailable” address in the United States is met if the address has either a complete city-style
address or a complete rural route address. A complete city-style address includes a house
number, street name, and ZIP code. The town or city and state fields are not required because
they can be derived from the ZIP code. A complete rural-route address includes a rural-route
number, box number, and ZIP code. About 95 percent of the 2005 sample addresses in the
United States meet the address completeness criteria and were designated as mailable (U.S.
Census Bureau, 2005a).
The requirement for a mailable address differs slightly in Puerto Rico. In addition to the criteria
for mailable addresses in the United States, sample addresses in Puerto Rico must also have an
urbanizacion name to be considered mailable. About 70 percent of the addresses in Puerto Rico
were considered mailable in 2005 (U.S. Census Bureau, 2005a).
Examples of unmailable addresses are descriptions about a HU and its location (for example,
white house with blue shutters, one mile east of intersection of Routes 15 and 29), post office
(PO) box addresses, and addresses that are missing ZIP codes. PO box addresses are considered
unmailable because of the unknown location of the HU that uses the PO box. Addresses that are
missing ZIP codes are also considered unmailable, because they also would be missing the place
name. HU addresses that do not meet one of the completeness criteria are still included in the
sample frame, but they bypass the mail and telephone phases and are only eligible for CAPI.
Mailout
Because a high level of mail response is critical, the mail phase used in the ACS consists of at
least three and as many as four mailings to each sample address, depending on when a return is
received. ACS materials sent to U.S. addresses are printed in English, and Puerto Rico
Community Survey (PRCS) materials sent to Puerto Rico are printed in Spanish. U.S.
respondents can request Spanish mailing packages, and Puerto Rico respondents can request
English mailing packages via telephone questionnaire assistance (TQA). Refer to Chapter 9 for
information about the ACS language assistance program. The address label file that includes all
mailable sample addresses defines the universe for the first three mailings: a pre-notice letter, an
initial mail package, and a reminder postcard. A replacement mail package is sent to sample
addresses from which there is no response within three weeks of mailing the initial mail
package. Details of each are provided below.
Pre-notice letter. The first mailing includes only a pre-notice letter, signed by the
Census Bureau’s director, alerting residents that they will receive the ACS questionnaire
in a few days and encouraging them to return the questionnaire promptly (see Appendix
B.1 for the U.S. version and Appendix B.2 for the Puerto Rico version). The pre-notice
letter is mailed on the Thursday before the last Monday of the month, unless the last
7-4

Monday is one of the last two days of the month. In that case, the entire mailout
schedule is one week earlier. The pre-notice letter is one of two ACS items that is
printed in-house using print-on-demand technology, which merges the letter text and the
sample address from the address label file.
Initial mail package. The next mailing is the initial mailing package. On the front of
the envelope used for this package is a boxed message informing recipients that the ACS
form is enclosed, and in bold, upper-case type, that their response is required by law.
Appendices B.3 (U.S. version) and B.4 (Puerto Rico version) show the front of an
outgoing envelope. This initial mail package is mailed on the last Monday of the month,
or on the previous Monday if the last day of the month is a Monday or a Tuesday. The
first mail package includes a cover letter, the questionnaire, a guide, a brochure, and a
return envelope.
Cover letter. The cover letter is signed by the Census Bureau’s director (see
Appendix B.5 for the U.S. version and Appendix B.6 for the Puerto Rico version.)
The letter reminds householders that they received the pre-notice letter a few days
earlier and encourages them to return the completed questionnaire as soon as
possible. The letter then explains the purpose of the ACS and how the data are
used. Finally, a toll-free telephone number is included for respondents to call if
they have questions or need help completing the questionnaire.
ACS questionnaire. The 2005/2006 questionnaire is a 24-page, two-color
booklet-style form (see Appendix B.7 for the U.S. version and Appendix B.8 for
the Puerto Rico version). The questionnaires are printed on white paper with
black and green ink. The ink (green for the U. S. form, yellow for the Puerto
Rico form) is used in three different levels of shading. The cover of the
questionnaire includes information on how to obtain assistance in English and
Spanish. The questionnaire includes questions about the HU and the people
living in the HU. Refer to Chapter 6 for details. Space is provided to record
detailed information for up to five people. Follow-up by telephone occurs for
households with six or more people.
Guide to the ACS. The guide instructs respondents how to complete the survey.
See Appendix B.9 for the U.S. version and Appendix B.10 for the Puerto Rico
version.
Frequently Asked Questions (FAQs) brochure. The color brochure (available
in both English and Spanish) provides answers to questions that are frequently
asked about the ACS. Examples of the FAQs are “What is the American
Community Survey?”, “Do I have to answer the questions on the American
Community Survey?”, and “Will the Census Bureau keep my information
confidential?” A similar brochure about the PRCS is substituted in packages
mailed to Puerto Rico. See Appendix B.11 for the U.S. version and Appendix
B.12 for the Puerto Rico version.
7-5

Return envelope. The postage-paid envelope is for the respondent to use in
returning the questionnaire to the Census Bureau.
Reminder postcard. The third mailing is a postcard, signed by the director of the
Census Bureau. See Appendix B.13 for the U.S. version and Appendix B.14 for the
Puerto Rico version. The postcard is mailed on a Thursday, three days after the initial
mail package, and reminds respondents to return their questionnaires. The reminder
postcard is the second of two items that are printed in-house, using print-on-demand
technology to merge text and addresses.
Replacement mail package. The last mailing is sent only to those sample addresses
from which the initial questionnaire has not been returned. It is mailed about three and a
half weeks after the initial mail package. The content is the same as the initial mail
package, except that it contains a different cover letter. The final cover letter is signed by
the Director of the Census Bureau, reminds the household of the importance of the ACS,
and asks them to respond soon. See Appendix B.15 for the U.S. version and Appendix
B.16 for the Puerto Rico version.
The Census Bureau’s National Processing Center (NPC) is responsible for assembling and
mailing the packages for addresses selected for mailout. All of the components of the mail
packages described above are printed under contract by outside vendors with the two exceptions
noted earlier. The vendors who print the mailout materials are selected through a competitive
bidding process conducted by the Government Printing Office (GPO) on behalf of the Census
Bureau. As the vendors print the materials, trained quality control staff from the NPC monitor
the work and reject batches of work if they do not meet contractual standards for quality. The
NPC maintains about a six-month inventory of mailout materials in its Jeffersonville, IN facility,
and print jobs are put out for bid about twice each year.
The NPC is responsible for labeling the outgoing mail packages. Several months before each
sample’s mailings, Census Bureau staff provide an address file to the NPC to use in creating
address labels for the first three mailings listed above. An updated address file is provided to the
NPC about three days before the replacement mail package is sent. This file excludes addresses
from which a questionnaire was returned in the first three weeks, which is usually about 25
percent to 30 percent of the sample addresses for the United States, and about 10 percent of the
sample addresses for Puerto Rico (U.S. Census Bureau, 2005b).
Most mail responses are received within five weeks of the date the initial mail package is sent,
but the NPC will continue to accept questionnaires for about three months from the start of each
monthly sample. After a specified cutoff date for each monthly sample, late mail returns will not
be included in the data set.

7-6

Check In
The United States Postal Service (USPS) returns all completed ACS questionnaires to the NPC.
The check-in unit receives mail deliveries two or three times each business day. Each
questionnaire contains a unique bar code in the address label area, which shows through a
window on the front of the return envelope. The mail returns are sent through a laser sorter
where the bar code is initially scanned. The bar code permits forms to be sorted according to
monthly sample and within each monthly sample, by whether they are United States or Puerto
Rico addresses. During this step, the return envelopes are mechanically slit for clerks to remove
the contents.
After clerks remove the forms from the return envelopes, the forms are taken to a unit where
clerks look at each page of every returned questionnaire. The clerks look for enclosed
correspondence, which they forward to headquarters, if necessary. The clerks then scan the bar
code on each questionnaire to officially check in the form, and organize the forms into batches of
50 questionnaires. The check-in staff have three days to check in a form. Usually the staff can
check in all the forms they receive within one day, but during particularly high volume days,
sometimes that is not possible. Each day, NPC staff transmit a file of the checked-in cases and
headquarters staff update the status of each checked-in case on the control file. See Chapter 10
for information about the control file.
Some of the mailed forms are returned to the NPC as undeliverable as addressed (UAA) by the
USPS. In 2005 UAA returns accounted for 12 percent to 13 percent of mailed questionnaires in
both the United States and in Puerto Rico (U.S. Census Bureau, 2005a). UAAs occur for many
reasons, including a bad or unknown address, an address corresponds to a vacant HU, or a
resident’s refusal to accept mail delivery. Sample addresses which are UAAs for the initial mail
package remain eligible for the replacement mail package because often the delivery process for
an address is successful on this second attempt without any change to the address. UAAs are
eligible for the CATI and CAPI operations.
Telephone Questionnaire Assistance (TQA)
TQA is a toll-free, interactive voice recognition (IVR) telephone system that respondents can
call if they have questions about completing the questionnaire, or to request a questionnaire in
the other language. The telephone number for TQA is listed on the questionnaire itself, as well
as on all of the letters, brochures, and postcards. Alternate TQA numbers are listed on the
questionnaire for Spanish speakers and for telephone device for the deaf (TDD) use.
When respondents call TQA, they enter the IVR system, which provides some basic information
on the ACS and directions on using the IVR. Respondents may obtain recorded answers to
frequently asked questions, or, during business hours, they can choose to speak directly to an
agent. Respondents can furnish their ACS identification number from the questionnaire (or one
of the other mailing pieces), which allows them to hear a customized message about the current
status of their questionnaire. The IVR can indicate whether the NPC has received a
questionnaire for the sample address, and, if not, state that an ACS interviewer may call or visit
7-7

the sample address. If a respondent chooses to speak directly to an agent, the agent answers the
caller’s questions and encourages the respondent to complete the questionnaire over the
telephone. Agents use an automated survey instrument to capture the respondent’s answers.
In 2005, a household member called the toll-free number for assistance for approximately six
percent of the mailable addresses (U.S. Census Bureau, 2005c). For less than one percent of the
mailable addresses in 2005 (U.S. Census Bureau, 2005a), a household member agreed to
complete the survey over the telephone. All calls to TQA are logged, and the system can record
up to five reasons for each call. Even though TQA interviews are conducted by telephone, they
are considered mail responses, not CATI responses, because the call was initiated by the sample
household upon receiving the questionnaire in the mail.
Data Capture
After the questionnaires have been checked in and batched into groups of 50, they move to the
data entry, or keying, unit in the NPC. The keying unit has a goal of keying the responses from
the questionnaires within three weeks of receipt, but the NPC usually keys the responses within
two weeks. Data keyers enter the information from the forms directly into a data capture file.
The NPC’s data keying operation uses stringent quality assurance procedures to ensure that nonsampling errors from keying data are minimized.
Data keyers progress through three levels of quality assurance verification. When new keyers
begin data entry for ACS questionnaires, they are considered to be in a training stage and 100
percent of their work is checked to determine whether they keyed the data correctly. An
experienced keyer independently re-keys the same batch of 50 questionnaires entered by the new
keyer, and the work of the two keyers is compared to check for keying errors. A keying error is
defined as an incorrectly keyed data item. If the new keyer’s error rate (the percentage of all
keyed data items that are in error) in one of the first two batches of questionnaires is equal to or
less than 1.5 percent, the keyer is moved to the pre-qualified stage. If the keyer’s error rate in
both of the first two batches of 50 questionnaires is above 1.5 percent, the keyer is immediately
retrained and then is eligible to advance to the pre-qualified stage. These keyers are still subject
to 100-percent verification. Once pre-qualified keyers key a batch at an error rate equal to or
less than 1.5 percent, they are moved to the qualified stage. If they exceed the error rate of 1.5
percent, keyers are provided feedback on their errors immediately. A supervisor eventually
decides whether to move keyers to the qualified stage. Supervisors verify a sample of each
qualified keyer’s work, with an acceptable error rate of 1.5 percent or less. Keyers at all levels
are subject to removal from the project and administrative action if they fail to maintain an error
rate of less than 0.80 percent, but most keyers have a rate far below that level.
Each day, NPC staff transmit a file with the data from the keyed questionnaires, and
headquarters staff update the status of each keyed case on the control file. See Chapter 10 for
more information on data processing operations.
The Census Bureau is currently exploring the possibility of moving to a key-from-image (KFI)
data capture system, which involves imaging the questionnaire, interpreting the check box
7-8

entries with optical mark recognition (OMR), and keying write-in responses from the images
using a computerized system. The advantages of KFI include reduced costs and potential for
increased data-capture accuracy.
Failed Edit Follow-up
After the data on the mail-returned questionnaires are keyed, the data files are processed in
batches through a computerized edit for coverage consistency and content completeness. The
purpose of the edit is to identify cases requiring additional information. Cases that fail the edit
are eligible for the telephone failed-edit follow-up (FEFU) operation, and become part of the
FEFU workload if we have a telephone number for the sample address. This operation is
designed to improve the final quality of mail-returned questionnaires.
Cases can fail the edit for two broad reasons: coverage failures and content failures. Coverage
failures can take two forms. First, since the ACS questionnaire is designed to accommodate
detailed answers for households with five or fewer people, a case will fail the coverage edit
when a respondent indicates that there are more than five people living in the household. Also, a
case will fail the coverage edit if the reported number of people differs from the number of
people for whom responses are provided. Content failures occur if the edit determines that two
or more critical items have not been answered, or if a specific number of other required items
have not been answered.
Approximately 33 percent of the keyed mail-return questionnaires in 2005 (U.S. Census Bureau,
2005b) failed either the coverage or the content edit and required FEFU. A new set of FEFU
cases is generated each business day with work constantly flowing into the FEFU operation.
Staff in the Census Bureau’s three telephone centers call respondents and attempt to obtain the
missing data. The interview period for each FEFU case is three weeks.

7.3 TELEPHONE PHASE
The second of the three data collection phases is the telephone phase, or CATI. The automated
data collection instrument (the set of questions, the list of response categories, and the logic that
presents the next appropriate question based on the response to a given question) is written in an
open-source scripting software language called Blaise. The CATI instrument is available in
English and Spanish in both the United States and Puerto Rico. Chapter 9 provides more
information on the ACS language assistance program.
To be eligible for CATI, a HU that did not respond by mail must have a mailable address and a
telephone number. The Census Bureau contracts with vendors who attempt to match the
Bureau’s sample addresses to their database of addresses and provide telephone numbers. There
are two vendors for United States addresses and one for Puerto Rico addresses. Since the
vendors use different methodologies and sources, one of the vendors may be able to provide a
telephone number while the other one may not. This matching operation occurs each month
before a sample is mailed. About a month later, just prior to the monthly CATI work,
7-9

headquarters staff transmit a file of the CATI-eligible sample addresses and the corresponding
telephone numbers to a common queue for all three telephone call centers.
The Census Bureau conducts CATI out of its three telephone call centers located in
Jeffersonville, IN; Hagerstown, MD; and Tucson, AZ. The CATI operation begins about five
weeks after the first mail package was sent out. A control system, called WebCATI, is used to
assign the cases to individual telephone interviewers. As CATI interviewers begin contacting
the households, the WebCATI system evaluates the skills needed for each case (for example, a
language skill or a refusal conversion skill) and delivers the case to only those interviewers who
possess the matching skill.
If a CATI interviewer reaches a person, the first task is to verify that the interviewer has
contacted the correct address. If so, the interviewer attempts to complete the interview. If the
householder refuses to participate in the CATI interview, a different CATI interviewer specially
trained in dealing with refusals will call the household after a few days to attempt to complete
the interview. If the household again refuses, CATI contact attempts are stopped, and the case is
coded as a non-interview. If at any time during the CATI operation a household’s questionnaire
is received, that case is removed from the CATI sample and the case is considered a mail
response.
The CATI operation has a strong quality assurance program, including CATI software-related
quality assurance and monitoring of telephone interviewers. The CATI instrument has a
sophisticated, integrated set of checks to prevent common errors. For example, a telephone
interviewer cannot input out-of-range responses, skip questions that should have been asked, or
ask questions that should have been skipped. New telephone interviewers, in particular, but also
experienced interviewers, are subject to random monitoring by supervisors to ensure that they
follow procedures for asking questions and effectively probe for answers. In addition, they are
monitored to ensure that the answer they enter into the computerized instrument matches the
answer provided by the respondent.
Approximately 650 interviewers conduct CATI interviews from the Census Bureau’s three
telephone call centers. Interviewers participate in a three-day classroom training session to learn
and practice the appropriate interviewing procedures. Interviewers have 25-26 calendar days to
complete the monthly CATI caseload, which averages about 85,000 cases each month. At the
end of the CATI interview cycle, all cases receive a CATI outcome code in one of three general
categories: interview; non-interview; or ineligible for CATI. This final category includes cases
with an incorrect telephone number. Cases in the last two categories are eligible for the personal
visit phase.
Each day, NPC staff transmit a file with the status of each case in the telephone mode, and
headquarters staff update the status of each case on the control file. See Chapter 10 for more
information on data processing operations.

7-10

7.4 PERSONAL VISIT PHASE
The last phase of ACS data collection is the personal visit phase, or CAPI. This phase usually
begins on the first day of the third month of data collection for each sample, and typically lasts
the entire month.
A sub-sample of the remaining non-response addresses is selected for CAPI. After mail and
CATI operations have been completed, the CAPI sub-sample is selected from two categories of
cases. Mailable addresses with neither a response to the mailout nor a telephone interview are
sampled at a rate of one-in-two, two-in-five, or one-in-three. Unmailable addresses are sampled
at a rate of two-in-three. See Table 4-3 and Chapter 4 for more information about CAPI subsampling.
The CAPI operation is conducted by Census Bureau field representatives (FRs) operating from
the Census Bureau’s 12 regional offices (ROs). The sampled cases are distributed among the 12
ROs based on the RO’s geographic boundaries. The Boston RO is responsible for CAPI data
collection in Puerto Rico.
After the databases containing the sample addresses to interview each month are loaded in the
ROs, the addresses are assigned to FRs. FRs can conduct interviews by telephone or personal
visit, using laptop PCs loaded with a survey instrument similar to the one used in the CATI
operation, written in Blaise. The CAPI instrument is available in English and Spanish in the
United States and Puerto Rico. See Chapter 9 for more information on language assistance
provided during CAPI.
If a telephone number is available, the FR will first attempt to contact the sample address by
telephone. There are two exceptions to this procedure. Unmailable addresses are not contacted
initially by telephone because an FR would not be able to verify the location of the address over
the telephone. Cases which were refusals during the CATI phase are also never contacted by
telephone because the residents at the sample address have already refused a telephone
interview. The FR will call the telephone number and confirm that he or she has reached the
sample address. If so, the FR uses the automated instrument on the laptop PC and attempts to
conduct the interview. If an FR cannot reach a resident of the sample address after calling three
to five times during the first few days of the interview period, at different times of the day, he or
she must make a personal visit to the address.
Approximately 80 percent of CAPI cases require that the FR visit the sample address. A
personal visit is needed to determine whether the HU exists and to determine the HU’s
occupancy status. If a HU does not exist at the sample address, that status is documented. If an
FR verifies that a HU is vacant, he or she will interview a knowledgeable respondent, such as the
owner, building manager, real estate agent, or as a last resort, a neighbor, and conduct a “vacant
interview” for the HU to obtain some basic information about the HU. If the HU is currently
occupied, the FR will conduct an “occupied” or “temporarily occupied” interview. An FR
conducts a temporarily occupied interview when there are residents living in HU at the time of
the FR’s visit, but no resident is living there for more than two months.
7-11

The FRs are trained to remain polite while being persistent and tenacious when attempting to
obtain responses. The FRs are also trained in how to handle almost any situation, ranging from
how to respond to a household that claims to have returned their questionnaire by mail to how to
handle an interview with a non-English speaking respondent.
When FRs are not able to obtain interviews, they must indicate the reason the interview was not
possible. Such non-interviews are taken seriously because they impact both sampling and nonsampling error. Non-interviews will occur when an eligible respondent (someone who lives at
the sample address) cannot be located, is unavailable, or is unwilling to provide the survey
information. Additional non-interviews occur when FRs are unable to confirm the status of a
sample HU due to restricted access to an area because of a natural disaster during the interview
period. Examples include floods, earthquakes, tornadoes, and fires. Note that initially, FRs
may be unable to gain access to a gated community, but they need to find appropriate methods to
obtain access and attempt to interview at the sample address. Some sample cases will be
determined to be ineligible for the survey. They include sample addresses identifying structures
that are under construction, structures that have been demolished, and non-existent addresses.
One of the important tasks for an FR is to check the geographic codes (state, county, tract, and
block) for each address to which he or she makes a personal visit. The FR confirms that the
codes are correct, corrects the codes, or records the codes if they are missing.
Each day, FRs transmit a file with the status of all their cases that are in the personal visit mode,
and headquarters staff update the status of each case on the control file. See Chapter 10 for more
information on data processing operations.
Approximately 3,500 FRs conduct CAPI interviews across the United States and Puerto Rico.
FRs participate in a four-day classroom training session to learn and practice the appropriate
interviewing procedures. Interviewers have almost the entire month to complete the monthly
CAPI caseload, which averages about 40,000 cases each month. At the end of the CAPI
interview cycle, all cases receive a CAPI interview or non-interview outcome code.
Supervisors travel with FRs during their first few assignments of work to observe their work and
to reinforce procedures learned in the classroom training sessions. In addition, a sample of FRs
is selected each month and supervisors reinterview a sample of those FRs’ cases. The primary
purpose of the reinterview program is to verify that FRs are conducting interviews for their
assigned cases and doing so correctly.
Data Collection in Remote Alaska
FRs conduct data collection at sample addresses in areas of Alaska designated as remote during
two separate periods each survey year, January through April and September through December.
This procedure is used because of the difficulties accessing these areas and the costs associated
with the data collection in these areas.

7-12

7.5 REFERENCES
Dillman, Don A., Clark, Jon R., and Treat, James B. (1994), “Influence of 13 Design Factors on
Completion Rates to Decennial Census Questionnaires,” Paper presented at the 1994 Annual
Research Conference of the U.S. Bureau of the Census, Arlington, Virginia, 1994.
Griffin, Deborah H. (2004), “The American Community Survey: Developing a Continuous
Measurement Application in the United States,” Paper presented at a Workshop on Alternative
Approaches to Population and Housing Censuses at the Instituto Brasileiro de Geografia e
Estatistica, Rio de Janeiro, Brazil, 2004.
McGovern, Pamela, Griffin, Deborah, and McGinn, Larry (2003), “Language Action Plan for the
American Community Survey,” Meetings of the Census Advisory Committee on the African
American Population, the American Indian and Alaska Native Populations, the Asian population,
the Hispanic Population, and the Native Hawaiian and Other Pacific Islander Populations, May
5-7, 2003.
U.S. Census Bureau (2001), “Meeting 21st Century Demographic Data Needs--Implementing the
American Community Survey: July 2001, Report 1: Demonstrating Operational Feasibility,”
Washington, DC, 2001.
U.S. Census Bureau (2002), “Meeting 21st Century Demographic Data Needs--Implementing the
American Community Survey: May 2002, Report 2: Demonstrating Survey Quality,”
Washington, DC, 2002.
U.S. Census Bureau (2004a), “Meeting 21st Century Demographic Data Needs--Implementing
the American Community Survey, Report 6: The 2001-2002 Operational Feasibility Report of
the American Community Survey,” Washington, DC, 2004.
U.S. Census Bureau (2004b), “Meeting 21st Century Demographic Data Needs--Implementing
the American Community Survey: Report 7: Comparing Quality Measures: The American
Community Survey’s Three-Year Averages and Census 2000's Long Form Sample Estimates,”
Washington, DC, June 2004.
U.S. Census Bureau (2004c), “Quality Assurance (QA) for DEC Data Entry Operations,”
Internal Census Bureau specification, Washington, DC, 2004.
U.S. Census Bureau (2005a), Internal ACSO management reports on data collection production,
Washington, DC, 2005.
U.S. Census Bureau (2005b), Internal production reports, Washington, DC, 2005.
U.S. Census Bureau (2005c), Internal telephone center management reports, Washington, DC,
2005.
7-13

Chapter 8.
Data Collection and Capture for Group Quarters
8.1 OVERVIEW
All living quarters are classified as either housing units (HUs) or group quarters (GQ) facilities.
An HU is a house, an apartment, a mobile home, a group of rooms or a single room occupied as
a separate living quarters or intended for occupancy as separate living quarters. Separate living
quarters are those in which the occupants live separately from any other people in the building
and which have direct access from outside the building or through a common hall.
GQ facilities are places where people live or stay that is normally owned or managed by an
entity or organization providing housing and/or services for the residents. These services may
include custodial or medical care as well as other types of assistance, and residency is commonly
restricted to those receiving these services. People living in GQ facilities are usually not related
to each other. GQ facilities include such places as college residence halls, residential treatment
centers, skilled nursing facilities, group homes, military barracks, correctional facilities,
workers’ dormitories, and facilities for people experiencing homelessness. GQ facilities are
defined according to the housing and/or services provided to residents and are identified by
Census GQ type codes. See Appendix C.1 for a copy of the GQ type codes and definitions.
In January 2006, the American Community Survey (ACS) expanded to include GQ facilities.
The ACS GQ sample includes 12 independent samples, like the HU sample, a new sample is
introduced each month. Data collection for each sample lasts for six weeks. The GQ data
collection operation is conducted in two phases. Census Bureau field representatives (FRs)
conduct interviews with a GQ contact or administrator of a selected GQ facility (GQ level) and
then conduct interviews with a sample of individuals from the selected GQ facility (person
level).
The GQ-level data collection instrument is an automated Group Quarters Facility Questionnaire
(GQFQ). The FRs will conduct GQ-level data collection at approximately 20,000 individual GQ
facilities each year. Information collected by the FR during the GQ-level interview determines
the type of facility the selected GQ facility represents, and the sample of individuals to be
interviewed.
An FR collects the GQ survey information from the sampled residents using a paper, bilingual
(English/Spanish) questionnaire during the person-level phase. The GQ questionnaire records
detailed information for one person. FRs will collect data from approximately 195,000 GQ
sample residents each year. All of the methods described in this chapter apply to the ACS GQ
operation in both the United States and Puerto Rico.

8-1

8.2 GROUP QUARTERS-LEVEL PHASE
Data collection is completed, primarily, through FR interviews at sample GQ facilities. The FRs
may also obtain the facility information by conducting a telephone interview with the GQ
contact. FRs collect data from approximately 20,000 GQ facilities each year. Each FR is
assigned approximately five sample GQ facilities each month. GQ-level interviews are
conducted for a period of six-weeks. The GQ-level interviews determine whether the FR will
sample all, some, or none of the residents at a sampled facility for person-level interviews. The
FR verifies the sample GQ information and can record up to two additional GQ types identified
at a sample GQ address in the GQFQ instrument. The GQFQ is programmed to determine the
appropriate GQ population to sample when more than one GQ type is identified at one sample
GQ address. The GQFQ assigns the correct GQ type code(s) based on GQ contact responses to
the GQFQ questions. The information obtained from GQ-level interviews is transmitted through
a secure file transfer to Census Bureau headquarters on a nightly basis.
Pre-Visit Mailings. The GQ-level data collection phase begins when an introductory letter is
mailed from the Census Bureau’s National Processing Center (NPC) to a monthly sample of GQ
facilities. This mailing includes a GQ FAQ brochure. A separate letter is mailed to sampled
state and local correctional facilities as an effort to obtain contact and security information prior
to making telephone contact with the GQ facility.
GQ Introductory Letter. NPC will mail an introductory letter that contains Regional
Office (RO) contact information to the sample GQ facility approximately two weeks
before the FRs begin each monthly GQ assignment. The letter explains that the FR will
visit the facility to conduct both GQ-level and person-level data collection. It also
describes the information that is asked for on the questionnaire, uses of the data, and the
Internet address where they can find more information about the ACS. This letter is
printed in-house using print-on-demand technology, which merges the letter text and the
sample GQ name and address and specific Regional Office contact information. There
are 12 RO-specific letters generated for each sample month. See Appendix C.2 for a
copy of this letter.
GQ Frequently Asked Questions (FAQ) Brochures. The color, tri-fold brochure
contains FAQs (see Appendix C.3) about the ACS and GQ facilities and is mailed to
sample GQ facilities with the GQ introductory letter. Examples of the FAQs are “What
is the American Community Survey?” “Do I have to answer the questions on the
American Community Survey?” and “Will the Census Bureau keep my information
confidential?” A similar Puerto Rico brochure (see Appendix C.4) is included with the
GQ introductory letter mailed to Puerto Rico sample GQ facilities, and a brochure that
specifically addresses the interests of populations in Remote Alaska (see Appendix C.5)
is mailed with introductory letter to those sample GQ facilities.
8-2

GQ State and Local Correctional Facilities Letter. FRs may mail another letter to
selected correctional facilities after the GQ introductory letter, described above, is sent
but before calling to schedule an appointment to visit the sample GQ facility. This letter
was developed during GQ pretesting and is designed to assist FRs with gaining access to
state and local correctional facilities. The GQ operation does not require FRs to send this
letter to all state and local correctional facilities. The letter asks for the name and title of
a person with the authority to schedule FR visits and to coordinate the GQ data
collection. It also provides information about the ACS, the dual nature of the FR visit to
the facility, and includes a form to return to the RO with the contact name, title, and
phone number of a designated GQ contact. See Appendix C.6 for a copy of this letter.
Group Quarters Facility Questionnaire - Initial GQ Contact
The GQ FRs conduct GQ-level interviews for the assigned GQ facility prior to selecting the
individual person-level sample and collecting the GQ person-level data. The FR will always try
to make initial contact with the GQ facility by telephone. If successful in reaching the GQ
contact (the GQ contact is usually the facility administrator), the FR will collect information
about the GQ facility and schedule an appointment to visit the GQ facility and complete the GQlevel data collection phase.
Using a laptop computer, the FR conducts an interview with the sample GQ contact person to
complete an automated GQFQ. The GQFQ instrument is available to FRs in both English and
Spanish languages. Information collected by the FR during the GQ-level interview verifies and
assigns the sample GQ type code. This type code identifies the population that will be included
in the person-level phase.
The GQFQ is initially used by the FRs to verify or update the name and the address of the GQ
facility and to set up an appointment to visit the GQ facility to select the person-level sample and
conduct interviews. If the GQ contact refuses to schedule an appointment for a visit, the FR
notifies the RO and another attempt at gaining the GQ contact’s cooperation is made. If the RO
is unsuccessful at scheduling an appointment, the FR goes to the GQ facility without an
appointment and attempts to talk to the GQ contact or administrator to obtain cooperation and
the information needed to generate the sample of residents and conduct the person-level
interviews. If the GQ contact again refuses, the RO/FR will attempt to explain the mandatory
nature of the survey, what the FR is attempting to do at the facility, and why. The FR or RO will
also attempt to determine or verify the GQ type code during this contact.
Group Quarters Facility Questionnaire - GQ Visit
Once a visit has been successfully scheduled and the FR arrives at the GQ facility, the FR will
update or verify the Special Place (SP) and GQ name, mailing and physical address, facility
telephone number, contact name(s), and telephone number(s). Using a flashcard (see Appendix
C.7), the FR asks the GQ administrator to indicate which type of facility best describes the GQ
8-3

facility. The GQ contact is asked to identify up to three different GQ type codes at one sample
address.
The FR will generate a person-level sample from all, some, or none of the residents at the facility
depending on whether the GQ address can now be classified as the Census 2000 type code,
multiple type codes, or a type code that is out of scope for data collection. When multiple type
codes are verified and assigned to the address, only those people in the sampled GQ type code
are included in the universe for person-level sampling. The FR records any other GQ type codes
that are identified at the sample GQ address and the address information is updated for future
ACS GQ sample selection. If none of the GQ type codes are the same as the sampled GQ type
code, the type code that identifies the largest population, and is in-scope for ACS GQ data
collection, is used for determining the population for person-level sampling at the sample GQ
facility.
After determining that the GQ facility is in scope for GQ data collection, the FR asks for a list or
register of names and/or bed locations for everyone that is living or staying at the sample GQ
facility on the day of the visit. This register is used to generate the sample of residents to be
interviewed. If a register is not available, the FR will create one using the GQ listing sheet. The
listing sheet contains preprinted GQ contact and facility address information. See Appendix C.8
for a copy of the listing sheet.
The FR uses the sampling component of the GQFQ instrument to verify the register provided to
them by the GQ contact person. The instrument will proceed automatically to the beginning of
the sampling component after the FR has entered all the required facility information and the GQ
contact person verifies that there are people living or staying there at the time of the GQ visit. If
there are no residents living or staying at the GQ facility at the time of the visit, the FR will
complete the GQ-level interview to update the GQ information and the GQ type code will be
determined. No person-level data collection follows.
The person-level sample of GQ residents is generated from the GQFQ instrument through a
systematic sample selection. See Section 8.3 for information about data collection from
individuals. The FR matches the line numbers generated by the instrument for the person sample
to the register of current residents. A grid up to 15 lines long appears on the GQFQ laptop
screen with a place for name and sample person location description, the line number
corresponding to the register, a telephone number, a telephone extension, and a GQ control
number (assigned by the GQFQ sampling program). To complete the sampling process, the FR
enters information that specifically identifies the location of each sample person into the GQFQ.
The FR must select an interim or final outcome to record the status of the GQ-level interview.
The reason for a GQ refusal or non-interview is specified, when applicable. The FR can enter an
interim GQ-level interview status reason to allow them to close the case and reenter it at a later
time. The FR will select the appropriate reason for exiting the interview from a list provided in
8-4

the GQFQ. When the FR does this, an outcome code will be assigned by the GQFQ that reflects
the current status of the interview.
There are several reasons why the GQ-level data collection may not be completed. These noninterviews occur when the FR is unable to locate the facility, finds there are no residents living
or staying at the sample GQ facility during the data collection period, the FR determines there
are no living quarters at the sample GQ facility, or when the sample GQ facility no longer exists.
The FRs will ask the GQ contact one reinterview question in the GQ-level GQFQ interview
when GQ production begins. The full series of questions for the GQ Reinterview (RI) has been
developed, programmed, and tested. GQ RI implementation began in April 2006. The purpose
of the reinterview is to detect and deter falsification at the GQ-level.
All information collected during the GQ-level phase is transmitted through secure, electronic file
transfer to the Census Bureau on a nightly basis from each FR.

8.3 PERSON-LEVEL PHASE
This section describes the activities needed to conduct person-level interviews at sample GQ
facilities. During the person-level data collection phase, the FR will collect data for 10 to 15
sample residents at each assigned GQ facility. The FR will prepare person-level survey
packages from GQ-level survey packages assembled at NPC, interview sample residents or
distribute survey packages to them, review and edit completed questionnaires, and assign a final
outcome code to all questionnaires and GQ assignments.
Preparation
The Census Bureau’s NPC is responsible for assembling GQ survey packages and delivering
them to the ROs two weeks before the start of each survey month. Most of the GQ materials are
printed under contract by outside vendors; however, due to the smaller scale of the GQ data
collection, some forms that are only needed at the GQ level are printed in-house. The vendors
who print the survey materials are selected through a competitive bidding process conducted by
the Government Printing Office (GPO) on behalf of the Census Bureau. Trained quality control
staffs from NPC monitor the work as the contractors print the materials. NPC rejects batches of
work if they do not meet contractual quality standards. NPC maintains about a six-month
inventory of ACS GQ survey materials in its Jeffersonville, IN facility, and print jobs are put out
for bid about twice a year.
The NPC is also responsible for DocuPrinting and/or addressing the GQ introductory letters,
Survey Package Control List for Special Sworn Status (SSS) individuals, instruction manual for
SSS individuals, listing sheets, and FR folder labels. Contractors print all the questionnaires, the
questionnaire information guide booklet, brochures, information card booklet, and Privacy Act
8-5

notices. The geocoding forms and geocoding packages are printed at the Census Bureau. NPC
also labels ACS GQ sample questionnaires with address and control numbers.
The Census Bureau provides label/address files to NPC for DocuPrinted materials on a monthly
basis. NPC receives the files approximately eight weeks prior to the sample months when they
are needed for data collection. On each FR assignment folder, NPC pre-prints a label containing
the SP name, GQ name, GQ address, state, city, county, tract-block, RO name, and GQ type
code. Each of 10 to15 personal interview survey packages included in the assignment folder
contain a GQ questionnaire (pre-printed with the previously described folder label information),
questionnaire instruction guide, an unlabeled GQ introductory letter, a return envelope, and a
supply of FAQ brochures and Privacy Act notices. Other materials the FR may need, such as the
SSS form and the instruction manual for SSS individuals is provided to the FRs by the ROs, as
needed.
The FR prepares the number of survey packages needed to interview the sample residents. There
are 10 sample residents selected at large sample GQ facilities. All residents will be interviewed
at GQ facilities identified as small GQ facilities. The FRs will use the register information from
the GQFQ to prepare the survey packages needed for person-level interviews. The GQFQ also
generates a questionnaire control number to track the questionnaires from the beginning of the
person-level phase through keying. The GQ questionnaire contains blank lines below the preprinted GQ address where the FR will manually record specific information to locate the sample
residents such as: name and floor, wing, room, and/or bed location. This information helps the
FR organize the order of personal interviews to efficiently conduct the person-level interviews,
and also enables another FR to locate the sampled residents at the GQ facility if a case is
reassigned.
Person-level Survey Materials
This section provides details about the materials needed to conduct ACS GQ person-level
interviews.
Introductory Letter for the Sample Resident. The FR will give each sampled person
an introductory letter (see Appendix C.9) at the time of the person-level interview. It
provides information about the ACS and describing why it is important that they
complete the GQ questionnaire, describes uses of ACS data, stresses the confidentiality
of their individual responses, and also includes the Internet address for the ACS website.

ACS GQ Questionnaire. The FR uses a paper GQ questionnaire for person-level data
collection. This questionnaire is a bilingual, 12-page, four-color, flip style booklet. Six
blue pages make up the English language GQ questionnaire (see Appendix C.10), and
when flipped over, six green pages make up the Spanish language GQ questionnaire.
Each page measures 10-1/4" wide by 10-1/2" high, the same dimensions as the ACS
8-6

questionnaire. The GQ questionnaire is designed to record detailed population
information for one person. The GQ questionnaire does not include housing questions
except for the food stamp benefit question. When a questionnaire is damaged or missing,
the FR uses case management assignment software to obtain the control number, SP/GQ
name, and address information. The FR transcribes this information into the label area of
a blank questionnaire and uses this new copy for the data collection. A PRCS GQ
bilingual questionnaire is used for person-level data collection in Puerto Rico. See
Appendix C.11.
GQ Questionnaire Instruction Guide. The FR will provide a copy of the questionnaire
Instruction Guide booklet (see Appendix C.12) to sample residents when a personal
interview cannot be conducted and the resident is completing the questionnaire
him/herself. This guide provides respondents with detailed information about how to
complete the GQ questionnaire. It explains each question, with expanded instructions
and examples, and instructs the respondent on how to mark the check boxes and record
write-in responses.
GQ Question and Answer Brochure. When beginning person-level data collection, the
FR will have a supply of brochures to give sample residents. This brochure provides
answers to questions about the ACS GQ program. See Appendix C.13 for a copy of this
brochure.
GQ Return Envelopes. The envelopes (see Appendix C.14) used to return completed
questionnaires to the FR or GQ contact are not designed for delivery through the U.S.
Postal Service. The FR personally collects and reviews all completed questionnaires.
Completing the GQ Questionnaire
There are several ways for an FR to obtain completed GQ questionnaires. The preferred method
is for the FR to fill out the questionnaire in a face-to-face interview with the sample resident.
However, other data collection methods may be necessary for the FR to obtain a completed GQ
questionnaire. The FR may fill out the questionnaire during a telephone interview with the
sample resident; conduct a face-to-face proxy interview with a relative, guardian or GQ contact;
leave the questionnaire with the sample resident to complete; or leave the questionnaire(s) with
the GQ contact to distribute the questionnaires to sampled residents and collect them when
completed. If the questionnaires are left with the residents to complete, the FR will make
arrangements with the sample resident, or GQ contact, to return and pick up the completed
questionnaire(s) within two days. The FR must be certain that the sample resident is physically
and mentally able to understand and complete the questionnaire him/herself.
Before a GQ contact or a GQ employee has access to the names of the sample residents and the
sample residents’ answers to the GQ questionnaire, they must be sworn in by the FR as an SSS
individual. Generally, an SSS individual is needed when the sample person is not physically or
8-7

mentally able to answer the questions. An FR must swear in a person under Title 13, United
States Code (U.S.C.) whether it is a social worker, administrator or GQ employee, any time that
person is going to see a sampled resident’s responses to the GQ questionnaire. In taking the
Oath of Nondisclosure, SSS individuals agree to abide by the same rules that apply to other
Census Bureau employees regarding safeguarding of Title 13 respondent information and other
protected materials and acknowledge that they are subject to the same penalties for unauthorized
disclosure. Legal guardians do not need to be sworn in as SSS individuals. If the sample person
gives a GQ employee permission to answer questions or help to answer questions on their behalf,
the GQ employee does not need to be sworn in.
Questionnaire Review
After data collection has been completed for each sample resident, the FR will conduct two
separate edit reviews: one review of the person-level questionnaires and a second review of all
questionnaires within a GQ-level assignment. The first review is a manual edit check of the
responses that are recorded on the questionnaire. The FR verifies that all responses are legible
and that the write-in entries and check boxes contain appropriate responses according to the skip
patterns on the questionnaire.
The FR will determine whether a person-level interview is complete, a sufficient partial, or
incomplete based on certain criteria. An interview is complete when all or most of the questions
have been answered. An interview is a sufficient partial interview when enough questions are
answered to define the basic characteristics of the sample person. A case is classified as a noninterview when answers were not obtained to meet the criteria of a complete or a sufficient
partial interview. The FR will verify that the correct outcome code has been assigned to each
questionnaire. The FR records the status of the questionnaire review as an interim or final
outcome code by marking the appropriate check box on the last page of the questionnaire.
The FR will conduct a GQ-level assignment review after completing the questionnaire review.
This review is necessary to ensure that all questionnaires within each GQ assignment are
accurately coded and accounted for. The FR determines if all questionnaires for the GQ facility
have been completed or if a return visit will be necessary. The FR marks any unused
questionnaires with an “X” and writes the word “UNUSED” on the front. The FR will ship the
unused and completed questionnaires to their RO on a flow basis throughout each six-week data
collection period. The ROs all conduct a final review of the questionnaires prior to sending
completed questionnaires to NPC for keying.

8.4 CHECK IN AND DATA CAPTURE
The RO will check in all questionnaires returned by the FRs. Based on the final outcome code
recorded for each questionnaire, the RO separates any blank questionnaires from those with data.
Only questionnaires that contain data, identified by the outcome code assignment, are shipped to
8-8

NPC for check in and keying on Tuesday of each week. The forms are sorted according to the
sample month and by whether they were interviewed in the United States or Puerto Rico.
Check In
The NPC check-in staff is given three days to check in a form. Usually the staff can check in all
the forms they receive within one day, but during particularly high volume days, sometimes that
is not possible. The check-in process results in batches of 50 questionnaires for data capture.
The NPC will continue to accept completed questionnaires, shipped from the RO on a weekly
basis for a period of six weeks from the start of the sample month. Each RO will close out the
sample month GQ assignments, account for all questionnaires, and send the remaining
completed questionnaires to NPC on the last day of the six-week data collection period. The
NPC will complete the sample month check in within seven days of receipt of the final shipment
from each RO. Each questionnaire contains a unique bar code in the address label area. The bar
code is scanned and permits forms to be sorted according to monthly sample panel and within
each panel, by whether they are were interviewed in the United States or Puerto Rico. The forms
for the United States and Puerto Rico contain slightly different formatting; therefore, the forms
are keyed in separate batches at NPC.
The ACS GQ questionnaires are taken to a unit where clerks look at each page of every returned
questionnaire. The clerks look for correspondence, which they forward to headquarters, if
necessary. The clerks then scan the bar code on each questionnaire to officially check in the
form, retain the English or Spanish pages of the questionnaire, depending on which side has been
completed, and organize the forms into batches of 50 questionnaires.
Data Capture
After the questionnaires have been checked in and batched into groups of 50, they move to the
keying unit in NPC. NPC clerical staff key the data from the questionnaires and transmit data
files to Census headquarters each night. Final keying of each GQ sample month is scheduled for
the last day of the month following the sample month. This schedule allows approximately 2 ½
weeks to complete all GQ keying after the final delivery of questionnaires for a sample month
has been received. The NPC keying quality assurance verification process mirrors the ACS keyfrom-paper process used for ACS mail returns. See Section 7.2 for additional information about
keying quality assurance verification.
The GQ questionnaire will be designed to accommodate key-from-image data-capture system in
the future. See Chapter 10 for more information on data processing operations.

8-9

8.5 SPECIAL PROCEDURES
A few exceptions to the data collection procedures outlined above are necessary to efficiently
collect data from all GQ facilities. Exceptions to the GQ data collection were made due to the
geographic location of a GQ facility and because of GQ security requirements.
Biannual Data Collection in Remote Alaska
FRs will conduct data collection at sample GQ facilities in Remote Alaska during two separate
periods each survey year. FRs will visit a sample of GQ facilities from January through midApril, and from September through mid-January. This exception is needed because of
difficulties accessing these areas of the United States at certain times of the year. The two time
periods designated for GQ interviewing are the same as those used for ACS data collection from
sample housing units in Remote Alaska. See Chapter 7 for more information about data
collection in Remote Alaska.
Annual Data Collection Restrictions in Correctional and Military Facilities
The FRs will conduct all data collection at state prisons, local jails, halfway houses, military
barracks, military disciplinary barracks, and correctional institutions classified as “other” only
once each survey year. These GQ types, when selected for the sample multiple times throughout
the survey year, will have each instance of selection clustered into one random month for data
collection. The Census Bureau agreed to a Department of Justice request to conduct data
collection at each sampled State prison and local jail only once a year. Limiting FR visits to
military GQ facilities during the year is also necessary as a result of the increased potential for
periods of “no-access” at military installations due to heightened national security.
When these GQ types are selected for the sample more than once in a year, the FR (or group of
FRs) will make one visit and conduct all interviews at the GQ facilities during one randomly
assigned month in the survey year, rather than multiple times each year as is the procedure for
data collection at non-correctional sample GQ facilities. The FR does not conduct the GQ-level
data collection phase after completing it once for the GQ sample facility. The GQFQ will
automatically take the FR to the person-level sample selection screen for each multiple sample
occurrence of the GQ facility. The FR (or group of FRs) will conduct the person-level data
collection.
Survey Period and Security Restrictions in Federal Correctional Facilities
The FRs will conduct person-level data collection during a 4-1/2 month period for selected
federal prisons and federal detention centers. The Census Bureau will provide the Bureau of
Prisons (BoP) with a file containing the federal prisons and federal detention centers that were
sampled for the survey year. The BoP will update the GQ-level information and generate the
person-level samples for these GQ facilities. All FR data collection at the federal correctional
8-10

facilities will be conducted from September until mid-January each survey year. Each sampled
facility will be visited once during the 4-1/2 month data collection period.
The BoP sample GQ list is provided to the ROs one year prior to the start of data collection at
these sample GQ facilities. The BoP will deliver the person-level sample to the Census Bureau
for processing before sending these GQ assignments to the appropriate ROs. The BoP conducts
the security clearances on a list of FR names provided to them by the ROs. This process takes
eight to 10 weeks. The FRs cannot contact any federal prison or federal detention center until
informed by their RO that all clearances and BoP contact notifications have taken place. The
BoP will provide the GQ contact names and phone numbers to the ROs prior to the start of data
collection. The RO will contact the FR after clearance, provide them with the contact
information for their BoP assignments, and give the FR permission to begin contacting the
federal prison or federal detention centers in their assignment list.
The FR will call the GQ contact to make an appointment to visit and conduct all interviews in
one day. Appointments may be scheduled in advance for any time during the federal
prison/detention center data collection period. The FR cannot visit a BoP prison or detention
center without calling and setting up an appointment first. The FR will not be authorized to enter
the facility without an appointment. Each facility will have different periods of time when there
is limited or no access allowed. FRs will prepare their survey packages before entering the
federal prison. The person-level interview process continues from this point in the same way as
described for all other GQ facilities.

8.6 REFERENCES
U.S. Census Bureau (2005a), “2006 ACS Group Quarters Field Representative’s Manual,” U.S.
Census Bureau, Washington, DC, 2005.
U.S. Census Bureau (2005b), “2006 Census Test Group Quarters Definitions and Code List,”
2010 Census Test Memorandum Series, Decennial Management Division, U.S. Census Bureau,
Washington, DC, 2005.
U.S. Census Bureau (2005c), “2006 PRCS Group Quarters Field Representative’s Manual,” U.S.
Census Bureau, Washington, DC, 2005.

8-11

Chapter 9.
Language Assistance Program
9.1 OVERVIEW
The language assistance program for the American Community Survey (ACS) includes the full
set of methods and procedures designed to assist sample households with limited English
proficiency in completing the ACS interview. Language assistance can be provided in many
forms, including the development of translated instruments and other survey materials, the
recruiting and training of bilingual interviewers, and the provision of telephone or Internet
assistance in multiple languages. Providing language assistance is one of many ways that the
ACS can improve survey quality. Language assistance can reduce levels of survey non-response
and the potential for non-response bias. It can also reduce the introduction of response errors by
ensuring that individuals with limited English skills more fully understand the survey questions.
The ACS language assistance program reflects the use of several key tools to support each of the
three modes of data collection - mail, telephone, and personal visit. The development of these
tools is based on research that was conducted to assess the current performance of the ACS for
non-English speakers. McGovern (2004) found that despite the limited availability of mail
questionnaires in languages other than English, non-English speakers were successfully
interviewed by telephone and personal visit follow-up modes. She also found that the level of
item non-response for households speaking languages other than English were consistent with
the low levels of item non-response in English-speaking households. These results focused
language assistance priorities on the development of tools to improve the quality of data
collected in the telephone and personal visit data collection modes. The program includes
assistance in a wide variety of languages during the telephone and personal visit non-response
follow-up stage.1 Efforts to expand language assistance in the mail mode were moved to a
slightly later time table. The greatest emphasis in the mail mode at this time is to support
Spanish-language speakers.
This chapter provides greater detail on the current language assistance program. It begins with
an overview of the language support, translation, and pretesting guidelines. Current methods are
then discussed for the mail mode and the telephone and personal visit follow-up modes.
Research and evaluation activities are discussed in Section 9.7.

In 2003 interviewer language capabilities included English, Spanish, Portuguese, Chinese, Russian, French, Polish,
Korean, Vietnamese, German, and Japanese.

9-1

9.2 BACKGROUND
The 2010 Decennial Census program has placed a priority on developing and testing tools to
improve the quality of data collected from people with limited English proficiency. This
population represents a growing share of the total population. The 2004 ACS found that
8.4 percent of the total population who speak a language other than English at home, speak
English less than “very well.” This is an increase from 7.6 percent in 2000 (U.S. Census Bureau,
2004b). The ACS is an integral part of the decennial census and the 2010 Census includes many
common data collection activities and quality objectives. Staff involved in the ACS and the
2010 Census have been working jointly to study language barriers and effective methods for data
collection.

9.3 GUIDELINES
The Census Bureau does not require that all survey instruments or materials be translated into
any specific set of languages. Each census and survey determines the appropriate set of
translated materials and language assistance options needed to ensure high quality survey results.
The Census Bureau does, however, require that guidelines be followed whenever a decision is
made to translate a data collection instrument or a respondent letter.
In 2004 the Census Bureau released guidelines for language support translation and pretesting.
These guidelines state that data collection instruments that are translated from a source language
into a target language should be reliable, complete, accurate, and culturally appropriate. Reliable
translations convey the intended meaning of the original text. Complete translations neither add
new information nor omit existing information provided in the source document. When a
translation is free of both grammatical and spelling errors, it is considered to be an accurate
translation. Cultural appropriateness considers the culture of the target population in developing
the text for translation. In addition to meeting these criteria, translated Census Bureau data
collection instruments and related materials should also have semantic, conceptual, and
normative equivalence. The Census Bureau guidelines recommend that a translation team
approach be followed to ensure equivalence. The language support guidelines include
recommended practices to prepare, translate, pretest, and revise materials and to ensure sound
documentation of findings. See U.S. Census Bureau (2004a) for details on these guidelines. The
ACS utilizes the Census Bureau guidelines in the preparation of data collection instruments, as
well as advance letters and other respondent communication.

9.4 MAIL DATA COLLECTION
The Census Bureau currently mails out ACS questionnaires in a single language to each sample
address. In the United States, English language forms are mailed, and in Puerto Rico, Spanish
language forms are mailed. The cover of the English mailout form (in the United States) and the
Spanish mailout form (in Puerto Rico) contains a message written in the alternative language
requesting that people who prefer to complete the survey in the alternate language call a toll-free
9-2

assistance number to obtain assistance or to request a form in that language. Upon request, the
Census Bureau will mail a Spanish form to people in the United States and an English form to
people in Puerto Rico. In 2005 we received requests for Spanish questionnaires from less than
0.01 percent of the mailout sample (Griffin, 2006b).
Telephone questionnaire assistance (TQA) is provided in both English and Spanish. A call to the
toll-free Spanish help number reaches a Spanish speaker directly. The interviewer will either
provide general assistance or conduct the interview. Interviewers are encouraged to convince
callers to provide the interview over the phone.

9.5 TELEPHONE AND PERSONAL VISIT FOLLOW-UP
The call centers and the regional offices that conduct the computer-assisted telephone
interviewing (CATI) and computer-assisted personal interviewing (CAPI) non-response
follow-up operations make every effort to hire bilingual staff to conduct interviews with
households that speak a language other than English. In addition, the CAPI interviewers are
instructed to search for interpreters either within the sample household or from the neighborhood
to assist in data collection. The regional offices maintain a list of interpreters skilled in many
languages who are available to assist the CAPI interviewer in the language preferred by the
respondent. A flashcard is available for interviewers to use to identify the specific language
spoken when they encounter a household with which they cannot communicate.
The ACS CATI and CAPI survey instrument is currently available in both English and Spanish.
Interviewers can conduct interviews in additional languages if they have those language
capabilities. Because a translated instrument is not available in languages other than English and
Spanish, interviewers translate the English version during the interview and record the results in
the English instrument. The Census Bureau is exploring the possibility of creating translated
instruments or guides for interviewer use in languages other than English and Spanish. Special
procedures and a unique training module about collecting data from respondents who do not
speak English exist for CATI and CAPI interviewers. All ACS interviewers are given this
training as part of their classroom interviewer training. The training is designed to improve the
consistency of procedures used in these situations and to remind interviewers of the importance
of collecting complete data for all households.
The CATI and CAPI instruments collect important data on language-related issues. This
information allows the Census Bureau to monitor how data are being collected, including
information on the frequency of use of interpreters and the frequency of use by the interviewers
of the Spanish instrument. The instrument also records how often an interviewer conducts
translations of their own into a different language. For example, Griffin (2006b) found that in
2005 over 86 percent of all CAPI interviews that were conducted with Spanish-speaking
households were conducted by a bilingual, Spanish-speaking interviewer. She also found that
about eight percent of the interviews conducted with Chinese-speaking households required
assistance of an interpreter who was not a member of the household.

9-3

Additional reports allow the call centers and the regional offices to identify CATI and CAPI
cases that could not be completed due to language barriers. A profile of this information by the
specific language highlights the languages for which greater support is needed. Griffin (2006b)
found that out of a total of 31,489 cases that were in the 2005 CATI workload and were
identified as requiring a language other than English, 9.3 percent could not be interviewed due to
a language barrier. The greatest language needs were Spanish, Vietnamese, Korean, and
Chinese. Call center managers used this information to identify specific language recruiting
needs and have hired additional staff with these language skills. Similar reports are used for
CAPI. Refer to Griffin (2006b) for more information on the results of these CATI and CAPI
evaluations.
Griffin and McGovern (2004) compared the language abilities of CAPI interviewers in each
regional office with the language needs of the population for that regional office. This
assessment was based on 2003 ACS data on language spoken at home and on 2003 regional
office staffing information. The regional offices used these data to assist in recruiting for to
support the full sample expansion in 2005. An update of this assessment for both CATI and
CAPI is planned to look at current staffing, now that the ACS sample has been expanded into all
counties across the United States.

9.6 GROUP QUARTERS
Chapter 8 describes the data collection methodology for people living in group quarters (GQ)
facilities. Two instruments are used in GQ data collection – a paper survey questionnaire that is
used to interview people living in GQ facilities and an automated instrument used to collect
administrative information from each facility. The Census Bureau designed and field tested a
bilingual (English/Spanish) GQ questionnaire in 2005. Interviewers used these bilingual
questionnaires to conduct interviews with a small sample of GQ residents. An interviewer
debriefing found that the interviewers had no problems using this questionnaire. This form is
currently used for GQ data collection. As with other personal visit data collection activities in
the ACS, the Census Bureau will hire bilingual interviewers to conduct interviews in GQ
facilities in Puerto Rico with non-English speakers. The Group Quarters Facility Questionnaire
(GQFQ) is available in both English and Spanish.

9.7 RESEARCH AND EVALUATION
Due to limited resources, research and development activities related to the language assistance
program were prioritized. Of critical importance was a benchmarking of the effectiveness of
current methods. The potential for non-response bias due to language barriers was assessed in
McGovern (2004) and Griffin and Broadwater (2005). In addition, ACS staff created a website
on quality measures with annual information about the effect of language barriers on survey nonresponse. See Chapter 15 for more information on quality measures. These evaluations and the
website show that current methods result in very low levels of non-interviews from the
interviewers’ inability to speak the respondent’s language. These non-response levels remain
9-4

low because of special efforts undertaken in the field to use interpreters and other means to
conduct these interviews. Item level non-response was also assessed by McGovern (2004). She
found that although most non-English speakers respond during an interviewer-assisted mode,
such as telephone or personal visit, the mail returns received from non-English speakers are
nearly as complete as those from English-speakers. She also found that the interviews conducted
by telephone and personal visit with non-English speakers are as complete as those from
English-speakers. The Census Bureau continues to monitor unit non-response due to language
barriers.
Language barriers can result in measurement errors when respondents do not understand the
questions or when interviewers incorrectly translate a survey question. The production of
accurate and reliable translated instruments in languages such as Spanish, Korean, Chinese,
Vietnamese, and Russian will reduce the potential for such errors to be introduced into ACS
data. Cognitive testing of the ACS Spanish instrument identified translation concerns (Carrasco,
2003). The Census Bureau is planning a more complete assessment of the Spanish instrument to
improve the quality of data collected from Spanish-speaking households. Staff are also
exploring options for developing either translated instruments or language guides for use by
telephone and personal visit interviewers who conduct interviews in Chinese, Korean,
Vietnamese, and Russian.
Future research is planned to develop and test additional language assistance materials for the
mail mode. Increasing levels of participation by mail can reduce survey costs and improve the
quality of final ACS data. Future updates of this report will reflect progress in this area.

9.8 REFERENCES
Carrasco, Lorena (2003), “The American Community Survey en Español: Using Cognitive
Interviews to Test the Functional Equivalency of Questionnaire Translations,” Statistical
Research Division Study Series Report, U.S. Census Bureau, Washington, DC, 2003.
Griffin, Deborah (2006a), “Method of Interview for ACS CAPI Households with Limited
English Proficiency,” American Community Survey Discussion Paper, U.S. Census Bureau,
Washington, DC, 2006.
Griffin, Deborah (2006b), “Requests for Alternative Language Questionnaires,” American
Community Survey Discussion Paper, U.S. Census Bureau, Washington, DC, 2006.
Griffin, Deborah and Broadwater, Joan (2005), “American Community Survey Noninterview
Rates Due to Language Barriers,” paper presented at the Meetings of the Census Advisory
Committee on the African American Population, the American Indian and Alaska Native
Populations, the Asian Population, the Hispanic Population, and the Native Hawaiian and Other
Pacific Islander Populations on April 25 - 27, 2005.

9-5

Griffin, Deborah and McGovern, Pamela (2003), “Language Action Plan for the American
Community Survey,” U.S. Census Bureau, Washington, DC, 2003.
Griffin, Deborah and McGovern, Pamela (2004), “Assessment of Language Needs and
Capabilities by Regional Office for the American Community Survey,” U.S. Census Bureau,
Washington, DC, 2004.
McGovern, Pamela D. (2004), “A Quality Assessment of Data Collected in the American
Community Survey for Households with Low English Proficiency,” U.S. Census Bureau,
Washington, DC, 2004.
U.S. Census Bureau (2003), “American Community Survey Operations Plan, Release 1: March
2003,” Washington, DC, 2003.
U.S. Census Bureau (2004a), “Census Bureau Guideline: Language Translation of Data
Collection Instruments and Supporting Materials,” Internal U.S. Census Bureau document,
Washington, DC, 2004.
U.S. Census Bureau (2004b), “Selected Social Characteristics: 2004," Washington, DC, 2004,
.

9-6

Chapter 10.
Data Preparation and Processing for Housing Units and
Group Quarters
10.1 OVERVIEW
The broad purpose of data preparation and processing is to take the response data gathered from
each survey collection mode to a point where the data can be used to produce survey estimates.
Files to be edited, known as edit input files, are created during the data preparation phase by
merging operational status information for each housing unit (HU) and group quarters (GQ)
facility with the files that include the response data. The combined data must go through a
number of processing steps before they are ready to be tabulated. Once the edit input files are
prepared, the edit and imputation process can be initiated. Editing and imputation ensure that the
final data are consistent and complete. Subject matter analysts thoroughly examine and approve
the results of the edit and imputation process.
Figure 10-1 depicts the overall flow of data as they pass from the data collection operations
through data preparation and processing and into data products development. All activities
leading to the creation of the edit input files are considered data preparation activities. The
activities that follow constitute data processing. Each of the files, operations, and processes
included in Figure 10-1 is described in Sections 10.2 and 10.3. The American Community
Survey (ACS) data preparation operations include a wide range of activities, which may occur
daily, monthly, or yearly. Data processing operations occur once a year, at the end of each year
of data collection, resulting in an annual data release. The annual release reflects data collected
in the months of the previous calendar year by mail, computer-assisted telephone interviewing
(CATI), and computer-assisted personal interviewing (CAPI).

10-1

Figure 10-1. American Community Survey (ACS) Data Preparation and Processing

This chapter summarizes the data preparation and data processing activities used for both HU
and GQ data. The processes and files used for GQ facilities are identical to those used for HUs.
GQ data are, however, limited to population variables, except for one housing variable, food
stamps. Section 10.2 details the data preparation activities, including the creation and processing
of the data capture files, coding files, and the edit input files. Section 10.3 details the data
processing activities, which include editing and imputation, the generation of recoded variables,
weighting, and the application of disclosure avoidance techniques, the review of edit results, and
the creation of input files for data products. Section 10.4 lists references which provide greater
detail for many of these activities.

10.2 DATA PREPARATION
The ACS control file includes detailed information documenting operational outcomes for every
ACS sample case. For the mail operations, this file documents the receipt and check-in date of a
questionnaire returned by mail. The status of data capture for these questionnaires returned by
mail and the results of the failed-edit follow-up (FEFU) operation are also recorded on the ACS
control file. For details of these mail data collection operations, refer to Chapter 7. For the
CATI and CAPI operation, the ACS control file stores information on whether or not the unit
was determined to be occupied or vacant. Refer to Chapter 7 for more information on the CATI
and CAPI operation. The ACS control file is updated daily, based on data from data collection
and data capture operations, to reflect the latest status of all sample cases.
Data preparation, which joins together control file information for each case with the raw,
unedited response data, involves the following three operations:

10-2

•
•
•

Creation and processing of data capture files,
Coding, and
Creation of edit input files.

Creation and Processing of Data Capture Files
Every day throughout the year, HU data are collected in the ACS by mail, CATI, CAPI, FEFU,
and telephone questionnaire assistance (TQA). GQ data are also collected throughout the year
by personal visit interviews. Each of these data collection activities is described in Chapters 7
and 8. The data from questionnaires received by mail are returned to headquarters from data
capture activities on a flow basis, with daily file deliveries. The data collected by the use of
CATI and CAPI instruments are returned to headquarters at the end of each month.
The ACS uses a centralized data capture file (DCF) to store all response data. At the end of each
month, all of the response data for that month from the various collection modes are added to the
DCF. These data represent all data collected during the previous month, regardless of the
sample month for which the HU or GQ facility was chosen. During the creation of the DCF,
responses are reviewed and illegal values and responses of “Don’t Know” and “Refused” are
identified and coded. These responses are blanked in the final edit input files. The ACS
questionnaire includes a set of questions that offer the possibility of a write-in response. Each of
these write-in entries requires coding to turn it into a numerical value. A part of processing
newly received data onto the DCF involves identifying these write-in responses and placing
them in a series of files that serve as input to the various coding operations. The DCF is a
collection of files, including a housing data file, a population data file, a GQ data file, and a
separate file for each of the various write-in entries. The GQ write-ins are stored along with
write-ins from the HU data. At the end of the year, the response data stored in the DCF will
have been updated 12 times and will become a principal source for the edit-input process.
Coding input files are created from the DCF files of write-in entries. Edit input files combine
data from the remaining DCF files and the returned coding files.
Coding
The ACS form contains several questions that either require or permit respondents to write in
their responses. These write-in responses must be coded before they can be processed in editing
and imputation. This second step in the data preparation phase is performed on a monthly basis,
and follows the monthly run of the DCF. The coding input files provide monthly coding
workloads, with the expectation that the files will be completely coded within the space of one
month. The coded files are called coding output files and are accumulated throughout the year in
preparation for the annual processing operation.
The 2005 ACS questionnaire contains the following write-in fields which require coding: race,
Hispanic origin, place of birth, ancestry, migration, language, place of work, industry, and
occupation. In the coding phase, fields with write-in values are translated to a prescribed list of
valid codes. For more details, see U.S. Census Bureau (2006). The write-in responses are coded
10-3

in slightly different ways depending on the subject matter. Place of birth, migration, and place
of work responses require coding of a geographic location. This geocoding is accomplished in
an automated first pass at Census Bureau headquarters using automated geocoding software.
Cases that cannot be coded automatically are coded clerically by geocoding experts at the
National Processing Center (NPC). Questions dealing with industry and occupation are all
coded clerically at the NPC. All other coding, referred to as general coding, is accomplished at
Census Bureau headquarters. The first attempt is automated, and the cases that cannot be
automatically coded are coded clerically. All of these operations are subject to quality assurance
processes to ensure that the coding is consistent and accurate.
Creation of Edit Input Files
The final data preparation operation involves creating edit input files to be used for data
processing. To create these edit input files, a number of steps must be followed. Data from the
DCF are combined with data returned from the coding operations, and operational information
for each case is merged from the ACS control file.
This combined set of data is referenced to determine the number of people living in a HU and to
identify the response type of the HU. The data are examined to determine the total number of
people living in the HU, which is not always straightforward. For example, on a mail return,
sometimes the count of people on the cover of the form does not match the number of people
reported on the inside of the form. For a CATI or CAPI interview, this uncertainty also arises
when people are initially identified as residents, but later during the interview it is determined
that they really are not residents of the HU. Each HU is assigned a response type that describes
it as either an occupied unit, a temporarily occupied unit, a vacant unit, a delete1, or a noninterview. While this type classification already exists on the DCF, it can be changed from
“occupied” to “vacant” or even to “non-interview” under certain circumstances, depending on
the final number of persons in combination with other variables.
The Census Bureau calculates an acceptability index at the form level to determine whether there
is enough information about an occupied HU or a GQ person to consider the record an interview.
The acceptability index is based on the number of responses to several questions. All records for
vacant units are considered to be acceptable. Records that do not meet this acceptability index
are classified as non-interviews and will not be included in further data processing. Noninterviews are accounted for in the weighting process. For details of this process, refer to
Chapter 11.
Once the universe of acceptable interviews is determined, the HU data are reviewed to
unduplicate multiple returns for a single HU. More than one response can exist for a HU for
several reasons. A household might return two mail forms, one in response to the initial mailing

Deleted HUs are units that are determined to be non-existent, demolished, or commercial units.

10-4

and a second in response to the replacement mailing. A household might return a mailed form
but also be interviewed in CATI or CAPI if the mail form was returned after a CATI or CAPI
interview is conducted. If more than one return exists for a HU, a quality index is used to choose
one of the returns. The quality index is calculated as the percentage of items with responses out
of the total number of items that should have been completed. This index considers responses to
both population and housing items. Unduplication of multiple returns is not necessary in
processing GQ data.
After multiple returns are resolved, each sample case is assigned a value for three critical
variables – data collection mode, month of interview, and case status. The month in which data
were collected from each sample case is determined and used to define the universe of cases to
be used in the production of survey estimates. For example, data collected in January 2005 are
included in the 2005 ACS data products. The majority of sample cases fall into one of the
following statuses: occupied, vacant, temporarily occupied, non-interview, delete, and subsampled out. Additional statuses exist to describe rare conditions that result in a sample case not
being classified in one of these six categories.
The final step in edit input creation is the creation of the actual edit input files. They contain
essentially the information in the DCF housing and person files, but only for the unduplicated
response records that met the requirements for an interview. At this point, a first name recode
and a Spanish surname recode are added to each person record. For cases where the number of
person records is less than the determined number of person, extra person records are created
which are assigned values during the edit. This addition of blank records does not apply to GQ
facilities. Failure to collect data for a GQ sample person is addressed in the non-interview
adjustment step of the GQ weighting process. For more information on this process, refer to
Chapter 11. The edit input files are fully verified.

10.3 DATA PROCESSING
Once the edit input files have been generated and verified, the edit and imputation process can be
initiated. The main steps in processing are:
•
•
•
•
•

Editing and imputation,
Generating recoded variables,
Weighting and the application of disclosure avoidance techniques,
Reviewing edit results, and
Creating input files for data products.

This chapter does not include a discussion of weighting or of the application of disclosure
avoidance methods. Chapter 11 provides details about the derivation and application of person
and housing weights. Chapter 13 includes information on disclosure avoidance methods.

10-5

Editing and Imputation
The edit input files described in the previous section are the basis for this processing step. The
HU edit input files contain one data record for each interviewed sample address. Cases
classified as non-interviews are not included on the edit input file. There is a separate edit input
file for GQ data. Only the data collected during the calendar year being processed are included.
The files contain the raw data as collected, so responses to some questions are missing and some
responses are not consistent. HU data are expanded to include blank records for people who
were not included in the initial response records but are determined to be part of the household.
The data are separated by state, since the HU editing and imputation operations are completed on
a state-by-state basis.
Edit and imputation rules are designed to ensure that the final edited data are as consistent and
complete as possible, and thus ready for tabulation. Subject matter experts develop these rules,
and processing staff write and run the programs. Application of edit and imputation rules helps
maintain data quality when complete responses cannot be obtained. These rules are used to
identify and account for missing, incomplete, and contradictory responses. Application of these
rules in the ACS does not affect the total number of people or housing units in the data file, as
the rules are used only to supply responses for missing or inconsistent answers about a housing
unit’s characteristics, not to create new units or new people. In each case where a problem is
detected, consistent, pre-established rules govern its resolution.
There are two principal ways to deal with missing or inconsistent data--assignment and
allocation. Both are considered imputation methods. Assignment involves looking at other data
as reported by the respondent. For example, when determining the gender of a person, the first
name is often used. An indication that this person had children born in the past 12 months would
indicate that the person was female. This approach also uses data as reported by other people in
the household to fill a blank or inconsistent field. For example, if the reference person and the
spouse are both citizens, a child with a blank response to citizenship is assumed to be a citizen.
Assigned values are expected to have a high probability of being correct. Certain values, such as
whether a person has served in the military, are better if provided from another HU or from a
person with similar characteristics. This approach uses the technique known as hot deck
allocation.
Hot deck allocation uses a statistical approach to supply responses for missing or inconsistent
data from similar responding HUs or people in the sample. A hot deck matrix contains the data
for prospective “donors” that can be called upon when a “recipient” needs data. Specifications
for each question or item outline in detail how the hot deck matrices for that item are to be
structured in the editing system. Classification variables for an item are used to determine
categories of donors (referred to as cells) in the hot deck. These donors are HUs or people with
complete and consistent data. One or more cells constitute the matrix used for allocating one or
more items. Some hot deck matrices are simple and contain only one cell, while others may
have thousands of cells. For example, in editing the housing item known as tenure (which
identifies whether the housing unit is owned or rented), a simple hot deck of three cells is used,
10-6

where the cells represent responses from single-family buildings, multi-unit buildings, and cases
where a value for the question on type of building is not reported. Alternatively, there are
dozens of different matrices defined with thousands of cells specified in the joint economic edit,
where many factors are used to categorize donors for these cells, including sex, age, industry,
occupation, hours and weeks worked, wages, self-employment income, and so forth.
Sorting variables are used to order the input data prior to processing and thus determine the
nearest neighbors for hot deck allocation. In the ACS, the variables used for this purpose mainly
consist of geographic variables like state, county, census tract, census block, and basic street
address. This sequence is used since it has been shown that housing and population
characteristics are often more similar within a given geographic area. One exception to this rule,
is the item place of work, where a unique ordering is used to provide a better grouping of donor
values in processing this question. The sorting variables for this edit are used to combine similar
people together by industry groupings, means of transportation to work, minutes to work, state of
residence, county of residence, and state in which the person works.
For each cell in the hot deck, up to four donors are stored at any one time. The hot deck cells are
given starting values which have been determined in advance to be the most likely values for
particular categories. These starting values, known as cold deck values, are used as donor values
only in rare instances where there are no donors. Procedures are employed to replace these
starting values with actual donors from cases with similar characteristics in the current data file.
This step is referred to as hot deck warming.
The edit and imputation programs look at the housing and person variables according to a
predetermined hierarchy. Thus, each item in a response record is edited and imputed in an order
delineated by this hierarchy. The hierarchy includes the basic person characteristics first,
including sex, age, and relationship, followed by most of the detailed person characteristics, then
all the housing items. Lastly, the remainder of the detailed person items, such as migration and
place of work, are edited. For HUs, the edit and imputation process is performed for each state
separately, with the exception of the place-of-work item, which is done at the national level, by
necessity. For GQ facilities, the data are processed nationally by GQ type, with GQ facilities of
the same type (nursing homes, prisons, and so on) edited and imputed together.
In the edit and imputation system, a flag is associated with each variable to indicate whether or
not it was changed in any way and if so, the nature of the change. These flags support the
subject matter analysts in their review of the data and provide the basis for the calculation of
assignment and allocation rates. Allocation rates measure the proportion of values that required
hot deck allocation and are an important measure of data quality. The ACS provides these rates
on its website under the quality measures section. For the most recent set of rates, refer to
the ACS quality measures website () .
Chapter 15 provides more information about these quality measures.

10-7

Generating Recoded Variables
During data processing, new variables referred to as recoded variables, or recodes, are calculated
based on the response data. These recodes are used later in tabulating the data. There are many
recodes for both housing and person data. For example, one recode variable is called “Presence
of Persons 60 and Over.” In this case, the algorithm examines each person’s age in a HU and
sets the variable equal to a value depending on the results. There are four possible values: null
value, indicating that this variable does not apply because the HU is not occupied; 0, indicating
no people 60 years of age or older are in the HU; 1, indicating one person 60 years of age or
older is in the HU; or 2, indicating two or more people 60 years of age or older are in the HU.
For a complete list of recoded variables see Appendix D.
Reviewing Edit Results
After editing and imputation are complete, weighted, and swapped, subject matter analysts
review the resulting data files. They contain both unedited and edited data, together with the
accompanying imputation flag variables that indicate which missing, inconsistent, or incomplete
items were filled by imputation methods. The subject matter analysts undertake their own
analyses, looking for problems or inconsistences in the data, or differences from what is outlined
in the specifications. Analysts may also examine unusual individual cases that were changed
during editing.
In addition, the analysts receive a number of special reports based on the edit outputs and
multiple years of survey data that can be used in comparisons. These reports and data are used
to help isolate problems in the specifications or in processing. They include detailed information
on imputation rates for all data items, as well as tallies representing counts of the number of
times various logic was executed during editing. If editing problems are discovered in the data
during this review process, it is often necessary to re-run the programs and repeat the review.
Creating Input Files for Data Products
Once the files of edited data and their associated recodes have been approved by the subject
matter analysts, they are ready to serve as inputs to the data products processing operation.
Chapter 13 details the production of ACS data products. If errors detected during the creation of
data products are attributable to editing problems, it may be necessary to repeat the editing and
review processes.

10.4 REFERENCES
Raglin, David (2004), “Edit Input Specification 2004,” Internal U.S. Census Bureau technical
specification, Washington, DC, 2004.

10-8

U.S. Census Bureau (1997), “Documentation of the 1996 Record Selection Algorithm,” Internal
U.S. Census Bureau memorandum, Washington, DC, 1997.
U.S. Census Bureau (2003a), “American Community Survey Operations Plan Release 1: March
2003,” Washington, DC, 2003.
U.S. Census Bureau (2003b), “Data Capture File 2003,” Internal U.S. Census Bureau technical
specification, Washington, DC, 2003.
U.S. Census Bureau (2004a), “American Community Survey Control System Document,"
Internal U.S. Census Bureau documentation, Washington, DC, 2004.
U.S. Census Bureau (2004b), “Housing and Population Edit Specifications,” Internal U.S.
Census Bureau documentation, Washington, DC, 2004.
U.S. Census Bureau (2004c), “Housing Recodes 2004,” Internal U.S. Census Bureau data
processing specification, Washington, DC, 2004.
U.S. Census Bureau (2006), “American Community Survey 2004 Subject Definitions,”
Washington, DC, 2006,
.

10-9

Chapter 11.
Weighting and Estimation
11.1 OVERVIEW
Beginning in 2010, three sets of American Community Survey (ACS) estimates will be
calculated annually for specified geographic areas using data collected over three different time
periods. In general, the ACS will produce and publish estimates for the same set of statistical,
legal, and administrative entities that were produced and published for the Census 2000 long
form sample. The ACS will publish a variety of tabulation summaries for the nation, states,
American Indian and Alaska Native (AIAN) areas, counties and county equivalents, minor civil
divisions (MCDs), incorporated places, and census tracts among others (see Section 13.2). The
population of each geographic area determines whether estimates can be released based on five
years, three years, or a single year of sample data.
•

Multi-year estimates based on five years of sample data are published for all statistical,
legal, and administrative entities, including census tracts, block groups, and small
incorporated places such as cities and towns. The five-year period estimates are based on
data collected during the 60 months of the five most recent collection years.

•

Geographic entities with populations of at least 20,000 receive three-year estimates based
on data collected during the 36 months of the three most recent collection years.

•

Geographic entities with populations of at least 65,000 receive single-year estimates
based on data collected during the twelve months of the most recent calendar year.

When the three- and five-year period estimates are produced each year, data from the most
recent year replace the earliest year of data included in the previous years’ estimates.
The basic estimation approach is a ratio estimation procedure that results in the assignment of
two sets of weights: a weight to each sample person record and a weight to each sample housing
unit (HU) record. For any given tabulation area, a characteristic total is estimated by summing
the weights assigned to the people, households, families, or HUs in a specific area possessing the
characteristic of interest. Estimates of person characteristics are based on the person weight.
Estimates of family, household, and HU characteristics are based on the HU weight. As with
most household surveys, weights are used to bring, on the average, the characteristics of the
weighted sample more into agreement with the characteristics of the full population. The
weights compensate for differences in sampling rates across areas, for differences between the
full sample and the interviewed sample, and for differences between the sample and independent
estimates of basic demographic characteristics (Alexander, Dahl, and Weidman, 1997).
The next section describes in detail the methodology for computing the 2005 ACS single-year
estimates, which use both housing data and population data for people living in HUs. Later
sections discuss the single-year weighting methodology for total population–group quarters
(GQ) facilities and HU persons combined. This methodology is planned for the 2006 ACS and
11-1

beyond. The ratio estimation method will also be used to produce three- and five-year period
estimates. The general procedures for the production of these period estimates are expected to
be similar to the single-year period estimates. Additional details will be provided in future
versions of this document.
The input to the weighting process is the final edited data file. The estimation methodology is
implemented by “estimation area,” which is either a county or group of less populous counties.
Estimates will be calculated for HUs and people living in HUs for the 2005 ACS. Single-year
weighting can be thought of as occurring in three stages. The first stage is based on the sampling
rates used to select the HU sample. The second stage adjusts the weights of responding HUs to
compensate for the loss of weights of non-responding HUs. The third stage controls the
estimates of HUs and persons by age, sex, race, and Hispanic origin to estimates from the
Intercensal Population Estimates (IPEs) program of the Census Bureau.

11.2 2005 ACS WEIGHTING - PROBABILITY OF SELECTION
The first stage of weighting involves three steps. First, each HU is assigned a base weight as a
function of its initial probability of selection. The second step involves an adjustment to account
for the sampling of HUs prior to computer-assisted personal interviewing (CAPI). Then a
further adjustment is made to these weights to reduce variability in the monthly weighted totals.
Assignment of base weights and CAPI sub-sampling factor
Each HU address in the annual sample is selected with a final overall sampling rate as given in
Chapter 4. The base weight for a HU is the inverse of this overall rate. HUs that are sent to
CAPI are eligible to be sub-sampled at one of the rates as described in Chapter 4, Table 4.3.
Those selected for the CAPI sub-sample and for which no late mail return is received in the
CAPI month receive a CAPI sub-sampling factor equal to the inverse of their sub-sampling rate,
and those not selected receive a factor of 0.0. HUs that are eligible for CAPI and for which a
mail return is received in the CAPI month, and HUs not eligible for CAPI (cases whose
interviews were completed by mail or by computer-assisted telephone interviewing (CATI))
receive a CAPI sub-sampling factor of 1.0. A new weight equal to the base weight times the
CAPI sub-sampling factor is calculated for every HU. Similarly, after each of the subsequent
weighting steps, with one exception that will be noted, a new weight is calculated as the product
of the new factor and the weight following the previous step. For additional details about the
weighting steps discussed in this and the following section, see Asiala (2004).
Adjustment for variation in monthly response
The goal of ACS estimation is to represent the characteristics in a geographic area across the
specified time period. For single-year estimates this is 12 months. It is 36 months for three-year
estimates and 60 months for five-year estimates. An even distribution of HU weights is
therefore required by month. The data for HUs assigned to any sample month can be collected
during a three month time period. For example, the households in the January sample month can
have their data collected in January, February, or March. Each HU in sample belongs to a

11-2

tabulation month, which is the month the case is completed. This is the month given by a
respondent on the first page of the completed mail questionnaire, or the month the interview is
completed by CATI or CAPI.
Due to seasonal variations in response patterns, the number of HUs in tabulation months may
vary, thereby over-representing some months and under-representing other months in the singleyear and multi-year estimates. To smooth out the total weight for all sample months, a variation
in monthly response factor (VMS) is calculated for each month as:
VMS = (total sample base weights of all HUs in that sample month) / (total weights after
CAPI sub-sampling factor of all HUs in that sample month).
When this factor is applied, the total VMS weight of all HUs tabulated in a sample month will be
equal to the total base weight of all HUs selected in sample that month. The result is that each
month contributes about 1/12 to the total single-year estimates. Analogously, each month
contributes about 1/36 and 1/60 to the three-year and five-year estimates, respectively.

11.3 2005 ACS WEIGHTING - NON-INTERVIEW ADJUSTMENT
The non-interview adjustment uses two factors to account for sample HUs for which an
interview is not completed. These steps are discussed below. Once the two steps are completed,
factors are applied, and the sum of the weights of the interviewed HUs will equal the sum of the
VMS weights of the interviewed plus non-interviewed HUs.
Since nothing new is learned about the HU or person characteristics of non-interviewed HUs
during data collection, only characteristics known at the time of sampling can be used in
adjusting for them. Characteristics that have been shown in other surveys and censuses to be
related to HU response include census tract, single-unit vs multi-unit structures (building type),
and month of data collection. If, within counties, a sufficient number of sample HUs were
available to fill the cells of a three-way cross-classification table formed by these three variables,
we could simultaneously adjust for these three factors. But since tracts have fairly small sample
sizes, and the sample is evenly spread across all twelve months of a year, that will not usually be
the case. As a result, the non-interview adjustment is carried out in two steps for each estimation
area. Note that vacant units and deletes1 are excluded from the non-interview adjustment. The
weight corresponding to these types of HUs remain unchanged during this stage of the weighting
process.

Deletes or out-of-scope addresses fall into three categories - a) addresses of living quarters that have been
demolished, condemned, or are uninhabitable because they are open to the elements, b) addresses that do not exist,
and c) addresses that identify commercial establishments, units being used permanently for storage, or living
arrangements known as group quarters.

11-3

Calculation of First Non-Interview Adjustment Factor
In this step, all HUs are placed into adjustment cells based on the cross-classification of building
type and census tract. If a cell contains fewer than 10 interviewed HUs, it is collapsed with an
adjoining tract until the collapsed cell meets the minimum size of 10 interviewed HUs. Cells
with no non-interviews are never collapsed regardless of size. The first non-interview
adjustment factor (NIF1) for each eligible cell is:
NIF1 = (total HU weight after VMS of interviewed occupied and temporarily occupied
HUs and non-interviewed HUs) / (total HU weight after VMS of interviewed
occupied and temporarily occupied HUs).
All occupied and temporarily occupied interviewed HUs are adjusted by this first non-interview
factor. Vacant and deleted HUs get a factor of 1.0, and non-interviews get a factor of 0.0.
Calculation of Second Non-Interview Adjustment Factor
In this step, all HUs are placed into adjustment cells based on the cross-classification of building
type and tabulation month. If a cell contains fewer than 10 interviewed HUs it is collapsed with
an adjoining tabulation month until the collapsed cell has at least 10 interviewed HUs. Cells
with no non-interviews are never collapsed regardless of size. The second non-interview factor
(NIF2) for each eligible cell is:
NIF2 = (total HU weight after VMS of interviewed occupied and temporarily occupied
HUs and non-interviewed HUs) / (total HU weight after NIF1 of interviewed
occupied and temporarily occupied HUs).
NIF1 weights for all occupied and temporarily occupied interviewed HUs are adjusted by this
second non-interview factor. Vacant and deleted HUs get a factor of 1.0, and non-interviews get
a factor of 0.0.
Calculation of Mode Bias Non-Interview Factor
One property that has not been accounted for in the two non-interview factors is that systematic
differences exist between characteristics of households that respond by mail and those that do
not (Weidman, et al., 1995). The same property has been observed in the ACS across response
modes. Virtually all non-interviews occur among the CAPI sample, and people in these HUs
may have characteristics that are more similar to CAPI respondents than to mail and CATI
respondents. Since the non-interview factors (NIF1 and NIF2) are applied to all HUs
interviewed by any mode, there may be some mode-related non-interview bias that has not been
compensated for. The mode bias factor ensures that the total weights in the cells defined by a
cross-classification of selected characteristics are the same as if the weight of non-interview HUs
had been assigned only to CAPI HUs, but distributes the weight across all respondents to reduce
the effect on the variance of the resulting estimates.

11-4

The first step in the calculation of the mode bias non-interview factor (MBF) is to calculate an
intermediate factor referred to as the mode non-interview factor (NIFM). NIFM is not used
directly as a weighting factor. The cross-classification cells are defined within county by
building type and tabulation month. Only HUs interviewed by CAPI and non-interviews are
placed in the cells, not mail and CATI interview HUs. If a cell contains fewer than 10
interviewed HUs, it is collapsed with an adjoining month until the collapsed cell has at least 10
interviewed HUs. Cells with no non-interviews are never collapsed regardless of size. The
mode non-interview factor (NIFM) for a cell is:
NIFM = (total HU weight after VMS of CAPI interviewed occupied and temporarily
occupied HUs, and non-interviewed HUs) / (total HU weight after VMS of CAPI
interviewed occupied and temporarily occupied HUs).
All CAPI interviewed occupied and temporarily occupied HUs get this mode non-interview
factor. HUs whose interviews were completed by mail or CATI, vacant HUs, and deleted HUs
get a factor of 1.0. Non-interviews get a factor of 0.0. The product NIFM and all previous
factors are used in the next step only.
Next a cross-classification table is defined within each estimation area by tenure (HU owned,
rented, or temporarily occupied), tabulation month, and marital status of the householder
(married/widowed or single). All occupied and temporarily occupied interviewed HUs are
placed in their cells. If a cell has fewer than 10 interviewed HUs, the cells with the same tenure
and month are collapsed across all marital statuses. If there are still fewer than 10 interviewed
HUs, the cells with the same tenure are collapsed across all months. The mode bias factor
(MBF) for each cell is then calculated as:
MBF = (total weight after NIFM of interviewed occupied and temporarily occupied HUs)
/(total weight after NIF2 of interviewed occupied and temporarily occupied HUs).
All interviewed occupied and temporarily occupied HUs are adjusted by this mode bias factor
and the remaining HUs get the factor 1.0. The weight for each HU following the calculation of
the mode bias factor is the base weight times the product of all factors except the mode noninterview factor.

11.4 2005 ACS WEIGHTING - HOUSING UNIT AND POPULATION CONTROLS
This stage of weighting forces the ACS total HU and person weights to conform with estimates
from the Census Bureau’s IPEs. The ACS estimates are based on a probability sample and will
vary from their true population values due to sampling and non-sampling error (Chapter 12 and
15). In addition, we can see from the formulas for the adjustment factors in the previous two
sections that they will also vary based on the combination of interviewed and non-interviewed
HUs in each tabulation month. The IPE program of the Census Bureau annually produces
estimates of the population by sex, age, race, and Hispanic origin and total HUs for each county
in the United States as of July 1. As part of the process of calculating person weights for the
ACS, estimates of totals by sex, age, race, and Hispanic origin are controlled to be equal to IPEs

11-5

by county. There are two reasons for this: (1) to reduce the variability of the ACS HU and
person estimates; and (2) to reduce bias due to undercoverage of HUs and people within HU in
household surveys like the ACS. The bias that results from missing these HUs and people is
partly corrected by using these controls (Alexander, et al., 1997). However, biases due to
differences between the interviewed and missed people within control cells will remain.
Production of IPEs for Puerto Rico is limited to population totals by municipio and the sex-age
distribution at the commonwealth level. Therefore, for the PRCS, estimates of totals by
municipio, sex, and age for the commonwealth are controlled to be equal to the IPEs. Currently,
there are no HU controls available for Puerto Rico. HU controls are being developed based on
information from the Puerto Rico Master Address File and field outcome codes.
The assignment of final weights involves the calculation of three factors based on the use of HU
and population controls. The first adjustment involves the use of independent HU estimates. A
second and separate adjustment relies on the use of independent population estimates. The final
adjustment is implemented to achieve consistency between the ACS HU estimates and the HU
estimates from the IPEs program.
Calculation of First Housing Unit Control Factor
Both HU and population estimates used as controls have a reference date of July 1. If person
weights are controlled to the IPEs as of July 1, it is also logical that HUs are controlled to their
IPEs to achieve a consistent relationship between the two totals.
The first housing control factor is employed to adjust the estimated number of ACS HUs by
estimation area to agree with their IPEs. This factor (HUF1) is simply:
HUF1 = (IPE HU estimate) / (total HU weight of interviewed occupied, interviewed
temporarily occupied, and vacant HUs).
All HUs (including deleted HUs) except non-interviews are adjusted by the first HU control
factor or HUF1.
Calculation of Person Post-Stratification Factor
The next step in the weighting process is to assign weights to persons. Each person in an
interviewed occupied HU is assigned an initial weight equal to the HU weight after HUF1 is
applied. Single-year person estimates are controlled to the IPEs for that year by sex, age, race,
and Hispanic origin combinations by estimation area. Race and Hispanic origin are combined to
define six unique race-ethnicity groups consistent with the groups used in weighting the Census
2000 long form. These groups are created by crossing Non-Hispanic with the five major single
race groups. The race groups are White, Black, American Indian and Alaskan Native, Asian,
and Native Hawaiian or Pacific Islander. The sixth group is Hispanic. The assignment of a
single major race to a person can be complicated because people can identify themselves as
being of multiple races. People responding with either multiple races or “Other race” are

11-6

included in one of the six race-ethnicity groups for estimation purposes only. Subsequent ACS
tabulations are based on the full set of responses to the race question.
Initial estimates are obtained from the ACS sample for each of the weighting race-ethnicity
groups. The initial estimates are calculated based on the initial person weight. Estimates from
the IPEs are also available for each weighting race-ethnicity group. The initial sample and IPEs
are tested against a set of criteria. The criteria require a minimum of 10 sample people and the
ratio of the population control to the initial estimate to be less than 3.5 for each weighting raceethnicity group2. If there are weighting race-ethnicity groups that do not satisfy these
requirements, they are further collapsed until all groups satisfy the collapsing criteria. See
Asiala (2004) for further details.
1. If the requirements are not met for all non-Hispanic groups combined, then all weighting
race groups are collapsed together.
2. If the requirements are not met for Hispanics, then the Hispanic group is collapsed with
the largest non-Hispanic, non-White group.
3. For each non-Hispanic, non-White group, if the requirements are not met, then it is
collapsed with the largest (prior to collapsing) non-Hispanic, non-White group.
4. If the largest collapsed non-Hispanic, non-White group still does not meet the
requirements, then it is collapsed with the surviving non-Hispanic, non-White groups in
this order - Black, American Indian and Alaskan Native, Asian, and Native Hawaiian
and Other Pacific Islander, until the requirements are met.
5. If all non-Hispanic, non-White groups have been collapsed together but the collapsed
group still does not meet the requirements, then the group is collapsed with all nonHispanic groups together.
Within each collapsed weighting race-ethnicity group the persons are placed in sex-age cells
formed by crossing sex by thirteen age categories: 0-4, 5-14, 15-17, 18-19, 20-24, 25-29, 30-34,
35-44, 45-49, 50-54, 55-64, 65-74, and 75+ years. These cells are also collapsed, if necessary, to
meet the same requirement of a minimum sample size and a ratio of less than 3.5. The goals of
the collapsing scheme are to keep children age 0-17 together whenever possible by first
collapsing across sex within the first three age categories. In addition the collapsing rules keep
men age 18-54 together, women age 18-54 together, and seniors 55+ together, by collapsing
across age. For each resultant cell a person post-stratification factor (PPSF) is calculated as:
PPSF = (IPE) / (total initial person weight).
This factor is applied to all persons in the cell so that their weights become the product of their
initial weight and this factor.

This is done to reduce the effect of large weights on the variance of the estimates.

11-7

Single-year estimates are produced for geographic areas of at least 65,000 population, including
incorporated places, for which IPEs are also published annually. Since population controls are
applied at the estimation area level, occasionally the ACS estimate of total population for a large
place within an estimation area may be “far enough” from its IPE to cause confusion among data
users. To avoid these anomalies, methodologies are being investigated that control HU person
weights to total population for places of at least 65,000 population within estimation areas.
Calculation of Final Housing Unit Factors
Prior to the calculation of person weights, each HU had a single weight and the sum of these
weights by estimation area was equal to the intercensal HU estimates. The set of person weights
in a HU is used to define a new HU weight with this property. In each interviewed occupied HU
we identify a principal person as either the wife in a household where both the husband and wife
are present or the reference person (one of the persons who rents or owns the HU) otherwise.
The principal person factor (PPF) for a HU is the person post-stratification factor of the principal
person. Its use is based on the assumption that principal persons are missed only if their entire
household is missed, not because of a failure to list them on the questionnaire, which happens for
potential reference persons. So their person post-stratification factors give an indication of
undercoverage for households whose principal persons have the same demographic
characteristics, and applying the principal person factor adjust for the resultant bias. In
particular, single males, particularly those 18-34 years of age, tend to have a lower response rate
than females and married males (Hainer, et. al., 1988). Thus, husbands in married-couple
households tend to get larger person post-stratification factors than their wives to make up for
the single males. To prevent over-representation of these households, the person poststratification factor of the wife is used. Vacant HUs get a principal person factor of 1.0 because
they have no person weights.
A second HU control factor (HUF2) is calculated in a similar manner to HUF1 with the only
difference being the use of the HU weights after the principal person factor has been applied to
them.
HUF2 = (IPE HU estimate) / (total HU weight after principal person factor of
interviewed occupied, interviewed temporarily occupied, and vacant HUs).
This weighting procedure results in two separate sets of weights, one for HUs and one for
persons residing within HUs. This is not uncommon for household surveys but has the
shortcoming that a single set of weights cannot be used to produce all estimates. In particular,
the sum of reference person weights and HU weights for all occupied HUs are not necessarily
the same, although it is logical that they should be. New procedures are being researched to
achieve consistency that logically exists between a variety of estimates, for example, estimates
of householders and occupied HUs.

11-8

11.5 WEIGHTING FOR SINGLE-YEAR ESTIMATES OF TOTAL POPULATION
Estimates from the 2006 ACS will include data from both HUs and GQ facilities because people
living in GQ facilities are included in the ACS sample for the 2006 data collection year. A
research project was undertaken to address the issue of developing an adequate method for
single-year weighting including GQ residents when simultaneously controlling person estimates
to IPEs by demographic characteristics at the county level. The research compared four
alternative options for controlling GQ persons separately or in combination with HU persons to
help address this issue. Results of this research show that it is feasible to weight the GQ data at
the state level and combine those results with the weighting of the household population at the
county level to produce adequate estimates of the total population for all levels of aggregation.
The methodology for controlling HU and GQ person estimates will not affect the prior weighting
steps. These steps are described in the previous section for HUs. Fewer steps are needed for GQ
facilities because all their data are collected by personal visit and tabulation month is not used in
any adjustments of GQ weights. The base weight is the reciprocal of the sampling rate and will
be equal to 40 for all people sampled in GQ facilities. In the small GQ stratum, there are
sometimes more than 15 people in a GQ facility, and in the large GQ stratum, sometimes the
initial number of people selected for a group is larger than 10. These situations occur when the
GQ facility has more people than expected. In these cases, a sub-sample of the residents is taken
so that only 10 will be eligible for interview. The weight of these 10 people must be increased
equally by a GQ sub-sampling factor to account for those not selected for the sub-sample.
A non-interview factor is calculated to account for the GQ residents who do not complete an
interview. This will occur in a single step where the non-interview cells will be defined by
combinations of GQ types, and research will determine the most appropriate combinations. If a
cell contains fewer than 10 people it is collapsed with an adjoining combination until the cell has
at least 10 people. The GQ non-interview factor (GQNF) is similar to the non-interview factors
applied to non-interviewed HUs and weights the interviewed GQ persons in these adjustment
cells up to the total weights of the interviewed and non-interviewed GQ persons by cell.

11.6 MULTI-YEAR ESTIMATES
Weighting for the combined HU and GQ multi-year estimates will be similar to that used to
produce single-year estimates. The annual samples corresponding to the estimation period will
be combined together and all the weighting steps will be performed similar to the single-year
weighting. The main difference is that the ACS population and HU estimates are controlled to
the average of the current IPEs for the multi-year estimation period.
The weighting will be implemented based on the geographic definition of the estimation areas as
of the last year of the estimation period. That is, all period estimates will be produced for a
consistent set of geographic areas. Additional details of multi-year estimation will be included
in future versions of this document.

11-9

11.7 REFERENCES
Adeshiyan, S. (1998), “A Study of the Weighting Adjustment Procedures for the American
Community Survey”, Proceedings of the Survey Research Methods Section, American Statistical
Association, pp. 178-183, 1998.
Albright, K.(2001), “2000 American Community Survey (ACS) Comparison County Replicate
Factors,” Internal U.S. Census Bureau Memorandum to C. Alexander from A. Navarro, May 23,
2001.
Albright, K. (2002), “Effects of Housing Unit Controls on Survey Estimates”, Proceedings of the
Survey Research Methods Section, American Statistical Association, pp. 13-18, 2002.
Alexander, C., Dahl, S. and Weidman, L. (1997), “Making Estimates from the American
Community Survey”, Proceedings of the Social Statistics Section, American Statistical
Association, pp. 88-97, 1997, .
Alexander, C. and Wetrogan, S. (2000), “Integrating the American Community Survey and the
Intercensal Demographic Estimates Program”, Proceedings of the Survey Research Methods
Section, American Statistical Association, pp.295-300, 2000.
Asiala, M.. (2004), “Specifications for Weighting the ACS 2003 HU Sample (ACS-W-6B),”
Second Draft, Internal U.S. Census Bureau Memorandum to L. McGinn through R. Singh,
Washington, DC, 2004.
Dahl, S. (1998a), “Weighting the 1996 and 1997 American Community Surveys,” Presented at
American Community Survey Symposium, 1998.
Dahl, S. (1998b), “Weighting the 1996 and 1997 American Community Surveys,” Proceedings
of the Survey Research Methods Section, American Statistical Association, pp.172-177, 1998.
Fay, R. and Train, G. (1995), “Aspects of Survey and Model-Based Postcensal Estimation of
Income and Poverty Characteristics for States and Counties,” Proceedings of the Section on
Government Statistics, American Statistical Association, pp. 154-159, 1995.
Gage, L. (2004), “Comparison of Census 2000 and American Community Survey 1999-2001
Estimates, San Francisco and Tulare Counties, California.,” 2004,
.
Hainer, P., Hines, C., Martin, E., and Shapiro, G. (1988), "Research on Improving Coverage in
Household Surveys," in Proceedings of the Fourth Annual Research Conference, Washington,
DC, U.S. Bureau of the Census, pp 513-539, 1988.
Hough, G. C. Jr., and Swanson, D.A. (2004), “The 1999-2001 American Community Survey and
the 2000 Census Data Quality and Data Comparisons, Multnomah County, Oregon,” 2004,
.

11-10

Love, S., Dalzell, D. and Alexander, C. (1995), “Constructing a Major Survey: Operational Plans
and Issues for Continuous Measurement,” Proceedings of the Survey Research Methods Section,
American Statistical Association, pp.584-589, 1995.
Salvo, J., Lobo, P., and Calabrese, T. (2004), “Small Area Data Quality: A Comparison of
Estimates, 2000 Census and the 1999-2001 ACS, Bronx, New York Test Site,” 2004,
.
Starsinic, M. (2005), “American Community Survey: Improved Reliability for Small Area
Estimates,” Paper Presented at the American Statistical Association Annual Meeting, 2005.
Tersine, A. and Asiala, M. (2002), “Alternative Oversampling Options for Low Mail Response
Areas in the American Community Survey,” Proceedings of the Survey Research Methods
Section, American Statistical Association, pp. 3442-3446, 2002,
.
U.S. Census Bureau (2004a), “Accuracy of the Data (2003),” pp. 6-9, Washington, DC, 2004,
.
U.S. Census Bureau (2004b), “State and County Housing Unit Estimates,” Washington, DC,
2004, .
U.S. Census Bureau (2004c), “Population Estimates: Concepts,” Washington, DC, 2004,
.
U.S. Census Bureau (2003), “Technical Documentation: Census 2000 Summary File 4,”
Washington, DC, 2003, .
Van Auken, P.M., Hammer, R.B., Voss, P.R., and Veroff, D.L. (2004), “American Community
Survey and Census Comparison, Final Analytical Report, Vilas and Oneida Counties,
Wisconsin; Flathead and Lake Counties, Montana,” 2004,
.
Weidman, L., Alexander, C., Diffendahl, G., and Love, S. (1995), “Estimation Issues for the
Continuous Measurement Survey,” Proceedings of the Survey Research Methods Section,
American Statistical Association, pp. 596-601, 1995,
.

11-11

Chapter 12.
Variance Estimation
12.1 OVERVIEW
The American Community Survey (ACS) is committed to providing its users with measures of
reliability with each published estimate. To accomplish this, all ACS estimates are accompanied
by either 90 percent confidence intervals or margins of error, both based on direct estimates of
sampling error. ACS variance estimates take into account sampling error and some types of
non-sampling errors, such as simple response variance and intra-interviewer assignment
correlation bias. The estimates of variance are computed using replication methods. Although
the variance estimates are not completely unbiased, the current variance estimation method
produces reliability measures that are accurate enough for practically all uses and analysis of
ACS data.
Additionally, generalized variance parameters are provided for users of the Public Use
Microdata Sample (PUMS) data to approximate standard errors for PUMS-tabulated estimates.
The generalized variance parameters are derived from the direct variance estimates.

12.2 ACS HOUSING UNIT AND PERSON ESTIMATES
Due to the systematic sample design and the ratio adjustments used in estimation, unbiased
estimates of the variance do not exist. The basic idea under a replication method is to choose a
fixed number of replicates (the ACS uses 80 replicates). For each replicate, the base weight for
each housing unit (HU) is multiplied by a replicate factor, and the entire weighting process is rerun, creating a new set of replicate weights. Replicate estimates are created by using the same
tabulation method as the original estimate, except applying each set of replicate weights instead
of the original weights. Finally, the replicate estimates and the original estimate are used to
compute the variance estimate.
The ACS will use the same variance estimation method as has been used in all its testing phasesthe successive differences replication (SDR) method. The theoretical framework for this method
was originally discussed by Wolter (1984) and extended by Fay and Train (1995). The SDR was
designed specifically to be used with systematic samples where the sort order is informative.
Applications of this method were developed to produce estimates of variance for the Current
Population Survey (CPS) (U.S. Census Bureau, 2002) and Census 2000 long-form sample
estimates (Gbur and Fairchild, 2002).
Replicate Factors
The replicate factor for each HU is determined from two rows of an 80x80 Hadamard matrix. A
Hadamard matrix H is a k-by-k matrix with all entries either 1 or -1, such that H'H = kI (that is,
the columns are orthogonal). Each of the 80 columns represents one of the 80 replicates. A

12-1

repeating sequence of 780 pairs of rows in the Hadamard matrix is assigned to each HU, in sort
order. For HU I and replicate r, the replicate factor is
f i ,r = 1 + 2 −1.5 a c(i ),r − 2 −1.5 a c(i + 1),r

where c(I) and c(I+1) are the rows of the Hadamard matrix assigned to the I-th HU, and ac(i),r and
ac(i+1),r are the cell values (either 1 or -1) from the Hadamard matrix in row c(i) and c(i+1) and
column r.
The replicate factors fi,r can take one of three approximate values: 1.7, 1.0, or 0.3.
With the replicate factors assigned, the replicate weights and estimates can be created. For any
estimate X0, with replicate estimates X1 through X80, its variance is estimated as the sum of
squared differences between each replicate estimate Xr and the full sample estimate Xo. The
formula is as follows:

( )

$ X0
Var

4
=
80

∑ (X
r =1

− X0

This equation holds not just for simple count estimates, but for any type of estimate – percent,
ratio, or median, among others.

12.3 SPECIAL CASES
There are certain cases, however, where the formula does not apply. The most important class
are estimates which are "controlled" and have their standard errors set to zero. These are the
estimates that are forced to equal intercensal estimates during the weighting process – total HUs
for estimation areas and the collapsed age, sex, or Hispanic origin estimates. In general, race
group estimates are not controlled because the categories used in the weighting (see Chapter 11)
do not match the tabulation groups published (because of multiple race responses and the “Some
Other Race” category. Information on the final collapsing of the raking matrix's margins is
passed from the weighting process to the variance estimation process to identify estimates that
are controlled. This is done for all geographic areas. Standard errors for those estimates are set
to zero. Another special case is zero estimated counts of people, households, or HUs. For these
cases, as well as for percents with zero estimates, a model-based estimate of the standard error is
assigned. The model-based estimate is a function of the state or national average of final
weights. Once the standard errors have been created, 90 percent margins of error and 90 percent
confidence bounds are produced. These are the measures of reliability that are presented along
with each published ACS estimate. The "Accuracy of the Data" document (U.S. Census Bureau,
2005) accompanying the 2004 ACS data release, provides more information on variance special
cases, margins of error, and confidence bounds. It also includes formulas to allow users to
approximate standard errors for sums, differences, proportions, and ratios of published ACS
estimates.

12-2

12.4 2006 ACS AND BEYOND
For each year, the group quarters (GQ) and HU samples are selected independently, so the
groups of GQ sample persons – all the persons from a small GQ of maximal size 15, the subsample of 10 persons from a small GQ with more than 15 persons, or each separate group of ten
selected from a large GQ – are assigned to the replicates separately from HUs. Then the
replicate factors, replicate estimates, and replication variances are calculated in the same manner
as described in section the previous section.
A similar replication methodology will be used for the three- and five-year variance estimates.
An additional factor that needs to be taken into account is that separate sub-universes of HUs
and small GQ facilities are in sample each year and a sub-universe is repeated only every fifth
year. This is to prevent HUs from being selected in sample more than once every five years, and
introduces a small amount of negative correlation to the ACS estimates.

12.5 GENERALIZED VARIANCE ESTIMATES
Data users of microdata or PUMS files tabulate the data for an unlimited number of
characteristics. A presentation of standard errors for all possible characteristic estimates is just
impossible. At this point, replicate weights are not provided on the PUMS files. Instead,
generalized variance parameters are provided which allow users to produce approximate
standard errors for any estimate they may create using the PUMS data. This is a simple yet very
effective way to provide users with a mechanism to obtain an approximate standard error for
numerous characteristic estimates.
The generalized variance parameters are calculated using one of three weighted regression
models on groups (by characteristic) of direct variance estimates. Standard errors for almost all
counts and proportions of persons, HUs and households are approximated using design factors:
⎛
Y$ ⎞
SE Y$ =& 12
. × DF × F * ×Y$ ⎜ 1 − ⎟
N⎠
⎝

()

. × DF ×
SE ( p$ ) =& 12

F*
× p$ (1 − p$ )
B

where DF is the appropriate design factor based on the topic of the estimate or proportion, F* is
a factor based on the fraction of the universe in the PUMS sample (F* = 142 for the 2000-2004
ACS test, and is yet to be determined for the 2005 PUMS), N is the size of the geographic area,
and B is the base of the proportion.
Standard errors for the total number of households as well as aggregates (totals of values other
than persons, HUs or households, such as travel time to work or income) use a three-parameter
(a-b-c) model:

12-3

()

SE Y$ =& 12
. × a + bY$ + cY$ 2

Standard errors for means, medians, per capita amounts, and ratios (other than proportions) use a
two-parameter (a-b) model:

()

SE Y$ =& 12
. × a + b × Log ( N )
where Log is the natural logarithm, and N is the denominator of a mean, per capita amount or
ratio, or the universe count for a median.
More detailed instructions on calculating the standard errors, as well as the parameter tables, are
included in the PUMS Accuracy of the Data document (U.S. Census Bureau, 2004b).
For the 2000-2004 ACS, generalized variance parameters were calculated at a national level.
With the 2005 data release, generalized variance function parameters will be provided at the
state level. The PUMS generalized variance parameters will be computed for the 2005 data
release, and will be updated periodically in the future, but not necessarily on an annual basis.

12.6 CONFIDENCE INTERVALS
Once the standard errors have been calculated, margins of error and/or confidence intervals are
produced. The confidence intervals or margins of error are measures of reliability presented
along with each published ACS estimate. A sample estimate and its estimated standard error
may be used to construct confidence intervals about the estimate. These intervals are ranges that
will contain the average value of the estimated characteristic that results over all possible
samples, with a known probability.
For example, if all possible samples that could result under the ACS sample design were
independently selected and surveyed under the same conditions, and if the estimate and its
estimated standard error were calculated for each of these samples, then approximately 90
percent of the intervals from 1.65 times the estimated standard error below the estimate to 1.65
times the estimated standard error above the estimate would contain the average result from all
possible samples. The lower and upper bounds presented in the published ACS products are the
bounds based upon a 90 percent confidence interval.
Some products provide a margin of error instead of upper and lower bounds. A margin of error
is the difference between an estimate and its upper or lower confidence bounds. Confidence
bounds can be created by adding the margin of error to the estimate (for an upper bound) and
subtracting the margin of error from the estimate (for a lower bound). All published margins of
error are also based on a 90 percent confidence level.
Users can calculate the standard error using the confidence bounds or the margin of error. If the
bounds are given, the margin of error is obtained using the following formula:

12-4

Margin of Error = max(upper bound - estimate, estimate - lower bound)
Using the margin of error (as published or calculated from the bounds), the standard error is
approximated as follows:
Standard Error = Margin of Error / 1.65
Here is an example of deriving the standard error from the published bounds:
The estimated number of males, never married is 33,290,195 in year from summary table
B12001 for the United States. The lower bound is 33,166,192 and the upper bound is
33,414,198.
Margin of Error = max(upper bound - estimate, estimate - lower bound)
Standard Error = Margin of Error / 1.65
Calculating the standard error using the bounds, we have:
Margin of Error = max(33,414,198 - 33,290,195, 33,290,195 - 33,166,192)
= 124,003
SE(33,290,195) = 124,003 / 1.65 = 75,153.
The "Accuracy of the Data" document accompanying the 2004 ACS data release (U.S. Census
Bureau, 2005) provides more information on variance estimation for special cases, margins of
error, and confidence bounds, as well as additional examples. It also includes formulas to
approximate standard errors for sums, differences, proportions and ratios of published ACS
estimates. For example, for the sum or difference of two published ACS estimates, an
approximation is given by the following formula:
SE ( Est 1 + Est 2 ) = SE ( Est 1 − Est 2 ) = SE ( Est 1 ) 2 + SE ( Est 2 ) 2

For two published products-multi-year profiles and ranking tables-we provide an indicator of
whether two estimates are statistically significantly different at the 90 percent confidence level.
That determination is made by initially calculating,
Z=

Est 1 − Est 2
SE ( Est 1 ) 2 + SE ( Est 2 ) 2

If Z < -1.65 or Z > 1.65 then the difference between the estimates is significant at the 90 percent
level. Determinations of statistical significance are made using unrounded values of the standard
errors, so users may not be able to achieve the same result on their own using the standard errors
derived from the rounded and published confidence bounds or margins of error.

12-5

12.7 REFERENCES
Fay, R. and Train, G. (1995), “Aspects of Survey and Model-Based Postcensal Estimation of
Income and Poverty Characteristics for States and Counties,” Proceedings of the Section on
Government Statistics, American Statistical Association, 1995.
Gbur, P. and Fairchild, L. (2002), “Overview of the U.S. Census 2000 Long Form Direct
Variance Estimation,” Proceedings of the Section on Survey Research Methods, American
Statistical Association, 2002.
Navarro, A. (2001), “2000 American Community Survey (ACS) Comparison County Replicate
Factors,” Internal U.S. Census Bureau Memorandum, Washington, DC, 2001.
U.S. Census Bureau (2002), "Current Population Survey: Technical Paper 63RV - Design and
Methodology," Washington, DC, 2002, .
U.S. Census Bureau (2003), “Technical Documentation: Census 2000 Summary File 4,”
Washington, DC, 2003.
U.S. Census Bureau (2004a), “Accuracy of the Data (2003),” Washington, DC, 2004,
.
U.S. Census Bureau (2004b), “PUMS Accuracy of the Data (2004),” Washington, DC, 2004,
.
U.S. Census Bureau, (2005), “Accuracy of the Data (2004)”, Washington, DC, 2005,
.
Wolter, K. (1984), “An Investigation of Some Estimators of Variance for Systematic Sampling,”
Journal of the American Statistical Association, 79.

12-6

Chapter 13.
Preparation and Review of Data Products
13.1 OVERVIEW
American Community Survey (ACS) data products include the tables, reports, and files that the
Census Bureau publishes containing estimates of population and housing characteristics from the
ACS. These products cover geographic areas within the United States and Puerto Rico. Tools
such as the Public Use Microdata Sample (PUMS) files, which enable data users to create their
own estimates, are also data products.
ACS data products will continue to meet the traditional needs of those who used the decennial
census long-form sample estimates. However, as described in Section 14.2, the ACS will provide
data products that are more current than the once-in-ten-years estimates available from the census
long form, an especially important advantage toward the end of a decade.
Most surveys of the population have sufficient samples to support the release of data products for
only the nation, states, and, possibly, a few sub-state areas. Because the ACS is a very large
survey collecting data continuously in every county, products can be released for many different
types of geographic areas, including many smaller geographic areas, such as counties, townships,
and census tracts. Thus, geography is an important topic for all ACS data products. It is the first
section in this chapter.
The first step in data products processing is defining the topics and characteristics to be covered.
Once the initial set of characteristics is determined, it must be reviewed by the Census Bureau
Disclosure Review Board (DRB) to ensure that the responses of any particular individual,
household, or establishment are kept confidential. Based on the results of that review, the
specifications of the products may be revised. The DRB may also require that the microdata files
be altered in certain ways. In addition, the DRB may place restrictions on the population size of
the geographic areas for which these estimates are published. These activities are collectively
referred to as disclosure avoidance. See Section 13.6 for a detailed discussion of disclosure
avoidance.
The actual processing of the data products cannot begin until all the response records for a given
year (or multiple years) are edited and imputed in the data preparation and processing phases
(Chapter 10), the final weights are determined (Chapter 11), and disclosure avoidance techniques
are applied. Using the weights, the sample data are tabulated for a wide variety of characteristics
according to the pre-determined content. These tabulations are done for the geographic areas that
have sample size that is sufficient to support statistically reliable estimates, with the exception of
five-year period estimates available for small geographic areas down to the tract and block group
levels. In the case of PUMS data, which are a subset of the full sample data, a different set of
processes creates the PUMS files.
13-1

After the estimates are produced and verified for correctness, Census Bureau subject matter
analysts review them. When the estimates have passed the final review, they are released to the
public. Similarly, once the PUMS data have been reviewed and accepted by the subject matter
analysts, the PUMS files are released to the public.
While the 2005 ACS sample is limited to the housing unit (HU) population, it has been expanded
to include Puerto Rico, so products released in 2006 (for sample year 2005) will include singleyear period estimates for the HUs and their population in both the United States and Puerto Rico.
Population residing in group quarters (GQ) facilities will be added starting with the 2006 sample
year. Therefore, the ACS and PRCS estimates published in 2007 (sample year 2006) will cover
the entire resident population.
In 2008, the Census Bureau will begin releasing products based on three years of ACS samples,
2005 through 2007. Then, in 2010, the Census Bureau will release the first products based on
five years of consecutive ACS samples (2005 through 2009). The schedule for the annual release
of single-year and multi-year products is discussed in Chapter 14. Since several years of samples
form the basis of these multi-year products, reliable estimates can be released for much smaller
geographic areas than is possible in single-year products. This chapter describes the data
products based on a single year of ACS sample. Future revisions will address the details of the
multi-year products.
In addition to regularly scheduled data products, other data products may be requested by
government agencies, private organizations and businesses, or individuals. To accommodate such
requests, the Census Bureau operates a custom tabulations program for the ACS on a fee basis.
These tabulations requests are reviewed by the DRB to assure protection of confidentiality before
release.
Chapter 14 describes the dissemination of the data products discussed in this chapter, including
display of products on the Census Bureau’s website and topics related to data file formatting.
The remainder of this chapter is divided into seven sections. Section 13.2 discusses the
geographic areas for which ACS estimates are published and relevant geographic issues. Sections
13.3 through 13.7 deal with the definition of the data products, their generation, and, finally, their
review and clearance. Section 13.8 discusses the ACS program under which data users can
request special data products (custom tabulations).

13.2 GEOGRAPHY
The Census Bureau strives to provide products for the geographic areas that are most useful to
users of those data. For example, the Census Bureau presents data summaries for the nation’s
many legal and administrative entities, including states, American Indian and Alaska Native
(AIAN) areas, counties, minor civil divisions (MCDs), incorporated places, congressional
districts, and voting districts. The Census Bureau also provides data for a variety of other
geographic entities. In cooperation with state and local agencies, it identifies and delineates
13-2

geographic entities referred to as “statistical areas.” These include regions, divisions, urbanized
areas (UAs), census county divisions (CCDs), census designated places (CDPs), census tracts,
and block groups. The data user community, composed of individuals, businesses, and agencies
at all levels of government, each with somewhat different needs, can then select the geographic
entity or set of entities that most closely represent their geographic area of interest.
“Geographic summary level” is a term used by the Census Bureau to designate the different
geographic levels or types of geographic areas for which data are summarized. Examples include
the entities described above, such as states, counties, and places (the Census Bureau term for
cities and towns). Information on the types of geographic areas for which the Census Bureau
publishes data is available at .
Single-year period estimates of ACS data are published annually for recognized legal,
administrative, or statistical areas with a population of 65,000 or more. The population sizes are
based on the latest Census Bureau population estimates. Multi-year period estimates based on
three successive years of ACS samples are published for legal, administrative, and statistical areas
of 20,000 or more population. If a geographic area met this threshold in a previous year during
the multi-year period but dropped below it for the current year, it will continue to be published as
long as the population does not drop more than five percent below the threshold. Multi-year
period estimates based on five successive years of ACS samples will be published for all legal,
administrative, and statistical areas down to the block-group level regardless of population size.
The PRCS will provide estimates for legal, administrative and statistical areas in Puerto Rico.
Starting in 2006, single-year period estimates from the 2005 sample will be published for all of
Puerto Rico as well as for the municipios and other areas with populations of 65,000 or more.
Table 13-1 provides the approximate number of geographic areas for which estimates will be
published for some of the main geographic summary levels for the single-year, three-year, and
five-year data products. For example, the table shows that 761 counties in the United States will
receive single-year period estimates, while three-year products will be available for those 761
counties plus another 1,050 counties that have a population between 20,000 and 65,000. The
Census Bureau will release five-year products for all counties in the United States.

13-3

Table 13-1. Major Geographic Summary Levels To Be Published in Single-Year, Three-Year, and Five-Year
Data Products

Description of Geographic
Summary Level1

Single-Year Data
Products
Number of
Geographic Areas
with a Population of
65,000 or More2

Three-Year Data
Products
Number of
Geographic Areas
with a Population of
20,000 or More2

Five-Year Data
Products
Total Number of
Geographic
Areas 2

1
1
1
United States
Census Regions
4
4
4
Census Divisions
9
9
9
States
51
51
51
Counties
761
1,811
3,141
Minor Civil Divisions
97
592
16,536
Places
476
1,983
25,161
American Indian and Alaska
15
41
768
Native Areas
Metropolitan, Micropolitan,
561
905
923
and Consolidated Statistical
Areas
Congressional Districts
436
436
436
School Districts
879
3,290
14,505
Census Tracts
--65,443
Block Groups
--208,790
1
1
1
Puerto Rico
Municipios
12
65
78
Zonas Urbanas,
9
20
225
Communidads
Census Tracts
--861
Block Groups
--2,477
1
See glossary for definitions of geographic terms.
2
The numbers shown in the table are based on the latest sources available and may change slightly.

The ACS publishes estimates every year for hundreds of sub-state areas, many of which will
undergo boundary changes due to annexations, detachments, or mergers with other areas 1 . Each
year, the Geography Division of the Census Bureau, working with the state and local
governments, updates its files to reflect these boundary changes. Minor corrections to the
location of the boundary can also occur as a result of the Census Bureau’s ongoing MAF/TIGER
Enhancement Project. The ACS estimates must reflect these legal boundary changes. To
accomplish this goal, all ACS estimates are based on Geography Division files that show the
1

The Census Bureau conducts the Boundary and Annexation Survey (BAS) each year. This survey collects
information on local governments and federally recognized American Indian areas. The information collected
includes the correct legal names, type of government, legal actions that resulted in boundary changes, and up-to-date
boundary information. The BAS uses a fixed reference date of January 1st of the BAS year. In years ending in 8, 9,
and 0, all incorporated places, all minor civil divisions, and all federally recognized tribal governments are included in
the survey. In other years, only governments at or above various population thresholds are contacted each year.
More detailed information on the BAS can be found at .

13-4

geographic boundaries as they existed on January 1st of the sample year or at the beginning of the
final year of data collection for a multi-year interval in the case of multi-year data products.

13.3 DEFINING THE DATA PRODUCTS
For the 1999 through 2002 sample years, the ACS detailed tables were designed to be comparable
with Census 2000 Summary File 3 in order to allow comparisons between data from Census 2000
and the ACS. However, when Census 2000 data users indicated certain changes they would like
to see in many of these tables, ACS managers saw the years 2003 and 2004 as an opportunity to
revise the ACS products based on users’ advice.
Once a preliminary version of the revised suite of products had been developed, the Census
Bureau asked for feedback on the planned changes from data users (including other federal
agencies) via a Federal Register Notice (Fed. Reg. #3510-07-P). The Notice requested comments
on current products and on proposed new products, particularly, the basic concept of each product
and its usefulness to the data users. The data users provided a wide variety of comments, leading
to modification of the set of planned products.
The ACS managers determined the exact form of the new products in time for their use in 2005
for the ACS data release for sample year 2004. This schedule allowed data users time to become
familiar with the new products and to provide comments well in advance of the data release for
the 2005 sample. Improvements to these products will continue, especially when multi-year
period estimates are available.

13.4 DESCRIPTION OF THE DATA PRODUCTS
The ACS data products can be divided into two broad categories: aggregated data products and
the PUMS, which is described in Section 13.5.
Data for the ACS are collected from a sample of HUs as well as the GQ population and are used
to produce estimates of the actual figures that would have been obtained by interviewing the
entire population using the same methodology. The aggregated data products contain the
estimates from the ACS survey responses. Each estimate in these products is created using the
sample weights from respondent records that meet certain criteria. For example, the 2005 ACS
estimate of people under the age of 18 in Chicago is calculated by adding the weights from all
respondent records from interviews completed in 2005 for people who reside in Chicago and who
are under 18 years old. Groups of estimates organized to show a complete distribution of values
for a characteristic are referred to as tables.
This section provides a description of each of these aggregated products. Each product described
in this section is available as single-year period estimates, and most of them will be available as
three-year and five-year period estimates as well. Chapter 14 provides more detail on the actual
appearance and content of each of these products. Information on the new data products and the
13-5

changes to the existing products, including several downloadable documents, are also available at
.
The data products described here contain all the estimates planned for release each year, including
those from multiple years of data, such as the 2005-2007 products. The release rules described in
Section 13.6 will prevent certain single-year and three-year period estimates from being released
if they do not meet ACS requirements for statistical reliability.
Aggregated Data Products
Base Tables. The base tables provide basic distributions of characteristics. In the past these
tables were known as detailed tables. They are the foundation upon which other data products are
built. These tables display estimates and the associated lower and upper bounds of the 90 percent
confidence interval. The base tables include demographic, social, economic, and housing
characteristics, and provide single-year, three-year, or five-year period estimates for the nation
and the states, as well as counties, towns, and other small geographic entities such as census tracts
and block groups.
In the case of multi-year period estimates based on monetary characteristics (for example, median
earnings), inflation factors are applied to the data to create estimates that reflect the dollar values
at the end of the final year of the time interval.
The Census Bureau will continue to maintain a high degree of comparability between ACS base
tables and Census 2000 sample-based data products. In addition, characteristics not measured in
the Census 2000 tables will be included in the new ACS base tables. The new base tables product
includes over 400 tables covering a wide variety of characteristics, and another 300 race and
Hispanic-origin iterations covering about 40 of the key characteristics. In addition to the tables
on characteristics, 81 tables are included that summarize allocation rates from the data edits for
many characteristics. These tables provide measures of data quality by showing the extent to
which responses to various questionnaire items were complete. Altogether, almost 900 separate
base tables will be provided.
Data Profiles. Data profiles are high-level reports containing estimates for demographic, social,
economic, and housing characteristics. For a given geographic area, the data profiles include
many important measures, such as distributions for sex, age, type of household, race and Hispanic
origin, school enrollment, educational attainment, disability status, veteran status, language
spoken at home, ancestry, income, poverty, physical housing characteristics, occupancy and
owner/renter status, and housing value. The data profiles include the lower bound and upper
bound of a 90 percent confidence interval for each estimate. As we accumulate more sample
years, a form of data profiles allowing the comparison of the latest sample year’s estimates with
those from earlier years will also be published. These profile reports will contain the results of a
statistical significance test for each earlier year’s estimate compared to the current year estimate.
This test result tells the data user if the previous year’s estimate is significantly different (at a 90
percent confidence level) from the estimate of the current year.
13-6

Narrative Profiles. Narrative profiles cover only the current sample year. These narratives are
easy-to-read, computer-produced profiles that put into words main topics from the data profiles
for the general-purpose user. This is the only ACS product with no standard errors accompanying
the estimates.
Subject Tables. This is a new data product. Subject tables are similar to the Census 2000 quick
tables, and like the quick tables, they are derived from base tables. Quick tables and subject
tables are pre-defined tables with frequently requested information on a single topic for a single
geographic area. However, subject tables contain more detail than the Census 2000 quick tables
or the ACS data profiles. In general, they present distributions for a few key universes, such as
the race groups and people in various age groups, among others. The estimates for these
universes are displayed as whole numbers. The distribution that follows is generally displayed in
percents. Subject tables also contain other measures, such as medians, and aggregates where
appropriate, and include the imputation rates for relevant characteristics. Over 40 topic-specific
subject tables will be released each year. These topics are described at the website mentioned
above.
Ranking products. Ranking products contain ranked results of many important measures across
geographic areas. They are produced only as single-year products based on the current sample
year. The ranked results among the geographic areas for each measure are displayed in three
ways – charts, tables, and tabular displays that allow testing of statistical significance. For
example, one of the characteristics ranked is “Mean Travel Time to Work of Workers 16 Years
and Over Who Did Not Work at Home.”
The rankings show 81 selected measures. The data for ranking products are pulled directly from
a base table or a data profile line across geographic areas at the geographic summary level for
which a characteristic is being ranked. Currently, most characteristics are ranked only at the state
level.
Thematic Maps. Thematic maps are similar to ranking tables. They show mapped values for
geographic areas at a given geographic summary level. They have the added advantage of
visually displaying on a map the geographic variation of a key characteristic (referred to as
themes). Initially, thematic maps will be displayed for state-level results. An example of a
thematic map would be a map showing the percent of population 65 years and older by state.
Selected Population Profiles. This new product provides certain characteristics from the data
profiles for a specific race or ethnic group (for example, Alaska Natives) or some other selected
population group (for example, people aged 60 years and older). The ACS will provide selected
population profiles every year for most of the Census 2000 Summary File 4 iteration groups.
Selected population profiles were introduced on a limited basis starting in the fall of 2005 using
the 2004 sample. This product will be expanded when it can reflect the full sample. Groups too
small to get a selected population profile report based on a single year of sample data will get a
selected population profile based on the three-year or five-year accumulation of sample data.
More details on these profiles can be found in Hillmer (2005), which includes a list of selected
race, Hispanic origin, and ancestry populations.
13-7

13.5 PUBLIC USE MICRODATA SAMPLE
Microdata are the individual records that contain information collected about each person and
HU. Public Use Microdata Sample (PUMS) files are extracts from the confidential microdata
taken in a manner that avoids disclosure of information about households or individuals. These
extracts cover all of the same characteristics contained in the full microdata sample files. See
Chapter 14 for information on data and file organization for PUMS.
The only geography shown on a PUMS file is the Public Use Microdata Area (PUMA). PUMAs
are special non-overlapping areas that partition a state, and each PUMA contains a population of
about 100,000. State governments drew the PUMA boundaries at the time of Census 2000. They
were used for the Census 2000 sample PUMS files and are known as the “5 percent PUMAs.”
The Census Bureau plans to create PUMS data only as a single-year data product. However,
users will be able to combine annual PUMS files to create larger samples in each PUMA
(covering a longer period of time). For more information on these geographic areas go to
.

13.6 GENERATION OF DATA PRODUCTS
Following conversations with users of census data, the subject matter analysts in the Census
Bureau’s Housing and Household Economics Surveys Division (HHES) and Population Division
(POP) specify the organization of the ACS data products. This includes the logic to be used to
calculate every estimate in every data product and the exact textual description to be associated
with each estimate. Starting with the 2006 data release, only limited changes to these
specifications are expected. Changes to the data product specifications in 2006 and later years
must preserve the ability to compare estimates from one year to another and must be operationally
feasible. Changes to these specifications must be made no later than late winter of each year to
ensure that the revised specifications are finalized by spring of that year.
After the edited data with the final weights are available, generation of the data products begins
with the creation of the base tables data product with the single-year period estimates. Refer to
Chapters 10 and 11 for information on the production of edited data and final weights. The
programming teams of the American Community Survey Office (ACSO) generate these
estimates. Another staff within ACSO verifies that these estimates comply with the specifications
from the subject matter analysts. Both the generation and the verification activities are
automated.
The single-year data products are released on a phased release schedule throughout the summer.
Because of this release schedule, the data products must be generated and verified as soon as the
edited and weighted data are available for the sample year, usually in early June. The single-year
data products are spread across four thematic data releases beginning in mid-August and
extending to late October or early November. The four thematic releases are demographic and
13-8

social characteristics; economic characteristics; housing characteristics; and characteristics of
selected race, Hispanic origin, and ancestry groups. The release schedule for the three-year and
five-year data products is discussed in Chapter 14.
A distinguishing feature of the ACS data products system is that standard errors are calculated for
all the estimates and are released with them in tables. Subject matter analysts also use the
standard errors in their internal review of the estimates.
Disclosure Avoidance
Title 13 of the United States Code (U.S.C.) is the basis for the Census Bureau’s policies on
disclosure avoidance. Title 13 says “Neither the Secretary, nor any other officer or employee of
the Department of Commerce may make any publication whereby the data furnished by any
particular establishment or individual under this title can be identified...” 2 The Census Bureau’s
DRB reviews all data products planned for public release to ensure adherence to Title 13
requirements governing the release of data based on confidential information provided by the
ACS respondents. The DRB may insist on applying rules (the disclosure avoidance techniques)
that could result in the suppression of certain measures for small geographic areas. More
information about the DRB and its policies can be found at
.
Once the plans are finalized for the ACS data products, the DRB reviews them to assure that
confidentiality of respondents was protected.
In order to meet Title 13 U.S.C. for the ACS, the Census Bureau uses statistical methodologies
during tabulation and data review to ensure that individually identifiable data will not be released.
To that end, several statistical methodologies are used:
Swapping. The main procedure used for protecting Census 2000 tabulations was data
swapping. It was applied to both the short form (100 percent) data and the long form
(sample) data independently. It is also currently being used to protect ACS tabulations.
In each case, a small percent of household records is swapped. Pairs of households that
are in different geographic regions are swapped across those geographic regions. The
selection process for deciding which households should be swapped is highly targeted to
affect the records with the most disclosure risk. Pairs of households that are swapped
match on a minimal set of demographic variables. All data products (tables and
microdata) are created from the swapped data files.
For PUMS data the following techniques are employed in addition to swapping:
Top-coding is a method of disclosure avoidance in which all cases in or above a certain
percentage of the distribution are placed into a single category.
2

Title 13, United States Code, § 9(a).

13-9

Geographic population thresholds prohibit the disclosure of data for individuals or HUs
for geographic units with population counts below a specified level.
Age perturbation, that is, modifying the age of household members, is required for large
households (households containing ten people or more) due to concerns about
confidentiality.
Detail for categorical variables is collapsed if the number of occurrences in each
category does not meet a specified national minimum threshold.
For more information on the disclosure avoidance techniques see Section 5, “Current disclosure
avoidance practices” at .
In addition, the DRB may determine that certain tables are so detailed that other restrictions are
required to assure that there is sufficient sample to support the table without revealing
information on individual respondents. In such instances, a restriction may be placed on the size
of the geographic area for which the table can be published. Current DRB rules require that base
tables containing more than 100 detailed cells not be released below the census tract level.
The data products released in the summer of 2006 for the 2005 sample will cover only the HU
population of the United States and Puerto Rico. Data collection from the population living in
GQ facilities began in January 2006. Thus, the data products released in the summer of 2007 will
cover the resident population of the United States and Puerto Rico. Thus, most estimates for
person characteristics covered in the data products will be affected by this expansion from the
household population to the resident population of the United States. For the most part, the actual
characteristics will remain the same, and only the description of the population group will change
from HU population to resident population. Future updates of this report will detail will changes
for GQ facilities.
Data Release Rules
Even with the population size thresholds described earlier, in certain geographic areas some very
detailed tables might include estimates whose reliability is unacceptable. Data release rules,
based on the statistical reliability of the survey estimates, will be used starting with the 2005 ACS
data released in the summer of 2006. These release rules apply only to the single-year and threeyear data products.
The main data release rule for the ACS tables works as follows. Every base table consists of a
series of estimates. If more than half the estimates are not statistically different from 0 (at a 90
percent confidence level), then the table fails. Each estimate is subject to sampling variability
that can be summarized by its standard error. Dividing the standard error by the estimate yields
the coefficient of variation (CV) for each of the estimates. (If the estimate is 0, a CV of 100
percent is assigned.) To implement this requirement for each table at a given geographic area,
CVs are calculated for each of the table’s estimates, and the median CV value is determined. If
13-10

the median CV value for the table is less than or equal to 61 percent, the table passes for that
geographic area; if it is greater than 61 percent, the table fails. Tables that are too sparse will fail
this test. In that case, the table will not be published for that geographic area.
Whenever a table fails, a simpler table that collapses some of the detailed lines together can be
substituted for the original, more detailed table. The data release rules are then applied to the
simpler table. If it passes, the simpler table is released. If it fails, none of the estimates for that
particular table is released for this geographic area. These release rules are applied to single-year
period estimates and multi-year period estimates based on three years of sample data. No data
release rules are applied to the estimates based on five years of sample data.

13.7 DATA REVIEW AND ACCEPTANCE
After the editing, imputation, data products generation, disclosure avoidance, and application of
the release rules have been completed, subject matter analysts perform a final review of the ACS
data and estimates before they can be released. Data review and acceptance are the last
opportunity to ensure that there are no missing values, obvious errors, or other anomalies in the
data.
Each year, the ACS staff and the subject matter analysts carry out the generation, review, and
clearance of all ACS estimates in approximately eight weeks. At a minimum, subject matter
analysts subject their data to a specific multi-step review process before it is cleared and released
to the public. Because of the short time available to review such a large amount of data, an
automated review tool (ART) was developed to facilitate the process.
ART is a computer application that enables subject matter analysts to detect statistically significant
differences in estimates from one year to the next using several statistical tests. The initial version
of ART was used for two successive years of data review, 2003 and 2004. It featured pre-designed
reports as well as ad hoc, user-defined queries for hundreds of estimates and for about 350
geographic areas. While ART proved helpful, several issues emerged. The ART workgroup
defined a new version of ART, called ART II, to address these issues. ART II has been used by
the analysts since June 2005. The improvements of ART II are characterized by the ability to
work on much larger data sets, a wider range of capabilities, and faster response time to user
commands.
In addition to ART II, the ACSO staff, together with the subject matter analysts, has developed
two automated tools to facilitate documentation and clearance for the steps required in the data
review process: the edit management and messaging application (EMMA), and, the PUMS
management and messaging application (PMMA). Both of these tools are used to track the
progress of analysts’ review activities. These tools enable analysts and managers to see the current
status of files under review and determine which review steps can start.

13-11

13.8 CUSTOM DATA PRODUCTS
As explained above, the Census Bureau offers a wide variety of general-purpose data products
from the ACS designed to meet the needs of the majority of data users. They contain pre-defined
sets of data for standard census geographic areas.
For users whose data needs are not met by the general-purpose products, the Census Bureau offers
customized special tabulations on a cost-reimbursable basis through the ACS custom tabulation
program. Custom tabulations are created by tabulating data from ACS edited and weighted data
files. Custom tabulation projects vary in size, complexity, and cost depending on the needs of the
sponsoring client.
Each custom tabulation request is reviewed in advance by the DRB to ensure that confidentiality
of respondents’ information is protected. The requestor may be required to modify the original
request to meet the Census Bureau’s disclosure avoidance requirements.
For more detailed information on the ACS Custom Tabulations program go to
.

13.9 REFERENCES
Hillmer, Douglas (2005), “An Overview of Selected Population Profiles—A New Data Product
Proposed for the American Community Survey,” Presentation given at the Meetings of the Census
Advisory Committee on the African American Population, the American Indian and Alaska Native
Populations, the Asian Population, the Hispanic Population, and the Native Hawaiian and Other
Pacific Islander Populations, Washington, DC, 2005.

13-12

Chapter 14.
Data Dissemination and User Assistance
14.1 OVERVIEW
The American Community Survey (ACS) has been tabulating and publishing single-year
estimates for specific areas with populations of 250,000 or more since 2000. ACS data products
and supporting documentation are released in several series and several places on the Internet.
The primary website for ACS data is the American FactFinder (AFF) website. Supporting
documentation is found on the ACS website and the File Transfer Protocol (FTP) website. This
chapter deals with single-year data products. Future versions of this document will include a
discussion of multi-year data products.
In 2005, the ACS expanded to its full sample size. The ACS will release data annually for areas
with populations of 65,000 or more beginning in summer 2006. For smaller areas, it will take
three or five years, depending on their population size, to accumulate a large enough sample to
produce releasable estimates. Once those data are collected, the Census Bureau will release
tabulations annually based on three-year period data for areas with populations of 20,000 or
more, and five-year period data for areas as small as census tracts and block groups (see time
line and additional materials at ).
Billions of dollars are distributed by federal agencies among states, tribal governments, and
population groups based on their social and economic profiles. In the past, the statistics for
funding formulas and projects, such as the location of services and program planning,
evaluation, and improvement, have come in large part from the long form sample of the
decennial census. As the ACS is completely phased in, ACS data products will provide updated
versions of many of the long form data products from Census 2000. Beginning in 2010, the
decennial census will no longer include a long form sample, ACS data products will provide
high quality, updated statistics every year for comparisons of the demographic, social, economic,
and housing characteristics of areas and population groups. The ACS statistics will also show
directions and levels of change over time, and relative differences between areas and population
groups. ACS data products will continue to meet the traditional needs of those who used the
decennial census sample statistics and will provide statistics that are more current than the onepoint-in-time statistics available from the census long form sample.

14.2 SCHEDULE
Data Release Timetable
By 2010, information on social, economic, and housing characteristics once available every 10
years will be available annually for all areas through the ACS. Then, each year thereafter, these

14-1

geographic areas will get new estimates of these characteristics based on the five-year interval
ending in the latest completed sample year.
Figure 14-1 summarizes the data products release schedule. In 2006, the first set of single-year
estimates will be released for specific areas with 65,000 and greater populations. These areas
will continue to receive single-year estimates annually. Beginning in 2008, data collected over a
three-year period (2005-2007) will be released for areas with at least 20,000 population. A new
three-year product will be released in 2009 based on data collected in 2006-2008. In 2010, the
first five-year products will be released based on data collected in 2005-2009. The five-year
products are produced for areas down to census tracts and block groups. Once three-year and
five-year products are produced, annual updates follow, as indicated by the figure below.
Figure 14-1. Data Release Schedule

14.3 PRESENTATION OF TABLES
American FactFinder
The AFF website contains data in the form of maps, tables, and reports from a variety of
censuses and surveys organized into data sets. AFF lists these data sets by program areas and
survey years. Currently, AFF contains data for the 1990 and 2000 Decennial Censuses, the
2000-2004 ACS, the Population Estimates Program, the 1997 and 2002 Economic Censuses, and
the Annual Economic Surveys.
The main page of the AFF website contains links that provide access to information stored in the
system. A banner is always at the top of the page, and always displays the same seven links:
14-2

main, search, feedback, FAQs, glossary, site map, and help. The left column provides links to
additional information about the AFF and other AFF options. Beginning with the 2004 release,
the AFF became the primary web access for ACS data. ACS data products include: base tables,
data profiles, narrative profiles, ranking charts, subject tables, selected population profiles, and
downloadable PUMS files. Chapter 13 contains detailed summaries of each of these products.
Figure 14-2 below is a screen shot of part of the Data Sets section of AFF. The image depicts
AFF’s organization.
Figure 14-2. The AFF Data Sets Page

For more information on ACS data on AFF, see the Quick Guide to 2003 ACS data in AFF
located on the following URL: .

14-3

Presentation of Base Tables
Figure 14-3 displays a sample image from AFF of a single-year table. The ACS table title
describes the variables in the table, and any combinations of them for which estimates are
presented and the universe. The data set indicates the year the data were collected and the type
of product. The table has two additional columns containing the lower and upper bounds for the
90 percent confidence interval of the estimate.
Figure 14-3. American FactFinder Base Table Presentation

14-4

Presentation of Data Profiles
Figure 14-4 displays only part of an image of a 2004 ACS data profile. The profiles appear as
html files on the AFF website. The data set indicates the year the data were collected and the
type of product. The profile has two additional columns containing the lower and upper bounds
for the 90 percent confidence interval produced of the estimate.
Figure 14-4. American FactFinder Data Profile Presentation

14-5

Presentation of Narrative Profiles
Figure 14-5 displays a section of a narrative profile. Narrative profiles include data profile
information displayed in a narrative format that summarizes data of on a wide array of subjects.
These narratives are easy-to-read, computer-produced profiles that put into words the main
topics from the tabular profiles for the general-purpose user.
Figure 14-5. American FactFinder Narrative Profile Presentation

14-6

Ranking Product Presentation
Ranking Chart. Figure 14-6 displays only part of a ranking chart. The characteristics in the
ranking tables and charts are displayed from highest to lowest. The dot represents the estimate,
and the 90 percent confidence intervals around each estimate are represented by the brackets.
Like the upper and lower bounds shown in the base tables, the “wings” or “arms” here represent
how close or far apart the bounds are around the estimate: the shorter the wing span, the lower
the variance on the estimate. The ranking tables and charts can be viewed and downloaded in
portable data format (PDF).
Figure 14-6. American FactFinder Ranking Chart Presentation

14-7

Ranking Table. Like the other tables, the ranking table shows the percentage estimate and the
associated lower and upper bounds of the confidence interval. The table is for a single
characteristic, and the geographic areas are ranked from highest to lowest. See Figure 14-7.
Figure 14-7. American FactFinder Ranking Table Presentation

14-8

Statistical Significance View of a Ranking Table. Users can look at the chart of statistical
significance to see how two geographic areas compare. The presence of a • indicates
geographies whose estimate is not statistically significant from the selected geography, while ••
indicates the selected geography. See Figure 14-8 for a chart of statistical significance.
Figure 14-8. American FactFinder Statistical Significance Presentation

14-9

Thematic Maps. Thematic maps on AFF are based on the state-level geographic ranking tables.
They have the added advantage of visually displaying on a map the geographic variation and
patterns of a key summary or derived measure. The shading intensity on a thematic map is
directly related to the value associated with the derived measure listed in the map’s legend under
data classes. Lighter shading is used for the lower derived measure values. As the shading
becomes darker and more intense, the derived measure’s values increase. Figure 14-9 depicts a
sample of a thematic map on AFF.
Figure 14-9. American FactFinder Thematic Maps Presentation

14-10

Presentation of Subject Tables
Figure 14-10 displays only part of a subject table. Subject tables are derived from the base
tables. They show more detail than is available in the data profiles. Generally, they present
distributions for a few key universes, with universes displayed as numeric estimates with margin
of error, but not upper and lower bounds. Subject tables display measures such as medians and
aggregates where appropriate, and include tables of imputation rates for relevant measures. The
AFF will contain about 40 summarized topic-specific subject tables.
Figure 14-10. American FactFinder Subject Table Presentation

14-11

Presentation of Selected Population Profiles
Selected Population Profiles provide the user with a ready-made report on a population group
and margin of error. Examples of selected population profiles include Native Hawaiians and
Other Pacific Islander Race Groups at the national or state level, children under 18, and the
elderly (60 and older). Figure 14-11 displays a portion of a selected population profile.
Figure 14-11. Sample Select Population Profile

14-12

Public Use Microdata Sample Files
Public use microdata sample (PUMS) files are data files that contain records of a sample of all
housing units that responded to the survey. PUMS files are available as comma delimited files
and as SAS datasets. For a more detailed description of the PUMS see Chapter 13. Figure 14-12
shows the layout of the PUMS main page on the AFF website.
Figure 14-12. AFF PUMS Main Page

14-13

ACS Website
The ACS website contains a wealth of information, documentation, and research papers on the
ACS (Figure 14-13 depicts the organization of the ACS website’s main page). The site contains
important metadata, including a set of subject definitions that includes more than 50 population
concept definitions and more than 40 housing concept definitions.
Figure 14-13. ACS Main Page

14-14

Documentation on the accuracy of the data is also included and provides information about the
sample design, confidentiality, sampling error, non-sampling error, and estimation methodology.
The site’s quality measures section displays detailed information and data on the quality of the
ACS estimates. Refer to Chapter 15 for more information on quality measures. The errata
section lists updates made to the data to correct errors. The geography section gives a brief
explanation of the Census Bureau’s geographic hierarchy, common terms, and specific
geographic areas presented in the ACS.
Figure 14-14 is a screen-shot of the ACS quality measures first page. Quality measures are
produced annually to accompany each ACS data release.
Figure 14-14. Presentation of the ACS Quality Measures Main Page

14-15

File Transfer Protocol Website
The FTP application is intended for users of census and ACS data. This site provides quick
access to data users who need to begin their analysis immediately upon data release.
Each state directory provides all files available for the identified state. Once downloaded, the
data are in Excel or PDF. No software is provided. Users of the FTP application can import the
files into the spreadsheet/database software of their choice for data analysis and table
presentation. Three types of data products are provided on this site: base tables, data profiles,
and ranking tables.
All of the data products (data profiles, ranking charts, and base tables) will also be available on
the Census Bureau’s FTP site. For example, the 2004 products are located on the following
URL:

Figure 14-15 displays the main page of the 2004 FTP site. The main directory contains
documentation describing the layout of the FTP site in the README file. The README file is
available in the main directory on the FTP server in 3 formats: MS Word, WordPerfect, and
text.

14-16

Figure 14-15. Main Page of 2004 FTP Site

14.4 EDUCATION AND TRAINING
Education and training will help data users, respondents, and the general public make the
transition from the decennial long form to the ACS. ACS staff conduct a wide range of activities
that are designed to inform and educate ACS stakeholders about the data. The staff makes
presentations, attends meetings, and conducts workshops for a variety of audiences, including
representatives of federal, state, local, and tribal governments, businesses, advocacy
organizations, a variety of national organizations, and the media.
General Outreach
The Census Bureau has developed the following tools for education and outreach:
•

ACS Alert. The ACS Alert is a periodic electronic newsletter that provides the latest
information on ACS products, research, and activities. It is disseminated via e-mail to
more than 5,000 census stakeholders in both the public and private sectors.

•

Congressional Tool Kits. In 2004, the Census Bureau supplemented its ongoing
congressional communication and education activities with a large-scale initiative
14-17

designed to prepare congressional offices for full implementation of the ACS. ACSO and
Congressional Affairs Office staff members visited the offices of each Member of
Congress (Washington, DC, states, and local districts) and delivered Congressional Tool
Kits that describe ACS operations and data products and provided other information
related to the ACS program.
The toolkits consist of binders with ACS data for the Member’s district or state, a
description of ACS operations, justifications for each question on the form, and other
information that Members of Congress may use to explain the ACS program to
constituents. The electronic versions of the binders are also available on CD-ROM and
on-line.
Versions of the toolkit are being developed for American Indian and Alaska Native
audiences and for local officials and businesses in Puerto Rico.
•

Staff of the Assistant Director for Marketing, Services and Customer Liaison Offices offer
standard and specialized training for audiences ranging from social service providers to
government officials. They answer data user queries, staff exhibit booths, and conduct
workshops for national, state, local, and tribal organizations. They also prepare and
distribute informational brochures, flyers, and electronic communications to keep
constituents informed of ACS developments. The State Data Center (SDC), Census
Information Center (CIC), and Federal-State Cooperative Program for Population
Estimates (FSCPE) networks, which are administered by the same directorate, are also
key communications resources for the ACS.

•

The Public Information Office (PIO) organizes and conducts news conferences and
special appearances for the Census Bureau. PIO also develops general and targeted
materials (for example, drop-in articles) for the news media.

•

The Field Division’s Partnership and Data Services Staff (located both at Census Bureau
headquarters and in the regional offices) and the Field Division’s Information Services
Staff conduct workshops, attend meetings, and develop and distribute materials to support
the ACS.

The Census Bureau is establishing an ACS education program. The first priority of this program
is to provide stakeholders with the information they need to use ACS data as they used decennial
census long form data in the past. The Census Bureau plans to develop separate modules or
training packages that will consist of PowerPoint slides, workbooks, and similar educational
media that will provide basic information about the ACS program. The Census Bureau also plans
to provide more advanced training for more experienced data users (for example, stakeholders
who used census data for programmatic applications in government or the private sector).
The Census Bureau also envisions the development of a reference manual that will help
stakeholders use ACS data appropriately. Currently the Census Bureau is coordinating the
production of an ACS User Guide for this purpose. The Census Bureau will add information on
14-18

ACS applications and case studies to the ACS website as they
become available.
Federal Agency Information Program
Federal agencies have special requirements for ACS education. Because of the methodological
differences between the ACS and the decennial census long form, federal agencies’ many and
varied uses of decennial census long-form data, and the need for early planning to evaluate the
effect of using ACS data for critical agency programs, the Census Bureau has adopted a proactive
approach to meeting federal agencies’ needs. As the ACS program has evolved, staff have
regularly consulted with federal agencies. In the mid-1990s, several colloquia were held with
federal agencies to explore how they would be affected by ACS data. The availability of annual
data from the ACS provides opportunities for improving programs that serve the public. It also
presents challenges, in that programs that are based on the “point-in-time” data from the use of
decennial census long form must now use multi-year estimates from the ACS.
Working relationships with federal agencies were established early in the ACS’s development.
The goal was to provide information on the continuous measurement concept and show how a
continuous measurement program differs from a once-a-decade long form data collection effort in
terms of the sample design, survey methods, operations, and data products. In all cases, federal
agencies were encouraged to discuss their concerns and questions about how their programs
would be affected with Census Bureau staff.
Some federal agencies demonstrated an early readiness to consider the detailed methodological
and design aspects of the continuous measurement plan and the implications for their agencies.
In a report published in 1996, the Department of Transportation (DOT) explored the impact of the
ACS on its programs. In 2003, the Department of Housing and Urban Development (HUD)
produced a comprehensive report with a detailed analysis of the opportunities, resource effects,
and HUD’s research needs for the timely data that the ACS provides.
During the spring of 2003, the Census Bureau started developing the ACS Federal Agency
Information Program (FAIP) to address the unique needs of each federal agency as it transitions
from using decennial-census long-form data to using ACS data. The goal of the program is to
have each interested federal agency identify one or more liaisons with whom Census Bureau staff
can work to identify transition issues. To launch the program, the Census Bureau sent letters to
more than 40 federal agencies inviting them to participate and identifying a contact person at the
Census Bureau with whom the liaisons could work.
The Census Bureau held a kick-off meeting for the FAIP on June 26, 2003 and started meeting
with federal agencies during the summer of 2003. In many cases, these meetings represented a
renewal of the commitment to work on ACS transition issues, since a number of meetings had
already taken place between federal agencies and the Census Bureau. The Census Bureau
developed a series of workshops customized for each agency. In April 2004, the first of these
ACS Federal Agency workshops was held with the DOT. Additional meetings and workshops
were held in 2004 with the U.S. Postal Service (USPS), the Equal Employment Opportunity
14-19

Commission (EEOC), the Department of Veterans Affairs, the Department of Education, and the
Department of Labor. In 2005 the Census Bureau met with the Department of the Interior, the
Department of Homeland Security (DHS), and a federal inter-agency group interested in using
ACS data to measure English proficiency. In early 2006, the Census Bureau held an ACS
workshop with HUD.
The long range objectives of the ACS FAIP are to:
•

Identify transition issues and challenges that affect the use of ACS estimates;

•

Provide technical assistance, information, or other resources, to assist federal agencies in
using the ACS statistics; and

•

Help federal agencies identify new ways to use the ACS so that they can get the maximum
benefits from the data.

The Census Bureau has produced three reports that describe progress on the FAIP. Copies of the
reports were provided to Members of Congress, the Government Accountability Office (GAO),
and the Office of Management and Budget (OMB).
The Census Bureau will continue to work with federal agencies on ACS transition issues. At the
same time, the Census Bureau looks forward to learning more about federal agencies’ needs from
a study currently underway at the National Academies of Sciences (NAS) concerning the most
effective ways to use ACS data.

14.5 REFERENCES
Griffin, D.H. (2004), “Continuous Measurement of Quality in the American Community Survey,”
Proceedings of the European Conference on Quality and Methodology in Official
Statistics (Q2004).
Hillmer, Doug (2005), “The American Community Survey Initial Briefing for the Federal
Reserve–Data Products Overview, April 7, 2005,” PowerPoint presentation by American
Community Survey Office, Washington, DC, 2005.
ORC Macro (2002), “The American Community Survey: Challenges and Opportunities for
HUD,” Paper prepared under contract for the U.S. Department of Housing and Urban
Development, Washington, DC, September 2002.
Torrieri, Nancy (2003-2004), “The American Community Survey Federal Agency Information
Program: Progress Reports I, II, and III,” Internal U.S. Census Bureau reports, Washington, DC,
2003-2004.

14-20

U. S. Census Bureau (2004), “2004 American Community Survey Data Products Document,”
Internal Census Bureau document, Washington, DC, 2004.
U. S. Census Bureau (2005), “American Community Survey Data User Training Guide,” Internal
Census Bureau document, Washington, DC, 2005.
U. S. Census Bureau, DADS Program Office (2005), “Quick Guide to 2003 ACS Data in
American FactFinder,” Washington, DC, January 6, 2005.
U. S. Department of Transportation, Bureau of Transportation Statistics (1996), “Implications of
Continuous Measurement for the Uses of Census Data in Transportation Planning,” Washington,
DC, April 1996.

14-21

Chapter 15.
Improving Data Quality by Reducing Non-Sampling Error
15.1 OVERVIEW
As with all surveys, the quality of the American Community Survey (ACS) data reflects how
well our data collection procedures address and reduce potential sources of non-sampling error,
including coverage error, non-response error, measurement error, and errors that may arise
during data capture and processing. Chapters 4 and 11 provide information regarding steps the
ACS takes to reduce sampling error while still managing costs.
There are four primary sources of non-sampling error (Groves, 1989, 1991):
•

Coverage error: the failure to give some units in the target population any chance of
selection into the sample, or giving a unit more than one chance of selection.

•

Non-response error: the failure to collect data from all units in the sample

•

Measurement error: the inaccuracy in responses recorded on survey instruments,
arising from:
•
the effects of interviewers on the respondents answers to survey questions,
•
the error due to respondents, from inability to answer questions, lack of requisite,
effort to obtain the correct answer, or other psychological or cognitive factors,
•
the error due to faulty wording of survey questionnaires, and
•
the error due to data collection mode effects

•

Processing error: the error introduced after the data are collected, arising from:
•
the error occurring while capturing data,
•
the error from the coding and classification of data, and
•
the error from the editing and item imputation of the data

This chapter identifies the operations and procedures designed to reduce these sources of nonsampling error and thus improve the quality of the data. In addition, this chapter includes tables
from the ACS Quality Measures website ()
that provide measures of non-sampling error. The ACS releases the annual survey estimates as
well as the measures provided on the ACS Quality Measures website at the same time each year
so that users can consider the data quality in conjunction with the survey estimates.

15.2 COVERAGE ERROR
All surveys experience some degree of coverage error. It can take the form of under-coverage or
over-coverage. Under-coverage occurs when units in the target population do not have a chance
15-1

of selection into the sample (for example, addresses not listed on the frame or people erroneously
excluded from a household roster). Over-coverage occurs when units or people have multiple
chances of selection (for example, addresses listed more than once on the frame, or people
included on a household roster at two different sampled addresses). In general, coverage error
can affect survey estimates if the characteristics of the individuals or units excluded or included
in error differ from the characteristics of those correctly in the frame. Over-coverage and undercoverage can sometimes be adjusted for as part of the post-stratification process (comparing
weights to external control totals).
The ACS uses the Master Address File (MAF) as its sampling frame. The ACS includes several
procedures for reducing coverage error in the MAF. See Chapter 3 for more details regarding
these procedures.
•

Twice a year, the Census Bureau receives the U.S. Postal Service (USPS) Delivery
Sequence File (DSF) and uses it to update the city-style addresses (that is, addresses
including a house number and street name rather than a rural route or post-office box) on
the MAF.

•

The ACS non-response follow-up operation provides ongoing address and geography
updates using the automatic listing and mapping instrument (ALMI).

•

The MAF includes address updates from special census operations.

•

The Community Address Updating System (CAUS) provides address updates as a
counterpart to the DSF updates that cover predominately rural areas where city-style
addresses are generally not used for mail delivery.

The ACS Quality Measures website presents tables with coverage rates as a measure of coverage
error in the ACS. National level coverage rates are calculated for Hispanics and for the five
major race categories for non-Hispanics – White, Black, American Indian and Alaska Native,
Asian, and Native Hawaiian and Other Pacific Islander. Coverage rates are calculated by sex at
the national and state levels. Data are available for every year since 2000.
The coverage rate is the ratio of the ACS population estimate of an area or group over the
population estimate for that area or group, multiplied by 100. The Census Bureau uses
independent data on births, deaths, immigration, and so on, to produce official population
estimates each year. The base for these independent estimates is the decennial census counts.
The numerator in the coverage rates is weighted to reflect the probability of selection into the
sample, sub-sampling for personal visit follow-up and adjusted for unit non-response. The
weight used for this purpose does not include post-stratification adjustments; that is, weighting
adjustments that make the weighted totals match external population estimates, since the control
totals serve as the basis for comparison for the coverage rates. The ACS, like all other household
surveys, corrects for potential over- or under-coverage by controlling to these official estimates
on specific demographic characteristics and at specific levels of geography. Refer to Chapter 11
for additional detail on weighting.
15-2

As the coverage rate for a particular subgroup drops below 100 percent, the weights of the
people in the subgroup need greater adjustment in the final weighting procedure to reach the
independent estimate. If the rate is greater than 100 percent, the ACS population estimates are
downweighted to match the independent estimates. A sample from the Quality Measures
website of the national-level coverage rates for 2000-2004 is shown in Figure 15-1.
Figure 15-1. ACS Quality Measures Website – Coverage Rates

15.3 NON-RESPONSE ERROR
There are two forms of non-response error: unit non-response and item non-response. Unit nonresponse results from the failure to obtain the minimum required data from a housing unit in the
sample. Item non-response occurs when respondents do not report individual data items or they
provide data considered invalid or inconsistent with other answers.
Surveys strive to increase both unit and item response thus reducing the possible bias introduced
into the survey estimates. Bias results from systematic differences between the non-respondents
and the respondents. Without data on the non-respondents, surveys-including ACS-cannot easily
measure differences between respondents and non-respondents. Instead the ACS attempts to
reduce the potential for bias by reducing the amount of unit and item non-response through
several different means.
•

Response to the ACS is required by law.

•

ACS survey operations include two stages of non-response follow-up: a computer–
assisted telephone interview (CATI) follow-up for mail non-respondents, and a
computer-assisted personal interview (CAPI) follow-up for a sample of remaining nonrespondents and non-mailable cases.

•

The mail operation implements a strategy suggested in research studies for obtaining a
high mail response rate (Dillman, 1978): a pre-notice letter, a message on the envelope
of the questionnaire mailing package stating that the response is “required by law”, a
postcard reminder, and a second mailing for non-respondents to the initial mailing.

15-3

•

The mailing package includes a frequently asked questions (FAQ) motivational brochure
explaining the survey, its importance, and its mandatory nature.

•

The questionnaire design reflects accepted principles of respondent friendliness and
navigation making it easier for respondents to understand what items apply to them, as
well as providing cues for a valid response at an item level (such as showing the format
for reporting dates, pre-filled ‘0’ to indicate reporting dollar amounts rounded to the
nearest whole number). See Appendix B.7 for a copy of the mail questionnaire.
Similarly, the automated instruments (CATI and CAPI) direct interviewers to ask the
appropriate questions, as well as provide functionality that facilitate valid responses. For
example, the instrument checks for values outside of the expected range and pops up a
window with a question for the interviewer to ask to make sure the reported amount
reflects the correct amount.

•

The questionnaire provides a toll-free telephone number for respondents who have
questions about the ACS in general or who need help in completing the questionnaire.

•

The ACS includes a telephone-edit follow-up (TEFU) interview with mail respondents
who either failed to respond to specific critical questions, or who indicated a household
size of five or more people. In regards to the latter group, the mail form only allows data
for five people, so the TEFU operation follows up to collect data for any additional
people.

•

ACS uses a permanent, professional CATI and CAPI interviewing staff trained in refusal
conversion methods.

•

Survey operations include providing support in other languages: a Spanish paper
questionnaire is available on request, a Spanish CATI/CAPI instrument, and CATI and
CAPI interviewers that speak Spanish (as well as other languages).

Unit Non-Response
The Census Bureau presents survey response rates on the ACS Quality Measures website as a
measure of unit non-response in the ACS. The survey response rate is the ratio of the estimate of
housing units interviewed after data collection is complete to the estimate of all units that should
have been interviewed. For the ACS, this means all interviews after mail, telephone, and
personal visit follow-up. Interviews include complete and partial interviews with adequate
information for processing.
To accurately measure unit non-response, the ACS uses data from the CATI and CAPI follow-up
operations to provide a weighted estimate of both the size of the universe of interview-eligible
cases and the total number of survey non-interviews; that is, all eligible units in CAPI follow-up
receive a weight used for estimating the total number of eligible units in the universe. Noninterview cases in CAPI also receive a weight for estimating the total number of non-interviews.
These weights reflect the probability of selection into the sample and sub-sampling for personal
15-4

visit follow-up. The estimate of the total number of eligible units becomes the denominator of
the unit response rate. As noted earlier, refer to Chapter 11 for more information about
weighting.
Figure 15-2 below provides an example of both national-level survey response rates and several
types of non-interview rates. Data are available on the website for every year since 2000 at the
national and state level.
Figure 15-2 also shows the percentage of cases that did not respond, broken out by the reason for
non-response. Those reasons include refusal, unable to locate the sample address, no one home
during the data collection period, temporarily absent during the interview period, language
problem, insufficient data (not enough data were collected to consider it a response), and other
(such as “sample address not accessible,” “death in the family,” as well as for cases not followed
up due to budget constraints, which last occurred in the winter of 2004).
Figure 15-2. ACS Quality Measures Website – Survey Response Rates

Item Non-Response
The ACS Quality Measures website includes data on item non-response. When respondents do
not report individual data items or they provide data considered invalid or inconsistent with other
answers, the Census Bureau imputes the necessary data. The Census Bureau uses imputation
methods that either use rules to determine acceptable answers or use answers from similar people
or housing units. The first of these two methods is known as "assignment," while the second is
referred to as "allocation." Assignment involves logical imputation where a response to one
question implies the value for a missing response to another question. For example, first name
can often be used to assign a value to sex. Allocation, on the other hand, involves using
statistical procedures to impute for missing values.
Figure 15-3 gives an example of the national-level allocation rates from 2000-2004 for a few
housing variables. Allocation rates are available for all published variables.
15-5

Figure 15-3. ACS Quality Measures Website – Item Allocation Rates

15.4 MEASUREMENT ERROR
All surveys experience some form of measurement error. Measurement error is the difference
between the recorded answer and the true answer. Measurement error may result in any mode of
data collection—vague or ambiguous questions easily misinterpreted by respondents, questions
that respondents cannot answer, interviewer actions such as the tone used in reading questions,
paraphrasing questions, leading the respondent to certain answers, as well as other similar
factors.
The ACS minimizes measurement error in several ways, some of which also help with reducing
non-response. Chapters 5 through 9 discuss in more detail the following methods or procedures
for reducing measurement error.
•

Per the Census Bureau Standard “Pretesting Questionnaires and Related Materials for
Surveys and Censuses (v 1.1)” ACS pretests new or modified content in all three modes
before introducing it into the ACS.

•

As noted in the previous section on non-response error, the ACS uses a questionnaire
design reflecting accepted principles of respondent friendliness and navigation.

•

Mail questionnaire respondents receive a questionnaire instruction booklet in their
mailing package that provides additional information on how to interpret and respond to
specific questions.

•

Respondents may also call the toll-free telephone questionnaire assistance (TQA) line
and speak with trained interviews for answers to general ACS questions as well as
answers to questions regarding specific items on the ACS questionnaire.

•

Differences among the mail, CATI, and CAPI questionnaires are reduced through
questionnaire and instrument design methods that reflect the strengths and limitations of
15-6

each mode of collection (for example, less complicated skip patterns on the mail
questionnaire, breaking up questions with long or complicated response categories into
separate questions for telephone administration, and including respondent flash cards for
personal visit interviews).
•

The CATI/CAPI instrument automates or directs skips through the instrument, only
showing interviewers questions appropriate for the person under discussion.

•

Training for the permanent CATI and CAPI interviewing staff covers reading the
questions as worded and answering respondent questions, including extensive roleplaying opportunities. As part of the training, all interviewers receive a manual
explaining each question in detail and providing detailed responses to questions often
asked by respondents.

•

Telephone interview supervisors and specially trained staff monitor CATI interviews and
provide feedback regarding verbatim reading of questions, recording of responses,
interaction with respondents, and so on.

•

Field supervisors and specially trained staff implement a quality reinterview program
with CAPI respondents to minimize falsification of data.

•

The CATI/CAPI instruments include a Spanish version and bilingual interviewers
provide language support in other languages as well.

Note that many of these methods are the same methods used to minimize non-response error.
Methods that make it easier for the respondent to understand the questions also increase the
chances the person will respond to the questionnaire.

15.5 PROCESSING ERROR
The final component of non-sampling error is processing error – the error introduced in the postdata collection process of taking the responses and turning them into published data. For
example, a processing error may occur in keying the data on the mail questionnaires. Miscoding
write-in responses, either clerically or by automated methods reflects another example of
processing error. Additionally, the degree to which imputed data differs from the truth also
reflects processing error, specifically imputation error.
Chapters 7 and 10 provide more detail for some of the methods used by the ACS for controlling
processing error.
•

Keying of mail questionnaires includes a quality control procedure to ensure the accuracy
of the final keyed data (see Chapter 7 for more detail).

•

Clerical coding includes a quality control procedure involving double coding of a sample
15-7

of the cases and adjudication by a third keyer.
•

Automated coding systems by design rely on manual coding by clerical staff to address
the most difficult or complicated responses.

•

Procedures for selecting one interview or return in the instance of multiple returns for an
address rely on a review of the quality of the data from each response, selecting the return
with the most complete data.

•

Questionnaires with insufficient data after completion of all three phases of data
collection (mail, CATI, and CAPI) do not continue in the survey processing but instead
receive a non-interview code and are accounted for in the weighting process.

•

Edit and imputation rules reflect the combined efforts and knowledge of subject matter
experts and experts in processing, including evaluation and subsequent improvements as
the survey progressed into full implementation.

•

Subject matter experts and survey experts complete an extensive review of the data and
data tables, comparing results with previous years’ data and other data sources.

15.6 REFERENCES
Dillman, Donald (1978), Mail and Telephone Surveys, New York, John Wiley and Sons, 1978.
Groves, Robert M. (1989), Survey Errors and Survey Costs, New York, John Wiley & Sons,
1978.
Groves, Robert M. (1991), “Measurement Error Across the Disciplines,” in Paul P. Biemer et al.
(Eds.), Measurement Errors in Surveys, A Wiley –Interscience Publication, 1991.

15-8

__ A mobile home or trailer
__ A one-family house detached from any
other house
__ A one-family house attached to one or
more houses
__ A building with 2 apartments
__ A building with 3 or 4 apartments
__ A building with 5 to 9 apartments
__ A building with 10 to 19 apartments
__ A building with 20 to 49 apartments
__ A building with 50 or more apartments
__ Boat, RV, van, etc.

H1. Which best describes this building? Include all
apartments, flats, etc., even if vacant.

1996-1998 QUESTION W ORDING

__ A mobile home
__ A one-family house detached from any
other house
__ A one-family house attached to one or
more houses
__ A building with 2 apartments
__ A building with 3 or 4 apartments
__ A building with 5 to 9 apartments
__ A building with 10 to 19 apartments
__ A building with 20 to 49 apartments
__ A building with 50 or more apartments
__ Boat, RV, van, etc.

Which best describes this building?
Include all apartments, flats, etc., even if
vacant.

A-1

The last category was modified because evaluation of the 1990 census data indicated
that respondents indicated “Other” when they were uncertain about the structure
classification of their unit. This design resulted in the proportion of units classified as
“Other” to be far larger than the number of units that were classified as “Boat, tent,
van, etc.” in 1980.

The words “or trailer” were deleted at the recommendation of the Manufactured
Housing Institute.

Two response categories differ from the 1990 census item. The first response category
was “A mobile home or trailer.” The last response category was “Other.”

In the first answer category, the words “or trailer” were deleted to match the Census
2000 item.

1999-2002 QUESTION W ORDING

February 2005

Unchanged

2003 - 2005 QUESTION W ORDING

CHANGES TO THE AMERICAN COMMUNITY SURVEY (ACS) QUESTIONNAIRES
(HOUSING QUESTIONS)

Appendix A.1 Changes to the American Community Survey (ACS) Questionnaires (Housing Questions)

_ _ (Month)

_ _ _ _ (Year)

H3. When did PERSON 1 (listed in the List of
Residents on page 2) move into this house
or apartment?

_ _ _ _ (Year)

__ 1999 or later
__ 1995 to 1998
__ 1990 to 1994
__ 1980 to 1989
__ 1970 to 1979
__ 1960 to 1969
__ 1950 to 1959
__ 1940 to 1949
__ 1939 or earlier

About when was this building first built?
__ 2000 or later
__ 1995 to 1999
__ 1990 to 1994
__ 1980 to 1989
__ 1970 to 1979
__ 1960 to 1969
__ 1950 to 1959
__ 1940 to 1949
__ 1939 or earlier

About when was this building first built?

__ 2005 or later
__ 2000 to 2004
__ 1990 to 1999
__ 1980 to 1989
__ 1970 to 1979
__ 1960 to 1969
__ 1950 to 1959
__ 1940 to 1949
__ 1939 or earlier

In 2003, the response categories were updated to include housing units built in
2000 or later.

____ (Month)

____ (Year)

When did PERSON 1 (listed in the List
of Residents on page 2) move into this
house, apartment, or mobile home?

A-2

The ACS response area includes “Month” to observe recent movers.

The words “or mobile home” were added to the question to be more inclusive of the
structure type. The question matches the Census 2000 question.

Unchanged

In 2005, the response categories were again updated, as follows:
Except for the first response category, where the ACS uses “1999 or later” and the
Census 2000 category is “1999 to 2000,” the ACS item matches the Census 2000 item. 2
About when was this building first built?
Research showed higher item response rates when categories were used rather than
write-ins.

The italicized instructions were dropped.

The response area was changed from a write-in space to response categories.

H2. About when was this building first built? If you 2
do not know the exact year, give your best estimate.

____ Months each year

c. How long does this household usually spend
at that residence?

____________________________

b. Where is that residence located?

__ No
__ Yes

H4a. Do all persons staying in this house or
apartment usually spend more than two
consecutive months of the year at another
residence?

A-3

This set of questions about household members’ other residence was changed. The
revised set of questions asks respondents to classify the sample residence when all of
the household members live somewhere else part of the year.

__ Primary residence?
__ Vacation home?
__ School residence?
__ Work residence?
__ Other - Specify ________________

c. Do you consider this house, apartment,
or mobile home, that uses the address on
the front cover, your -

__ All persons listed
__ Some persons - How many?
___ Persons - SKIP to the questions
for Person 1 on the
next page.

b. Of the persons listed on pages 2 and 3,
how many live somewhere else part of
the year?

__ Yes - SKIP to the questions for Person 1
on the next page
__ No -

28a. Do all of the persons listed on pages 2
and 3 live at this address year round?

In part c, the question and all answer categories were revised. In addition, the
answer category “Looking for permanent housing” was added.

In part b, the question was changed entirely.

In part a, the text “all of the persons listed on pages 2 and 3" is replaced with
“you or any member of this household”.

Question number changed

__ This is their permanent address
__ This is their seasonal or vacation address
__ To be close to work
__ To attend school or college
__ Looking for permanent housing
__ Other reason(s) - Specify
________________________________

c. What is the main reason members of this household are
staying at this address?

Months
__

b. How many months a year do members
of this household stay at this address?

__ Yes - SKIP to the questions for Person 1
on the next page
__ No -

25a. Do you or any member of this
household live or stay at this address
year round?

__ Yes
__ No

H7. Is there a business such as a store or barber
shop or a medical office on this property?

__ Yes
__ No

__ Less than 1 acre --SKIP to question 6
__ 1 to 9.9 acres
__ 10 or more acres

How many acres is this house or
mobile home on?

__ Yes
__ No

Is there a business (such as a store or
barber shop) or a medical office on
this property?

A-4

With the addition of parentheses, the 1999 - 2002 ACS item matches the 1990 census
item and the Census 2000 item.

The ACS monthly data collection necessitates that the reference period be “In the past
12 months” rather than “In 1999” for the Census 2000.

The question wording and response categories were changed to match the 1990 census
item and the Census 2000 item. This design is used primarily to identify farm
residences, and allows us to adapt to changes in the farm definition.

__ None
__ $1 to $999
__ $1,000 to $2,499
__ $2,500 to $4,999
__ $5,000 to $9,999
__ $10,000 or more

IN THE PAST 12 MONTHS, what were
the actual sales of all agricultural
products from this property?

The question wording was changed, and the second response category was modified to
match the Census 2000 item.

H6. IN THE PAST 12 MONTHS, were the sales of all 5
agricultural products from this property $1,000
or more?

__ Less than 1 acre --SKIP to question H7
__ 1 to less than 10 acres
__ 10 or more acres

H5. Is this house or mobile home on --

Unchanged

____ Bedroom(s)
OR
____ None

H9. How many bedrooms are in this house or
apartment? Count the number of bedrooms that
you would list if this house or apartment were for
sale or for rent.

____ Room(s)
__ 1 room
__ 2 rooms
__ 3 rooms
__ 4 rooms
__ 5 rooms
__ 6 rooms
__ 7 rooms
__ 8 rooms
__ 9 or more rooms

How many rooms are in this house,
apartment, or mobile home? Do NOT count
bathrooms, porches, balconies, foyers, halls, or
half-rooms.

__ No bedroom
__ 1 bedroom
__ 2 bedrooms
__ 3 bedrooms
__ 4 bedrooms
__ 5 or more bedrooms

How many bedrooms are in this house,
apartment, or mobile home; that is, how
many bedrooms would you list if this
house, apartment, or mobile home were
on the market for sale or rent?

A-5

The ACS question uses the phrase “are in” rather than the phrase “do you have” in the
Census 2000 item.

The response area was changed from a write-in space to response categories.

The italicized instructions were incorporated into the question.

The words “or mobile home” were added to the question to be more inclusive of the
structure type.

The ACS question uses the phrase “are in” rather than the phrase “do you have” in the
Census 2000 item.

The response area was changed from a write-in space to response categories to match
the Census 2000 item. Research showed that there was no significant difference in the
item response rates when categories were compared with write-in entries for number of
bedrooms.

The words “or mobile home” were added to the question to be more inclusive of the
structure type.

7
H8. How many rooms are in this house or
apartment? Do NOT count bathrooms, porches,
balconies, foyers, halls, half-rooms, or utility rooms.

Unchanged

__ Yes, has all three facilities
__ No

H11. Does this house or apartment have complete
kitchen facilities; that is 1) a sink with piped
water, 2) a stove or range, and 3) a
refrigerator?

__ Yes, has all three facilities
__ No

H10. Does this house or apartment have complete
plumbing facilities; that is 1) hot and cold
piped water, 2) a flush toilet, and 3) a bathtub
or shower?
__ Yes, has all three facilities
__ No

Does this house, apartment, or mobile
home have COMPLETE plumbing facilities;
that is, 1) hot and cold piped water, 2) a
flush toilet, and 3) a bathtub or shower?

__ Yes, has all three facilities
__ No

Does this house, apartment, or mobile
home have COMPLETE kitchen facilities;
that is, 1) a sink with piped water, 2) a
stove or range, and 3) a refrigerator?

A-6

The ACS question uses the phrase “Does this ... have” rather than the phrase “do you
have” in the Census 2000 item.

The term “stove” replaced the term “Cookstove” used in the 1990 census item.

The word “COMPLETE” was capitalized for emphasis.

The words “or mobile home” were added to the question to be more inclusive of the
structure type.

The ACS question uses the phrase “Does this ... have” rather than the phrase “do you
have” in the Census 2000 item.

The word “COMPLETE” was capitalized for emphasis.

The words “or mobile home” were added to the question to be more inclusive of the
structure type.

Unchanged

__ Yes
__ No

H16. Is there a telephone in this house or
apartment?

— Yes
__ No

H15. Does this house or apartment have a central
heating system; that is, one system that
heats all or most of the rooms?

__ Yes, a central air conditioning system
__ Yes, one or more individual room units
__ No

H14. Does this house or apartment have air
conditioning?

__ Yes, connected to a public sewer
__ No, connected to septic tank or cesspool
__ No, uses other means

H13. Is this house or apartment building connected
to a public sewer?

__ A public system such as a city water
department or private company?
__ An individual drilled well?
__ An individual dug well?
__ Some other source, such as a spring, creek,
river , cistern, etc.

H12. Does this house or apartment building get
water from -

__ Yes
__ No

Is there telephone service available in this
house, apartment, or mobile home from
which you can both make and receive
calls?

A-7

The 1999 - 2002 ACS item matches the Census 2000 item. On previous ACS
questionnaires, the question adopted the focus of the 1990 Census question that asked
a respondent to report if a telephone was in the unit. The Federal Communications
Commission recommended the change to telephone service availability because it is a
better measure of households that have service available to initiate and receive calls.

The words “or mobile home” were added to the question to be more inclusive of the
structure type.

This item was deleted starting with the 1999 ACS because it is not a Census 2000 item.

Deleted for 1999

This item was deleted starting with the 1999 ACS because it is not a Census 2000 item.

Deleted for 1999

This item was deleted starting with the 1999 ACS because it is not a Census 2000 item.

Deleted for 1999

This item was deleted starting with the 1999 ACS because it is not a Census 2000 item.

Deleted for 1999

Unchanged

__ Gas: from underground pipes serving the
neighborhood
__ Gas: bottled, tank, or LP
__ Electricity
__ Fuel oil, kerosene, etc.
__ Coal or coke
__ Wood
__ Solar energy
__ Other fuel
__ No fuel used

H18. Which FUEL is used MOST for heating this
house or apartment?

_____ Vehicle(s)
OR
__ None

H17. How many automobiles, vans, and trucks of
one-ton capacity or less are kept at home
for use by members of this household?

__ None
__ 1
__ 2
__ 3
__ 4
__ 5
__ 6 or more

How many automobiles, vans, and trucks
of one-ton capacity or less are kept at
home for use by members of this
household?

__ Gas: from underground pipes serving the
neighborhood
__ Gas: bottled, tank, or LP
__ Electricity
__ Fuel oil, kerosene, etc.
__ Coal or coke
__ Wood
__ Solar energy
__ Other fuel
__ No fuel used

Which FUEL is used MOST for heating this
house, apartment, or mobile home?

A-8

The 1999 - 2002 ACS item matches the Census 2000 item.

The words “or mobile home” were added to the question to be more inclusive of the
structure type.

With the change in the response area from a write-in space to response categories, the
1999 - 2002 ACS item matches the 1990 census item and the Census 2000 item. The
1990 census item had a response category of “7 or more”. The Department of
Transportation and the Census Bureau agreed that the response category of “6 or more”
is sufficient.

Unchanged

$_____.00 (Past 12 months)
OR
__ Included in rent or in condominium fee
__ No charge or these fuels not used

d. IN THE PAST 12 MONTHS, what was the cost
of fuel oil, wood, kerosene, coal, etc. for this
house or apartment? If you have lived here less
than 12 months, estimate the cost.

$_____.00 (Past 12 months)
OR
__ Included in rent or in condominium fee
__ No charge

c. IN THE PAST 12 MONTHS, what was the
cost of water and sewer for this house or
apartment? If you have lived here less than
12 months, estimate the cost.

$_____.00 (Last month)
OR
__ Included in rent or in condominium fee
__ No charge or gas not used

b. LAST MONTH, what was the cost of gas for
this house or apartment?

$_____.00 (Last month)
OR
__ Included in rent or in condominium fee
__ No charge or electricity not used

Past 12 months’ cost - Dollars
$_____.00
OR
__ Included in rent or in condominium fee
__ No charge or these fuels not used

d. IN THE PAST 12 MONTHS, what was the
cost of oil, coal, kerosene, wood, etc.,
for this house, apartment or mobile
home? If you have lived here less than 12
months, estimate the cost.

Past 12 months’ cost - Dollars
$_____.00
OR
__ Included in rent or in condominium fee
__ No charge

c. IN THE PAST 12 MONTHS, what was the
cost of water and sewer for this
house, apartment, or mobile home? If
you have lived here less than 12 months,
estimate the cost.

Last month’s costs - Dollars
$_____.00
OR
__ Included in rent or in condominium fee
__ Included in electricity payment
entered above
__ No charge or gas not used

b. LAST MONTH, what was the cost of
gas for this house, apartment, or
mobile home?

Last month’s cost - Dollars
$_____.00
OR
__ Included in rent or in condominium fee
__ No charge or electricity not used

LAST MONTH, what was the cost of
electricity for this house,
apartment, or mobile home?

A-9

The 1990 census item for the cost of water did not include sewage fees.

In part b. the response category “Included in electricity payment entered above”was
added for respondents who receive one bill for both costs.

The ACS differs from the census item. The ACS asks for monthly costs of electricity
and gas, while the census asks for annual costs. In both the ACS and census, we
ultimately produce average monthly costs. There is evidence to imply that we can do
this better in the census by asking for annual costs. We do not want last month’s costs
in April 2000 (March) since it may not be representative of the entire year. Since ACS
data are gathered all year long, the best estimate of average monthly costs will result
from an average of these monthly estimates. The approaches followed in the ACS and
the decennial are appropriate for their respective data collection environments.

The words “or mobile home” were added to each part to be more inclusive of the
structure type.

H19a. LAST MONTH, what was the cost of electricity 14a.
for this house or apartment?

Unchanged

__ No

$_______.00 (Monthly)
OR
__ None

__ Yes - What is the monthly condominium fee?

H22. Is this house or apartment part of a
condominium?

__ Yes - What was the value of the food stamps?
$_____.00 (12 month amount)
__ No

__ Yes
__ No

b. the Federal home heating and
cooling assistance program?

__ Yes
__ No

a. free or reduced-price meals at school through
the National School Lunch Program or the
School Breakfast Program?

At any time DURING THE PAST 12 MONTHS,
were you or any member of this household
enrolled in or receiving benefits from:

__ No

Monthly amount - Dollars
$_____.00
OR
__None

__Yes -What is the monthly condominium fee?
For renters, answer only if you
pay the condominium fee in addition
to your rent; otherwise, mark the
“None” box.

Is this house, apartment, or mobile home
part of a condominium?

A-10

An instruction, which had been tested, was added to clarify the follow-up question for
renters.

The words “or mobile home” were added to the question to be more inclusive of the
structure type.

The words “Food Stamps” were capitalized in the question following the Yes response
category.

__ Yes - What was the value of the
Food Stamps?
Past 12 months’ value - Dollars
$_____.00 (12 month amount)
__ No

At any time DURING THE PAST
12 MONTHS, did anyone in this
household receive Food Stamps?

In part a of the item, the names of the programs were corrected.

H21. At any time DURING THE PAST 12 MONTHS, 16
did anyone in this household receive Food
Stamps?

__ Yes
__ No

b. the Federal home heating and cooling
assistance program?

__ Yes
__ No

a. free or reduced-price meals at school through
the Federal School Lunch Program or the
Federal School Breakfast Program?

H20. At any time DURING THE PAST 12 MONTHS, 15
were you or any member of this household
enrolled in or receiving benefits from:

__ Yes - What was the value of the
Food Stamps received
during the past 12 months?
Past 12 months’ value - Dollars
$_____.00 (12 month amount)
__ No

At any time DURING THE PAST
12 MONTHS, did anyone in this
household receive Food Stamps?

__ No

Monthly amount - Dollars
$_____.00
OR
__None

__Yes -What is the monthly condominium
fee? For renters, answer only if you
pay the condominium fee in addition
to your rent; otherwise, mark the
“None” box.

Is this house, apartment, or mobile home
part of a condominium?

Question number changed

Question number changed
The words “during the past 12 months” were added to the explanation
following the “Yes” response category.

Deleted for 2003

__ Owned by you or someone in this
household with a mortgage or loan?
__ Owned by you or someone in this
household free and clear (without a
mortgage or loan)?
__ Rented for cash rent?
__ Occupied without payment of cash
rent? -- Skip to question H27

H23. Is this house or apartment --

Is this house, apartment, or mobile
home --

A-11

The 1999 - 2002 ACS item matches the Census 2000 item.

The words “or mobile home” were added to each question to be more inclusive of the
structure type.

Is this house, apartment, or mobile
home --

The skip instruction was modified because some of the subsequent were
deleted for 2003.

Question number changed

__ Yes
__ No

H26. Is the rent on this house or apartment
reduced because the Federal, state, or local
government is paying part of the cost?

__ Yes
__ No

H25. Does the monthly rent include any meals?

$____.00 (Monthly)

H24. What is the monthly rent for this house or
apartment?

__ Yes
__ No

Does the monthly rent include any
meals?

Monthly amount -- Dollars
$_____.00

What is the monthly rent for this
house, apartment, or mobile home?

__ The Section “8" program
__ Some other government program
__ Not sure

What government program provides
this reduced rent?

__ Yes
__ No - Skip to question 21

Is the rent on this house, apartment, or
mobile home reduced because the Federal,
state, or local government is paying part
of the cost?

A-12

The follow-up question was added in response to welfare reform programs.

The words “or mobile home” were added to the question to be more inclusive of the
structure type.

20a.

Although there is a difference in the lead in to the question, the 1999 - 2002 ACS
question essentially matches the 1990 census item and the Census 2000 item.

The words “or mobile home” were added to the question to be more inclusive of the
structure type.

19a.

__ Yes
__ No

Does the monthly rent include any
meals?

Monthly amount -- Dollars
$_____.00

What is the monthly rent for this
house, apartment, or mobile home?

Deleted for 2003

Question number changed

18a.

$_____.00 (Value)

H28. What is the value of this property; that is,
how much would this house or mobile home
and lot, or condominium unit sell for if it
were for sale?

__ Yes
__ No

H27. Is this house or apartment in a public housing
project; that is, is it part of a government
housing project for persons with low income?

__ Yes
__ No

Is this house, apartment, or mobile home
in a public housing project; that is, is it
part of a government housing project for
persons with low income?

__ Less than $10,000
__ $10,000 to $14,999
__ $15,000 to $19,999
__ $20,000 to $24,999
__ $25,000 to $29,999
__ $30,000 to $34,999
__ $35,000 to $39,999
__ $40,000 to $49,999
__ $50,000 to $59,999
__ $60,000 to $69,999
__ $70,000 to $79,999
__ $80,000 to $89,999
__ $90,000 to $99,999
__ $100,000 tp $124,999
__ $125,000 to $149,999
__ $150,000 to $199,999
__ $200,000 to $249,999
__ $250,000 or more -- Specify $ _______________ .00

What is the value of this property; that
is, how much do you think this house
and lot, apartment, or mobile home and
lot, would sell for if it were for sale?

A-13

The ACS item has 19 response categories, and a write-in space for homes valued at
250,000 or more. The Census 2000 item has 24 response categories, with the last
category being “$1,000,000 or more.” There are no write-ins for high valued homes.

The 1999 - 2002 ACS question matches the Census 2000 question. The ACS response
area was changed from a write-in space to categories because research showed a
consistently higher non-response rate for write-in values.

The question wording was modified slightly.

Responses were changed to listed categories instead of a write-in box. Also, the 1998
question included “condominium unit” rather than “apartment.” The 1999 version is
better because it is all inclusive. (A “condominium unit” could be a house or mobile
home)

The words “or mobile home” were added to the question to be more inclusive of the
structure type.

What is the value of this property; that
is, how much do you think this house
and lot, apartment, or mobile home and
lot, would sell for if it were for sale?

Question number changed

Deleted for 2003

$_____.00 (Annual)
OR
__ None

H30. What is the annual payment for fire, hazard,
and flood insurance on this property?

$_____.00 (Annual)
OR
__ None

H29. What are the annual real estate taxes on this
property?
Annual amount - Dollars
$_____.00
OR
__ None

What are the annual real estate taxes on
THIS property?

Annual amount - Dollars
$_____.00
OR
__ None

What is the annual payment for fire,
hazard, and flood insurance on THIS
property?

A-14

The Census 2000 item also uses this format; however, the reference period is “last
year.” Since the ACS in conducted every month, a reference to something other than
“last year” is more appropriate for the ACS.

The word “THIS” was capitalized for emphasis.

Annual amount - Dollars
$_____.00
OR
__ None

What are the annual real estate taxes on
THIS property?

Annual amount - Dollars
$_____.00
OR
__ None

What is the annual payment for fire,
hazard, and flood insurance on THIS
property?

Question number changed

__ Yes, insurance included in payment
__ No, insurance paid separately or no
insurance

H34. Does the regular monthly mortgage payment
include payments for fire, hazard, or flood
insurance on this property?

__ Yes, taxes included in payment
__ No, taxes paid separately or taxes not
required

H33. Does the regular monthly mortgage payment
include payments for real estate taxes on this
property?

$_____.00 (Monthly)
OR
__ No regular payment required -- SKIP to
question H35

H32. How much is the regular monthly mortgage
payment on this property? Include payments
only on FIRST mortgage or contract to purchase

__ Yes, mortgage, deed of trust, or similar debt
__ Yes, contract to purchase
__ No -- SKIP to question H35

H31. Do you or any member of this household have
a mortgage, deed of trust, contract to
purchase, or similar debt on this property?

A-15

In parts b, c, and d of this item, the ACS uses the word “the” while the Census 2000
item uses the word “your.”

In part a of this item the ACS has the phrase “or any member of this household” which
is not part of the Census 2000 item.

In parts a and b, the SKIP to question number was changed, from 26a to 23a,
to accommodate the deleted questions in 2003.

__ Yes, insurance included in mortgage
payment
__ No, insurance paid separately or no
insurance

Does the regular monthly mortgage payment include payments for fire, hazard, or
flood insurance on THIS property?

__ Yes, taxes included in mortgage payment
__ No, taxes paid separately or taxes not
required

Does the regular monthly mortgage
payment include payments for real
estate taxes on THIS property?

Monthly amount - Dollars
$_____.00
OR
__ No regular payment required -- SKIP to
question 23a

How much is the regular monthly
mortgage payment on THIS property?
Include payments only on FIRST mortgage or
contract to purchase.

__ Yes, mortgage, deed of trust, or similar
debt
__ Yes, contract to purchase
__ No -- SKIP to question 23a

Do you or any member of this
household have a mortgage, deed of
trust, contract to purchase, or similar
debt on THIS property?

The word “mortgage” was added for clarification in the response categories in parts c
and d of the item.

22a.

Question number changed

__ Yes, insurance included in mortgage
payment
__ No, insurance paid separately or no
insurance

Does the regular monthly mortgage payment include payments for fire, hazard, or
flood insurance on THIS property?

__ Yes, taxes included in mortgage payment
__ No, taxes paid separately or taxes not
required

Does the regular monthly mortgage
payment include payments for real
estate taxes on THIS property?

Monthly amount - Dollars
$_____.00
OR
__ No regular payment required -- SKIP to
question 26a

How much is the regular monthly
mortgage payment on THIS property?
Include payments only on FIRST mortgage or
contract to purchase.

__ Yes, mortgage, deed of trust, or similar
debt
__ Yes, contract to purchase
__ No -- SKIP to question 26a

Do you or any member of this
household have a mortgage, deed of
trust, contract to purchase, or similar
debt on THIS property?

The word “THIS” was capitalized for emphasis in all four questions.

25a.

$_____.00 (Monthly)
OR
__ No regular payment required

H36. How much is the regular monthly payment on
ALL second and third mortgages and home
equity loans?

__ Yes, home equity loan
__ Yes, second mortgage
__ Yes, second mortgage and home equity loan
__ No -- SKIP to H37

H35. Do you or any member of this household have
a second mortgage or a home equity loan on
this property?

Monthly Amount - Dollars
$_____.00
OR
__ No regular payment required

How much is the regular monthly payment
on all second or junior mortgages and all
home equity loans on THIS property?

__ Yes, home equity loan
__ Yes, second mortgage
__ Yes, second mortgage and home equity
loan
__ No - SKIP to D

Do you or any member of this household
have a second mortgage or a home equity
loan on THIS property?

A-16

In the response area, the Census 2000 item does not have the third ACS response
category (“Yes, second mortgage and home equity loan.”), but does have an instruction
to “Mark (X) all boxes that apply.”

In part b of this item, the question wording was changed from “second and third
mortgages” to “second or junior mortgages,” and the phrase “on THIS property” was
added. This question matches the Census 2000 question, except that the ACS question
uses the word “the” rather than “your.”

In part a, the word “THIS” was capitalized for emphasis. The question matches the
Census 2000 question, except that the it contains the additional phrase “...or any
member of this household...” as well as the additional answer category “Yes, second
mortgage and home equity loan”.

26a.

Monthly Amount - Dollars
$_____.00
OR
__ No regular payment required

How much is the regular monthly payment
on all second or junior mortgages and all
home equity loans on THIS property?

__ Yes, home equity loan
__ Yes, second mortgage
__ Yes, second mortgage and home equity
loan
__ No - SKIP to D

Do you or any member of this household
have a second mortgage or a home equity
loan on THIS property?

Question number changed

23a.

$_____.00 (Annual)

H37. What are the total annual costs for personal
property taxes, site rent, registration fees, and
license fees on this mobile home and its site?
Do not include real estate taxes.

Annual costs -- Dollars
$_____.00

What are the total annual costs for
installment loan payments, personal
property taxes, site rent, registration
fees, and license fees on THIS mobile
home and its site? Exclude real estate
taxes.

__ Yes
__ No

Do you or any member of this
household have an installment loan or
contract on THIS mobile home?

A-17

The Census 2000 item uses the reference period “last year.” Since the ACS is
conducted every month, a reference to something other than “last year” is more
appropriate for the ACS.

In part b of this item, the word “THIS” was capitalized for emphasis. The term
“Installment loan payments” was added to this question.

Part b of this item specifies that installment loan payments should be included in the
annual costs. The revised question more clearly identifies the mobile home fees that
account for the majority of the shelter costs of households residing in these types of
units.

Part a of this item was added to clarify the mobile home households that have an
installment loan or contract.

27a.

Annual costs -- Dollars
$_____.00

What are the total annual costs for
personal property taxes, site rent,
registration fees, and license fees on
THIS mobile home and its site?
Exclude real estate taxes.

The words ”installment loan payments” were deleted.

Question number changed
Part a was deleted, in its entirety, making this a one part question.

Part a. deleted for 2003

_____ Age (in years)

Month Day Year of Birth
____
___ __________

What is this person’s
date of birth and what
is this person’s age?

__ Male
__ Female

What is
this
person’s
sex?

1996-1998
QUESTION W ORDING

__ Male
__ Female

What
is this
person’s
sex?

Age (in years)
___

Month Day Year of Birth
__
__
____

What is this person’s
date of birth and what
is this person’s age?
Print numbers in boxes.

A-18

The 1990 census item had a label and an instruction for filling the item.

The 1999 - 2002 ACS differs from the Census 2000 question only in the ordering of age and date of
birth.

The italicized instruction was added for 1999.

The 1990 census item had a label and an instruction for filling the item.

The 1999 - 2002 ACS question matches the Census 2000 item, except that the census item includes
the instruction, “Mark (X) ONE box.”

1999-2002
QUESTION W ORDING

February 2005

Month Day Year of Birth
__
__
____

Age (in years)
___

What is this person’s
age and what is this
person’s date of birth?
Print numbers in boxes.

The order in which we ask for date of birth and age was reversed to be consistent with the Census
2000 item.

Unchanged

2003-2005
QUESTION W ORDING

CHANGES TO THE AMERICAN COMMUNITY SURVEY (ACS) QUESTIONNAIRES
(PERSON QUESTIONS)

Appendix A.2 Changes to the American Community Survey (ACS) Questionnaires (Person Questions)

__ Now married
__ Widowed
__ Divorced
__ Separated
__ Never Married

What is this
person’s current
marital status?

__ Husband or wife
__ Son or daughter
__ Brother or sister
__ Father or mother
__ Grandchild
__ In-law

__ Other relative
__ Roomer, boarder
__ Housemate, roommate
__ Unmarried partner
__ Foster child
__ Other nonrelative

How is this person related to Person 1?

1996-1998
QUESTION W ORDING

__ Husband or wife
__ Son or daughter
__ Brother or sister
__ Father or mother
__ Grandchild
__ In-law
__ Other relative

__ Roomer, boarder
__ Housemate,
roommate
__ Unmarried partner
__ Foster child
__ Other nonrelative

How is this person related to Person 1?

__ Now married
__ Widowed
__ Divorced
__ Separated
__ Never Married

What is this
person’s
marital
status?

A-19

The 1990 census item had a label and an instruction for filling the item.

The word “current” was dropped to match the Census 2000 item.

The 1990 and Census 2000 response categories differ. In 1990, natural born and adopted
son/daughter were combined, in-laws were included in “other relatives,” and “foster child” was
combined with “roomer, boarder.”

The response categories in the Census 2000 item use a slash, whereas the ACS uses the word “or”
(for example: husband/wife vs. husband or wife) because of spacing constraints.

The Census 2000 item includes separate response categories for “Natural-born son/daughter,”
“Adopted son/daughter,” “Stepson/stepdaughter,”
“Son-in-law/daughter-in-law,” and “Parent-in-law”. In addition, the “Other relative” line includes a
write-in response. The sample size of the ACS did not justify the addition of these separate
categories.

1999-2002
QUESTION W ORDING

Unchanged

2003-2005
QUESTION W ORDING

__ No, not Spanish/Hispanic/ Latino
__ Yes, Mexican, Mexican-Am., Chicano
__ Yes, Puerto Rican
__ Yes, Cuban
__ Yes, other Spanish/Hispanic/ Latino – Print group --> ____________

_ Some other race}
_Multiracial
}
Print the race(s) or
group below
_Other Asian/Pacific
Islander - Print group ---

_______________________________________________

_Vietnamese
_Japanese
_Samoan
_Guamanian
_Asian Indian _Native Hawaiian
_Chinese
_Guamanian or Chamorro
_Filipino
_Samoan
_ Japanese
_Other Pacific Islander - Print race below
_Korean
_Some other race - Print race below
_Vietnamese ____________________
_Other AsianPrint race -->____________________

A-20

The 1990 census item had a label and instructions for filling the item. The categories agreed with the
1996-98 ACS, except the “Black, African Am.” category was “Black or Negro” in the 1990 census
item.

The 1999 - 2002 ACS question on race reflects the Census Bureau’s implementation of the Office
and Management and Budget’s revised standards for the classification of race and ethnic data by
Federal agencies, issued in October 1997. Specifically, the revised standards (1) allow respondents
to specify more than one race, and (2) split the category “Other Asian and Pacific Islander” into two
categories–“Other Asian” and “Other Pacific Islander.”

The 1999 - 2002 ACS item matches the Census 2000 item.

_White
_Black, African Am., or Negro
_American Indian or Alaska
Native - Print name of enrolled
or principal tribe
____________________
____________________

_Chinese
_Filipino
_Hawaiian
_Korean
_Asian Indian

_White
_Black, African Am.
_Eskimo
_Aleut
_Indian (Amer.)
Print the name of the
enrolled or principal
tribe. ----------------->

What is this person’s race? Mark (X) one or more races to indicate what this
person considers himself/herself to be.

The 1990 census item differs from the Census 2000 item in the following ways:
(a) in 1990, the item did not include the term “Latino” in the question nor in the response options;
(b) in 1990, no specific instruction was provided for non-Hispanics;
(c) in 1990, the first response option was “No (not Spanish/Hispanic);”
(d) in 1990, the second response option included a hyphen in “Mexican-Am.;” and
(e) in 1990, the “Yes, other Spanish/Hispanic” category provided examples of six Hispanic origin
groups and instructed respondents to print one group.

The ACS item, including the addition of the italicized instruction, was changed to match the Census
2000 item.

__ No, not Spanish/Hispanic/ Latino
__ Yes, Mexican, Mexican Am.,
Chicano
__ Yes, Puerto Rican
__ Yes, Cuban
__ Yes, other Spanish/Hispanic/
Latino – Print group --> ____________

Is this person Spanish/
Hispanic/Latino?
Mark (X) the “No” box if
not Spanish/Hispanic/Latino.

NOTE: Please answer BOTH Questions 5 and 6.

1999-2002
QUESTION W ORDING

Is this person Spanish/Hispanic/Latino?

6. What is this person’s race? Mark [X] one box for the race that the person
considers himself/herself to be.

1996-1998
QUESTION W ORDING

Modified the second response category.

_Asian Indian _Native Hawaiian
_Chinese
_Guamanian or Chamorro
_Filipino
_Samoan
_ Japanese
_Other Pacific Islander - Print race below
_Korean
_Some other race - Print race below
_Vietnamese ____________________
_Other AsianPrint race -->____________________

What is this person’s race? Mark (X) one or more races to indicate what this
person considers himself/herself to be.

_White
_Black or African American
_American Indian or Alaska
Native - Print name of enrolled
or principal tribe
____________________
____________________

Unchanged

2003-2005
QUESTION W ORDING

_ _ _ _ (Year)

When did this person come to live in the
United States?

__ Yes, born in the United States -Skip to question 10
__ Yes, born in Puerto Rico, Guam, the U.S.
Virgin Islands, or Northern Marianas
__ Yes, born abroad of American parent(s)
__ Yes, U.S. citizen by naturalization
__ No, not a citizen of the United States

Is this person a CITIZEN of the United States?

_________________________

In what U.S. State, territory, commonwealth
or foreign country was this person born?

1996-1998
QUESTION W ORDING

__ Outside the United States – Print name of foreign
country, or Puerto Rico, Guam, etc.
____________________

__ In the United States – Print name of state.
____________________

Where was this person born?

__ Yes, born in the United States ---> Skip to 10a
__ Yes, born in Puerto Rico, Guam, the U.S. Virgin
Islands, or Northern Marianas
__ Yes, born abroad of American parent or parents
__ Yes, U.S. citizen by naturalization
__ No, not a citizen of the United States

Is this person a CITIZEN of the United States?

Year
____

When did this person come to live in the
United States? Print numbers in boxes.

A-21

The 1990 census item was worded “When did this person come to the United States to stay?” Also,
the 1990 item had ten response intervals while the 2000 item has space to write in the year.

The ACS item added the italicized instruction to match the Census 2000 item.

The ACS item was changed to match the Census 2000 item. The wording of the third response
option was changed from “... of American parent(s)” to “... of American parent or parents.”

The ACS item was changed to match the Census 2000 item.

1999-2002
QUESTION W ORDING

Unchanged

2003-2005
QUESTION W ORDING

__ Yes, public school or public college
__ Yes, private school or private college
__ Yes, vocational, technical, or business school
__ No, has not attended in the last 3 months.

10. At any time IN THE PAST 3 MONTHS, was this
person attending a school or college? Include
nursery or preschool, kindergarten, elementary
school, and schooling that leads to a high school
diploma, college degree, or vocational certificate.

1996-1998
QUESTION W ORDING

A-22

Neither the 1990 census nor the 1996-1998 ACS included part b. of this item. Results from the 1990
census indicated that a question on grade or level attending was necessary.

Except for the reference period, the ACS item was changed to match the Census 2000 item. Monthly
interviewing on the ACS requires a moving reference period.

__ Nursery school, preschool
__ Kindergarten
__ Grade 1 to grade 4
__ Grade 5 to grade 8
__ Grade 9 to grade 12
__ College undergraduate years (freshman to
senior)
__ Graduate or professional school
(for example: medical, dental, or law school)

b. What grade or level was this person
attending? Mark (X) ONE box.

__ No, has not attended in the last 3
months –>SKIP to question 11
__ Yes, public school, public college
__ Yes, private school, private college

10a.At any time IN THE LAST 3 MONTHS, has this
person attended regular school or college?
Include only nursery or preschool, kindergarten,
elementary school, and schooling which leads to a
high school diploma or a college degree.

1999-2002
QUESTION W ORDING
Unchanged

2003-2005
QUESTION W ORDING

____________________

12. What is this person’s ancestry? For example:
Italian, African Am., Cape Verdean, Ecuadorian,
Haitian, Irish, Jamaican, Korean, Lebanese,
Mexican, Nigerian, Polish, Taiwanese, Ukrainian,
or any other ancestry.

__ None, no schooling completed
__ Nursery or preschool
__ Kindergarten
__ Grade ____ (Write grade 1-11)
__ 12th grade, NO DIPLOMA
__ HIGH SCHOOL GRADUATE– high school DIPLOMA
or the equivalent (GED)
__ Some college but no degree
__ Vocational, technical, or business school degree
__ Associate degree in college
__ Bachelor’s degree (BA, AB, BS)
__ Master’s degree (MA, MS, MEng, MEd, MSW, MBA)
__ Professional school degree (MD, DDS, DVM, LLB, JD)
__ Doctorate degree (PhD, EdD)

11. What is the highest degree or level of school
this person has COMPLETED? Mark [X] ONE box
for the highest grade completed or degree received.

1996-1998
QUESTION W ORDING

A-23

The 1990 question was worded the same as the Census 2000 question. The list of examples were
changed to shorten the list and provide a better mix of geographic and cultural diversity.

The ACS item was changed to match the Census 2000 item.

(For example: Italian, Jamaican, African Am.,
Cambodian, Cape Verdean, Norwegian,
Dominican, French Canadian, Haitian, Korean,
Lebanese, Polish, Nigerian, Mexican, Taiwanese,
Ukrainian, and so on.)

____________________
____________________

12 What is this person’s ancestry or ethnic origin?

The ACS item was changed to match the Census 2000 item.

__ No schooling completed
__ Nursery school to 4th grade
__ 5th grade or 6th grade
__ 7th grade or 8th grade
__ 9th grade
__ 10th grade
__ 11th grade
__ 12th grade – NO DIPLOMA
__ HIGH SCHOOL GRADUATE – high school
DIPLOMA or the equivalent (for example: GED)
__ Some college credit, but less than 1 year
__ 1 or more years of college, no degree
__ Associate degree (for example: AA, AS)
__ Bachelor’s degree (for example: BA, AB, BS)
__ Master’s degree (for example: MA, MS, MEng,
MEd, MSW, MBA)
__ Professional degree (for example: MD, DDS, DVM,
LLB, JD)
__ Doctorate degree (for example: PhD, EdD)

11 What is the highest degree or level of school
this person has COMPLETED? Mark (X) ONE box.
If currently enrolled, mark the previous grade or
highest degree received.

1999-2002
QUESTION W ORDING

Unchanged

2003-2005
QUESTION W ORDING

(4) ZIP Code

__ Yes

__ No, lived outside city/town limits

c.Did this person live inside the city or town limits?

____________________ ________

(3) Name of county

____________________

(2) Name of city or town

____________________
If outside U.S., print answer above and SKIP to
question 14a.

(1) Name of U.S. State, territory, commonwealth
or foreign country

b.Where did this person live 5 years ago?

__ Yes– SKIP to question 14a
__ No

13a.Did this person live in this house or apartment
5 years ago?

1996-1998
QUESTION W ORDING

ZIP Code
_____

A-24

The Census 2000 item differs from the 1990 census item with the addition of a separate write-in line
and skip instruction for a foreign country response, and the reordering of the city, county, and state
response areas.

The ACS item was changed to match the Census 2000 item, except for the time reference. The ACS
time period of one year reflects the ongoing data collection on the ACS, and allows for annual
estimates of migration.

Name of state
_______________

Name of county
_______________

__ Yes
__ No, outside the city/town limits

c. Did this person live inside the limits of the
city or town?

Name of city, town, or post office
____________________

b. Where did this person live 1 year ago?

__ Person is under 1 year old --> SKIP to the
questions for Person 2 on page 10.
__ Yes, this house --> SKIP to F in the next
column
__ No, outside the United States – Print name of
foreign country, or Puerto Rico, Guam, etc.,
below; then SKIP to F in next column.
__ No, different house in the United States

13a. Did this person live in this house or
apartment 1 year ago?

1999-2002
QUESTION W ORDING

ZIP Code
_____

The numerical order of questions 13 and 14 were reversed so that part c of this question would not be
displayed in a separate column of the questionnaire

Name of state
_______________

Name of county
_______________

__ Yes
__ No, outside the city/town limits

c. Did this person live inside the limits of the
city or town?

Name of city, town, or post office
____________________

b. Where did this person live 1 year ago?

__ Person is under 1 year old --> SKIP to the
questions for Person 2 on page 10.
__ Yes, this house --> SKIP to F
__ No, outside the United States – Print name of
foreign country, or Puerto Rico, Guam, etc.,
below; then SKIP to F
__ No, different house in the United States

14a. Did this person live in this house or
apartment 1 year ago?

2003-2005
QUESTION W ORDING

__ Difficulty seeing (even with glasses)
__ Difficulty hearing (even with a hearing aid)
__ Difficulty walking, or
__ None of the above

15. If this person has difficulty seeing, hearing,
or walking, mark [X] the appropriate boxes.
If this person has no difficulty with these
activities, mark “None of the above.”

__ Very well __ Not well
__ Well
__ Not at all

c.How well does this person speak English?

____________________

b.What is this language? For example: Korean,
Italian, Spanish, Vietnamese

__ Yes
__ No, only English ---> SKIP to question 15

14a.Does this person speak a language other than
English at home?

1996-1998
QUESTION W ORDING

__ Yes __ No

A-25

The ACS item was changed to match the Census 2000 item.

b. A condition that substantially limits
one or more basic physical activities
such as walking, climbing stairs,
reaching, lifting, or carrying?
__ Yes __ No

a. Blindness, deafness, or a severe
vision or hearing impairment:

15 Does this person have any of the following
long-lasting conditions:

Compared to the 1990 census item, the Census 2000 item has different examples of languages.

The ACS item was changed to match the Census 2000 item. The only change was to drop the term
“only English” after the “No” response in item 14a.

__ Very well __ Not well
__ Well
__ Not at all

c.How well does this person speak English?

____________________
For example: Korean, Italian, Spanish, Vietnamese

b.What is this language?

__ Yes
__ No --> Skip to question 15

14a.Does this person speak a language other
than English at home?

1999-2002
QUESTION W ORDING

Unchanged

The numerical order of questions 13 and 14 were reversed so that part c of the previous question
would not be displayed in a separate column of the questionnaire

__ Very well __ Not well
__ W ell
__ Not at all

c.How well does this person speak English?

____________________
For example: Korean, Italian, Spanish, Vietnamese

b.W hat is this language?

__ Yes
__ No --> Skip to question 14

13a.Does this person speak a language other
than English at home?

2003-2005
QUESTION W ORDING

__ No

____ Babies

[ ] None

17. How many babies has this person ever had,
not counting stillbirths?

__ Yes

b. Prevents this person from working at a job
or business?

__ Yes

a. Makes it difficult to go outside the home
alone, for example, to shop or visit a
doctor’s office?

16. Does this person have a long-lasting physical
or mental condition that –

1996-1998
QUESTION W ORDING

A-26

This item is not part of the Census 2000 items because there were no required needs for these data.

The ACS item was changed to provide a more meaningful description of current and local area
fertility patterns than the previous question.

__ Yes
__ No

17 Has this person given birth to any children in
the past 12 months?

The ACS item was changed to match the Census 2000 item.

d. (Answer only if this person is 16 YEARS
OLD OR OVER) Working at a job or
business?
__ Yes __ No

c. (Answer only if this person is 16 YEARS
OLD OR OVER.) Going outside the
home alone to shop or visit a
doctor’s office?
__ Yes __ No

__ Yes __ No

Question number changed

__ Yes
__ No

18 Has this person given birth to any children in
the past 12 months?

This question was established as a result of the removal/separation of parts c and d in question 16,
preceding this one, to limit this question to persons 15 years old or older.

b. Working at a job or
business?

a. Going outside the home alone to shop or visit a
doctor’s office?
__ Yes __ No

17 Because of a physical, mental, or emotional
condition lasting 6 months or more, does this
person have any difficulty in doing any of the
following activities:

Parts c and d were removed/separated from this question in order to establish a new question
immediately following this one, for persons 15 years old or older.

__ Yes __ No

b. Dressing, bathing, or getting around
inside the home?
__ Yes __ No

a. Learning, remembering, or
concentrating?
__ Yes __ No

b. Dressing, bathing, or getting around
inside the home?
__ Yes __ No

16 Because of a physical, mental, or emotional
condition lasting 6 months or more, does this
person have any difficulty in doing any of the
following activities:

2003-2005
QUESTION W ORDING

16 Because of a physical, mental, or emotional
condition lasting 6 months or more, does this
person have any difficulty in doing any of the
following activities:

1999-2002
QUESTION W ORDING

n/a

1996-1998
QUESTION W ORDING

Less than 1 month
1 to 6 months
7 to 12 months
More than 12 months

A-27

The Census 2000 item will use these response categories.

These response categories were adopted in 2000.

Less than 6 months
6 to 11 months
1 or 2 years
3 or 4 years
5 years or more.

The ACS added this set of questions to match the set for the Census 2000, except for the response
categories in part c. Results from the Census 2000 Dress Rehearsal, which were not available in time
for the 1999 ACS, indicated that the response categories for part c. should be as follows:

__
__
__
__

c. How long has this grandparent been
responsible for the(se) grandchild(ren)? If
the grandparent is financially responsible for
more than one grandchild, answer the question
for the grandchild for whom the grandparent
has been responsible for the longest period of
time.

__ Yes
__ No –> SKIP to question 19

b. Is this grandparent currently responsible for
most of the basic needs of any
grandchild(ren) under the age of 18 who
live(s) in this house or apartment?

__ Yes
__ No –>SKIP to question 19

18a.Does this person have any of his/her own
grandchildren under the age of 18 living in
this house or apartment?

1999-2002
QUESTION W ORDING

Less than 6 months
6 to 11months
1 or 2 years
3 or 4 years
5 or more years
Question number changed

__
__
__
__
__

c.How long has this grandparent been
responsible for the(se) grandchild(ren)? If
the grandparent is financially responsible for
more than one grandchild, answer the question
for the grandchild for whom the grandparent
has been responsible for the longest period of
time.

__ Yes
__ No –> SKIP to question 20

b.Is this grandparent currently responsible for
most of the basic needs of any
grandchild(ren) under the age of 18 who
live(s) in this house or apartment?

__ Yes
__ No –>SKIP to question 20

19a Does this person have any of his/her own
grandchildren under the age of 18 living in
this house or apartment?

2003-2005
QUESTION W ORDING

__ Yes, now on active duty
__ Yes, on active duty in past, but not now
__ No active duty service - SKIP to question 21

18. Has this person ever served on ACTIVE DUTY in
the U.S. Armed Forces, military Reserves, or
National Guard? Include activation during
Operation Desert Shield/Storm and service in the
Merchant Marine during World War II. Do NOT
include training for the Reserves or National Guard.

1996-1998
QUESTION W ORDING

A-28

The item is essentially a simplified and updated version of the 1990 census item. The “No active
duty service” answer category was expanded to distinguish persons whose only military service was
for training in the Reserves or National Guard, from persons with no military experience whatsoever.

The ACS item was changed to match the Census 2000 item.

__ Yes, now on active duty
__ Yes, on active duty in past, but not now
__ No, training for Reserves or National Guard
only - SKIP to question 22
__ No, never served in the military - SKIP to
question 22

19 Has this person ever served on active duty in
the U.S. Armed Forces, military Reserves, or
National Guard? Active duty does not include
training for the Reserves or National Guard, but
DOES include activation, for example, for the
Persian Gulf War.

1999-2002
QUESTION W ORDING

Question number changed

__ Yes, now on active duty
__ Yes, on active duty during
the last 12 months, but not now
__ Yes, on active duty in past, but not
during the last 12 months
__ No, training for Reserves or National Guard
only - SKIP to question 23
__ No, never served in the military - SKIP to
question 23

20. Has this person ever served on active duty in
the U.S. Armed Forces, military Reserves, or
National Guard? Active duty does not include
training for the Reserves or National Guard, but
DOES include activation, for example, for the
Persian Gulf War.

2003-2005
QUESTION W ORDING

__ August 1990 or later (including Persian Gulf War)
__ September 1980 to July 1990
__ May 1975 to August 1980
__ Vietnam era (August 1964 to April 1975)
__ February 1955 to July 1964
__ Korean conflict (June 1950 to January 1955)
__ World War II (September 1940 to July 1947)
__ Some other time

19. When did this person serve on active duty in
the U.S. Armed Forces? Mark (X) a box for EACH
period in which this person served.

1996-1998
QUESTION W ORDING

A-29

The ACS modified the response categories by changing “Korean conflict” to “Korean War.”

__ April 1995 or later
__ August 1990 to March 1995 (including Persian Gulf War)
__ September 1980 to July 1990
__ May 1975 to August 1980
__ Vietnam era (August 1964 to April 1975)
__ February 1955 to July 1964
__ Korean War (June 1950 to January 1955)
__ World War II (September 1940 to July 1947)
__ Some other time

2001-2002 only
20 When did this person serve on active duty in the U.S. Armed Forces? Mark (X) a
box for EACH period in which this person served.

The Census 2000 response categories are different from those used in 1990, as follows: the 1990
category “September 1980 or later” was expanded; the 1990 category “World War I” was dropped;
and the 1990 category “Any other time” was replaced with “Some other time”.

The Census 2000 question is different from the 1990 census question (“Was active-duty military
service during --”).

The ACS item was changed to match the Census 2000 item.

__ April 1995 or later
__ August 1990 to March 1995 (including
Persian Gulf War)
__ September 1980 to July 1990
__ May 1975 to August 1980
__ Vietnam era (August 1964 to April 1975)
__ February 1955 to July 1964
__ Korean conflict (June 1950 to January 1955)
__ World War II (September 1940 to July 1947)
__ Some other time

1999-2000 only
20 When did this person serve on active duty in
the U.S. Armed Forces? Mark (X) a box for EACH
period in which this person served.

1999-2002
QUESTION W ORDING

Question number changed. Response categories modified

__ September 2001 or later
__ August 1990 to August 2001 (including
Persian Gulf War)
__ September 1980 to July 1990
__ May 1975 to August 1980
__ Vietnam era (August 1964 to April 1975)
__ March 1961 to July 1964
__ February 1955 to February 1961
__ Korean War (June 1950 to January 1955)
__January 1947 to June 1950
__ World War II (December 1941 to December 1946)
__ November 1941 or earlier

21 When did this person serve on active duty in the U.S. Armed Forces? Mark (X) a
box for EACH period in which this person served.

2003-2005
QUESTION W ORDING

__ No - SKIP to question 28a

This item was deleted starting with the 1999 ACS because it is not a Census 2000 item.

__ Actual Hours worked LAST WEEK

22. LAST WEEK, how many hours did this person
actually work at all jobs? Subtract any time
off; add overtime or extra hours worked.

__ Yes

21. LAST WEEK, did this person do ANY work for
pay or profit?

_ _ _ _ Years and _ _ Months

20. In total, how much time has this person spent
on active duty in the U.S. Armed Forces?

1996-1998
QUESTION W ORDING

Deleted 1999

A-30

This question is the key question in the set of questions that collect employment status. The Census
2000 questions replace the somewhat different set from the 1990 census.

The ACS item was changed to match the Census 2000 item.

__ Yes
__ No - SKIP to question 28

22 LAST WEEK, did this person do ANY work for
either pay or profit? Mark (X) the “Yes” box even if
the person worked only 1 hour, or helped without
pay in a family business or farm for 15 hours or more,
or was on active duty in the Armed Forces.

The 1990 census question was the same, but the response area required the respondent to write down
the number of years.

The ACS item was changed to match the Census 2000 item.

__ Less than 2 years
__ 2 years or more

21 In total, how many years of active-duty
military service has this person had?

1999-2002
QUESTION W ORDING

Deleted 1999

Question number changed

__ Yes
__ No - SKIP to question 29

23 LAST WEEK, did this person do ANY work for
either pay or profit? Mark (X) the “Yes” box even if
the person worked only 1 hour, or helped without
pay in a family business or farm for 15 hours or more,
or was on active duty in the Armed Forces.

Question number changed

__ Less than 2 years
__ 2 years or more

22 In total, how many years of active-duty
military service has this person had?

2003-2005
QUESTION W ORDING

__ No

e. Name of state
_________________

________________

d. Name of county

__ Yes

f. ZIP Code
________

c. Is the work location inside the limits of the city
or town?

_________________________________

b. Name of city, town, or post office

_________________________________________
If the exact address is unknown, give a description of
the location such as the building name or the
nearest street or intersection. For example: Town
Center Mall, 1st National Bank Building, Reno
Airport, 2nd Ave. and 4th St.
_______________________________

_______________________________

A-31

This item is basically the same as the 1990 census item. Labels on the write-in spaces were modified
to provide clarifications.

Question number changed

________

The ACS item was changed to match the Census 2000 item.

f. ZIP Code

___________________________________

e. Name of U.S. state or foreign country

f. ZIP Code

___________________________________

e. Name of U.S. state or foreign country

________________

d. Name of county

________________

__ Yes
__ No, outside the city/town limits

c. Is the work location inside the limits of that city or town?

b. Name of city, town, or post office

c. Is the work location inside the limits of that city or town?

____________________________________
If the exact address is not know, give a
description of the location such as the building
name or the nearest street or intersection.

a. Address (Number and street name)

24 At what location did this person work LAST WEEK?
If this person worked at more than one location, print
where he or she worked most last week.

2003-2005
QUESTION W ORDING

____________________________________
If the exact address is not know, give a
description of the location such as the building
name or the nearest street or intersection.

a. Address (Number and street name)

23 At what location did this person work LAST WEEK?
If this person worked at more than one location, print
where he or she worked most last week.

23. LAST WEEK, at what address or location did
this person work? If this person worked at more
than one address or location, print where he or she
worked most last week.

a. Address (Number and street)

1999-2002
QUESTION W ORDING

1996-1998
QUESTION W ORDING

__ Bicycle
__ Walked
__ Worked at home
SKIP to question 32
__ Other method

__ p.m.

____ Minutes – SKIP to question 32

27. LAST WEEK, how many minutes did it usually
take this person to get from home to work?

_____ : _____ __ a.m.
Hour Minute

26. LAST WEEK, what time did this person usually
leave home to go to work?

___ Person(s)

25. LAST WEEK, how many people, including this
person, usually rode to work in the car, truck,
or van?

__ Car, truck, or van
__ Public transportation
(bus, trolley, subway,
or railroad)
__ Taxicab
__ Motorcycle

24. LAST WEEK, how did this person usually get to
work? If more than one method was used during
the trip, mark (X) the box for the one used for most
of the distance.

1996-1998
QUESTION W ORDING

__ Motorcycle
__ Bicycle
__ Walked
__ Worked at home –
SKIP to question 32
__ Other method

__ a.m.
__ p.m.

A-32

The ACS item was changed to match the Census 2000 item which is the same as the 1990 census
item.

____ Minutes

27 How many minutes did it usually take this
person to get from home to work LAST WEEK?

The ACS item was changed to match the Census 2000 item which is the same as the 1990 census
item, except that the ACS has “Hour” and “Minute” labels in the answer space.

Hour Minute
_____ : _____

26 What time did this person usually leave home to
go to work LAST WEEK?

The Census 2000 response categories differ from 1990 as follows: The two categories
“5 people” and “6 people” were collapsed to “5 or 6 people”; and the two categories
“7 to 9 people” and “10 or more people” were collapsed to “7 or more people.”

In order to save space, the ACS response area allows for a write-in rather than labeled response
categories.

The ACS question was changed to match the Census 2000 question which is the same as the 1990
census question.

Person(s)
_____

25 How many people, including this person,
usually rode to work in the car, truck, or van
LAST WEEK?

The ACS item was changed to match the Census 2000 item, which is the same as the 1990 census
item except for the format of the skip instructions.

__ Car, truck, or van
__ Bus or trolley bus
__ Streetcar or trolley car
__ Subway or elevated
__ Railroad
__ Ferryboat
__ Taxicab

24 How did this person usually get to work LAST
WEEK? If this person usually used more than one
method of transportation during the trip, mark (X) the
box of the one used for most of the distance.

1999-2002
QUESTION W ORDING

__ Motorcycle
__ Bicycle
__ Walked
__ Worked at home
– SKIP to question 33
__ Other method

__ a.m.

Question number changed

____ Minutes

28 How many minutes did it usually take this
person to get from home to work LAST WEEK?

Question number changed

Hour Minute
_____ : _____ __ p.m.

27 What time did this person usually leave home to
go to work LAST WEEK?

Question number changed

Person(s)
_____

26 How many people, including this person,
usually rode to work in the car, truck, or van
LAST WEEK?

Question number changed

__ Car, truck, or van
__ Bus or trolley bus
__ Streetcar or trolley car
__ Subway or elevated
__ Railroad
__ Ferryboat
__ Taxicab

25 How did this person usually get to work LAST
WEEK? If this person usually used more than one
method of transportation during the trip, mark (X) the
box of the one used for most of the distance.

2003-2005
QUESTION W ORDING

__ Yes, if a job had been offered
__ Yes, if recalled from layoff
__ No, because of own temporary illness
__ No, because of all other reasons (in school, etc.)

30. LAST WEEK, could this person have gone to
work?

__ Yes
__ No - SKIP to question 31

29. Has this person been looking for work
during the last 4 weeks?

__ Yes, on vacation, temporary illness, labor
dispute, etc. - SKIP to question 31
__ No

b.LAST WEEK, was this person TEMPORARILY
absent from a job or business?

__ Yes, on temporary layoff from most recent job SKIP to question 30
__ Yes , permanently laid off from most recent job SKIP to question 29
__ No

28a.LAST WEEK, was this person on layoff from
a job?

1996-1998
QUESTION W ORDING

A-33

The 1990 census item differed slightly from the Census 2000 item.

The ACS item was changed to match the Census 2000 item.

__ Yes, could have gone to work
__ No, because of own temporary illness
__ No, because of all other reasons (in school, etc.)

30 LAST WEEK, could this person have started a job
if offered one, or returned to work if recalled?

The ACS item remains the same and matches the Census 2000 item and the 1990 census item.

__ Yes
__ No -> SKIP to question 31

29 Has this person been looking for work during
the last 4 weeks?

These three questions are part of a set of questions that collect employment status. The Census 2000
set of questions replaces a somewhat different set from the 1990 census.

The ACS items were changed to match the Census 2000 items.

__ Yes -> SKIP to question 30
__ No

c. Has this person been informed that he or she
will be recalled to work within the next
6 months OR been given a date to return to
work?

__ Yes, on vacation, temporary illness,
labor dispute, etc. -> SKIP to question 31
__ No -> SKIP to question 29

b LAST WEEK, was this person TEMPORARILY
absent from a job or business?

__ Yes -> SKIP to question 28c
__ No

28a LAST WEEK, was this person on layoff from
a job?

1999-2002
QUESTION W ORDING

Question number changed

__ Yes, could have gone to work
__ No, because of own temporary illness
__ No, because of all other reasons (in school, etc.)

31 LAST WEEK, could this person have started a job
if offered one, or returned to work if recalled?

Question number changed

__ Yes
__ No -> SKIP to question 32

30 Has this person been looking for work during
the last 4 weeks?

Question number changed

__ Yes -> SKIP to question 31
__ No

c. Has this person been informed that he or she
will be recalled to work within the next
6 months OR been given a date to return to
work?

__ Yes, on vacation, temporary illness,
labor dispute, etc. -> SKIP to question 32
__ No -> SKIP to question 30

b.LAST W EEK, was this person TEM PORARILY
absent from a job or business?

__ Yes -> SKIP to question 29c
__ No

29a.LAST WEEK, was this person on layoff from
a job?

2003-2005
QUESTION W ORDING

34 - 39. CURRENT OR MOST RECENT JOB ACTIVITY.
Describe clearly this person’s chief job activity or
business. If this person had more than one job,
describe the one at which this person worked the
most hours. If this person had no job or business last
week, give information for his/her last job.

____ Usual hours worked per WEEK

33. During the PAST 12 MONTHS, in the WEEKS
WORKED, how many hours did this person
usually work each WEEK?

_____Weeks (Including paid vacation, paid sick
leave, and military service)

32. During the PAST 12 MONTHS, in how many
WEEKS did this person work, even for a few
hours? Include paid vacation, paid sick leave,
and military service in the total.

__ Within the past 12 months
__ Between 1 to 5 years ago - SKIP to question 34
__ Over 5 years ago or never worked - SKIP to
question 40

31. When did this person last work, even for a
few days?

1996-1998
QUESTION W ORDING

A-34

34 - 39 CURRENT OR MOST RECENT JOB ACTIVITY.
Describe clearly this person’s chief job activity or
business last week. If this person had more than one
job, describe the one at which this person worked the
most hours. If this person had no job or business last
week, give information for his/her last job.

The ACS item was changed to be consistent with the Census 2000 item, which is basically the same
as the 1990 census item. The primary difference is the reference period, which is the past 12 months
for the ACS, and the previous calendar year for the Census 2000 item.

Usual hours worked each WEEK
____________

33 During the PAST 12 MONTHS, in the WEEKS
WORKED, how many hours did this person
usually work each WEEK?

____Weeks

32 During the PAST 12 MONTHS, how many
WEEKS did this person work? Count paid
vacation, paid sick leave, and military service.

The ACS item differs from the Census 2000 item and the 1990 census item in that this ACS question,
and the accompanying two questions on weeks worked and hours worked each week (questions 32
and 33) are used to collect data on work experience over the past 12 months , whereas the census
questions are used to collect data for the previous calendar year.

The ACS item was changed by deleting the word “between” in the second response category, to
clarify the time periods.

__ Within the past 12 months
__ 1 to 5 years ago - SKIP to question 34
__ Over 5 years ago or never worked - SKIP to
Question 40

31 When did this person last work, even for a
few days?

1999-2002
QUESTION W ORDING

Question number changed

35-40 CURRENT OR MOST RECENT JOB ACTIVITY.
Describe clearly this person’s chief job activity or
business last week. If this person had more than one
job, describe the one at which this person worked the
most hours. If this person had no job or business last
week, give information for his/her last job.

Question number changed

Usual hours worked each WEEK
____________

34 During the PAST 12 MONTHS, in the WEEKS
WORKED, how many hours did this person
usually work each WEEK?

Question number changed

____Weeks

33 During the PAST 12 MONTHS, how many
WEEKS did this person work? Count paid
vacation, paid sick leave, and military service.

Question number changed

__ Within the past 12 months
__ 1 to 5 years ago - SKIP to question 35
__ Over 5 years ago or never worked - SKIP to
Question 41

32 When did this person last work, even for a
few days?

2003-2005
QUESTION W ORDING

36. What kind of business or industry was this?
Describe the activity at the location where employed.
For example: hospital, newspaper publishing,
public high school.
_____________________________________

_____________________________________
Name of company, business, branch of the Armed
Forces or other employer

35. For whom did this person work?

__ An employee of a PRIVATE FOR PROFIT company
or business, or of an individual, for wages, salary,
or commissions?
__ An employee of a PRIVATE NOT-FOR-PROFIT,
tax-exempt, or charitable organization?
__ A local GOVERNMENT employee (city, county, etc)?
__ A state GOVERNMENT employee?
__ An active duty U.S. Armed Forces member?
__ A federal GOVERNMENT employee (excluding
active duty military)?
__ SELF-EMPLOYED in own NOT INCORPORATED
business, professional practice, or farm?
__ SELF-EMPLOYED in own INCORPORATED business,
professional practice, or farm?
__ Working WITHOUT PAY in family business or farm?

34. Was this person –

1996-1998
QUESTION W ORDING

_ an employee of a PRIVATE FOR PROFIT company
or business, or of an individual, for wages, salary,
or commissions?
_ an employee of a PRIVATE NOT FOR PROFIT,
tax-exempt, or charitable organization?
_ a local GOVERNMENT employee (city,
county, etc.)?
_ a state GOVERNMENT employee?
_ a Federal GOVERNMENT employee
_ SELF-EMPLOYED in own NOT INCORPORATED
business, professional practice, or farm?
_ SELF-EMPLOYED in own INCORPORATED business,
professional practice, or farm?
__ Working WITHOUT PAY in family business or farm?

Question number changed

The ACS was changed to match the Census 2000 item, except for a minor difference in one of the
examples.

A-35

The Census 2000 item is basically the same as the 1990 census item except for changes to the list of
examples.

37 What kind of business or industry was this?
Describe the activity at the location where employed.
(For example: hospital, newspaper publishing, mail
order house, auto engine manufacturing, bank)
_____________________________________

36 What kind of business or industry was this?
Describe the activity at the location where employed.
(For example: hospital, newspaper publishing, mail
order house, auto engine manufacturing, bank)
_____________________________________

Question number changed

Name of company, business, or other employer
_____________________________________

The ACS item was changed to match the Census 2000 item, which is basically the same as the 1990
census item.

If now on active duty in the
Armed Forces, mark (X) this box -> [ ]
and print the branch of the Armed Forces.

36 For whom did this person work?

If now on active duty in the
Armed Forces, mark (X) this box -> [ ]
and print the branch of the Armed Forces.

35 For whom did this person work?

The ACS continued to use “a” or “an” at the beginning of the response categories (where
appropriate) to emphasize that each response category is an extension of the first three words of the
question.

Question number changed

Mark (X) ONE box.

The ACS item was changed to be consistent with the Census 2000 item, which is basically the same
as the 1990 census item.

35 Was this person –

2003-2005
QUESTION W ORDING

34 Was this person –

1999-2002
QUESTION W ORDING

__ Retail Trade
__ Other (agriculture,
construction, service,
government, etc)?

Indicate the types of income this person received
during the PAST 12 MONTHS and enter the amounts
received. If you do not know the exact amount,
please give an estimate. For income received jointly,
report if possible, the appropriate share for each
person. Otherwise report the whole amount for
only one person and mark the “No” box for the
other person. If net income was a loss, mark the
“LOSS” box to the right of the dollar amount.

40. INCOME IN THE PAST 12 MONTHS.

39. What were this person’s most important
activities or duties? For example: patient care,
directing hiring policies, teaching 9th grade biology
_____________________________________

38. What kind of work was this person doing?
For example: registered nurse, personnel manager,
high school teacher
_____________________________________

__ Manufacturing?
__ Wholesale trade?

37. Is this mainly-

1996-1998
QUESTION W ORDING

If net income was a loss, mark the “Loss” box to the
right of the dollar amount.
For income received jointly, report the appropriate
share for each person – or, if that’s not possible,
report the whole amount for only one person and
mark the “No” box for the other person.

If net income was a loss, mark the “Loss” box to the
right of the dollar amount.

For income received jointly, report the appropriate
share for each person – or, if that’s not possible,
report the whole amount for only one person and
mark the “No” box for the other person.

A-36

Mark (X) the “No” box to show types of income
NOT received.

Question number changed

Mark (X) the “Yes” box for each type of income this
person received, and give your best estimate of the
TOTAL AMOUNT during the PAST 12 MONTHS.
(Note: The “past 12 months” is the period from
today’s date one year ago up through today.)

41 INCOME IN THE PAST 12 MONTHS.

Question number changed

The ACS item was changed to match the Census 2000 item, which is basically the same as the 1990
census item except for the examples.

40 INCOME IN THE PAST 12 MONTHS.

40 What were this person’s most important
activities or duties? (For example: patient care,
directing hiring policies, supervising order clerks,
typing and filing, reconciling financial records)
_____________________________________

Question number changed

The ACS item was changed to match the Census 2000 item, which is basically the same as the 1990
census item except for the examples.

39 What were this person’s most important
activities or duties? (For example: patient care,
directing hiring policies, supervising order clerks,
typing and filing, reconciling financial records)
_____________________________________

39 What kind of work was this person doing? (For
example: registered nurse, personnel manager,
supervisor of order department, secretary, accountant)
_____________________________________

Question number changed

__ manufacturing?
__ wholesale trade?
__ retail trade?
__ other (agriculture, construction, service,
government, etc.)?

38 Is this mainly – Mark (X) one box.

2003-2005
QUESTION W ORDING

38 What kind of work was this person doing? (For
example: registered nurse, personnel manager,
supervisor of order department, secretary, accountant)
_____________________________________

The first letter of the first word of each ACS response category was changed to lower case to
emphasize that each response category is an extension of the first three words of the question.

The ACS item was changed to be consistent with the Census 2000 item, which is basically the same
as the 1990 census item.

__ manufacturing?
__ wholesale trade?
__ retail trade?
__ other (agriculture, construction, service,
government, etc.)?

37 Is this mainly – Mark (X) one box.

1999-2002
QUESTION W ORDING

$
.00
(Past 12 months - Dollars)

$
.00
__ Loss
(Past 12 months - Dollars)

__ Yes --->
__ No

$
.00
(Past 12 months - Dollars)

d. Social Security or Railroad Retirement?

__ Yes --->
__ No

c. interest, dividends, net rental income, royalty
income, or income from estates and trusts?
Report even small amounts credited to an account.

__ Yes --->
__ No

b. self-employment income from own business
(farm or non-farm) including proprietorship
and partnership? Report NET income after
business expenses.

__ Yes --->
__ No

40a. wages or salary? Report commissions, bonuses, or
tips from all jobs BEFORE DEDUCTIONS FOR TAXES,
BONDS, DUES, OR OTHER ITEMS.

In the PAST 12 MONTHS, did this person
receive-

1996-1998
QUESTION W ORDING

$
.00
TOTAL AMOUNT for past
12 MONTHS

__ No

__ Yes --->

$
.00
TOTAL AMOUNT for past
12 MONTHS

A-37

$
.00
__ Loss
TOTAL AMOUNT for past
12 MONTHS

d. Social Security or Railroad Retirement.

__ No

__ Yes --->

$
.00
__ Loss
TOTAL AMOUNT for past
__ No
12 MONTHS
c. Interest, dividends net rental income, royalty
income, or income from estates and trusts.
Report even small amounts credited to an account.

__ Yes --->

b. Self-employment income from own nonfarm
businesses or farm businesses, including
proprietorships and partnerships. Report NET
income after business expenses.

__ No

__ Yes --->

40a. Wages, salary, commissions, bonuses, or tips
from all jobs. Report amount before deductions
for taxes, bonds, dues, or other items.

1999-2002
QUESTION W ORDING

$
.00
TOTAL AMOUNT for past
12 MONTHS

$
.00
__ Loss
TOTAL AMOUNT for past
12 MONTHS

$
.00
TOTAL AMOUNT for past
12 MONTHS
Question number changed

__ No

__ Yes --->

d. Social Security or Railroad Retirement.

__ No

__ Yes --->

$
.00
__ Loss
TOTAL AMOUNT for past
__ No
12 MONTHS
c. Interest, dividends net rental income, royalty
income, or income from estates and trusts.
Report even small amounts credited to an account.

__ Yes --->

b. Self-employment income from own nonfarm
businesses or farm businesses, including
proprietorships and partnerships. Report NET
income after business expenses.

__ No

__ Yes --->

41a. Wages, salary, commissions, bonuses, or tips
from all jobs. Report amount before deductions
for taxes, bonds, dues, or other items.

2003-2005
QUESTION W ORDING

$
.00
(Past 12 months - Dollars)

$
.00
__ None
(Past 12 months - Dollars)__ Loss

41. What was this person’s total income during
the PAST 12 MONTHS? Add entries 40a-g;
subtract any losses.

NOTE: The 1998 ACS questionnaire deleted references to AFDC because of welfare
law reforms.

__ Yes --->
__ No

g. Other sources of income received regularly
such as Veterans’ (VA) payments,
unemployment compensation, child support
or alimony, etc? Do NOT include lump sum
payments such as money from an inheritance or
the sale of a home.

__ Yes --->
__ No

f. Supplemental Security Income (SSI), Aid to
Families with Dependent Children (AFDC), or
other public assistance or public welfare
payments?

__ Yes --->
__ No

e. retirement, survivor, or disability pensions? Do
NOT include Social Security.

40. (Continued)

1996-1998
QUESTION W ORDING

$
.00
TOTAL AMOUNT for past
12 MONTHS

.00
__ Loss
TOTAL AMOUNT for
past 12 MONTHS

A-38

The ACS item was changed to be consistent with the Census 2000 item which is basically the same
as the 1990 census item. The ACS has a different reference period, asking about the past 12 months
while the census items ask about the previous calendar year.

__ None OR $

41 What was this person’s total income during the
PAST 12 MONTHS? Add entries in questions 40a to
40h; subtract any losses. If net income was a loss, enter
the amount and mark (X) the “Loss” box next to the
dollar amount

The questions for the Census 2000 differ from the 1990 census questions concerning selfemployment and public assistance items. In 1990, the self-employment income item was asked in
two parts (non-farm and farm self-employment). Questions about SSI and public assistance were
asked as one in the 1990 census. The public assistance item dropped references to AFDC because of
welfare law reforms.

The ACS questions were changed to be consistent with the questions for the Census 2000. The
instructions are slightly different to reflect differences in the reference periods. The ACS asks about
the past 12 months, and the questions for the Census 2000 ask about the previous calendar year.

__ No

__ Yes --->

$
.00
TOTAL AMOUNT for past
__ No
12 MONTHS
h. Any other sources of income received regularly
such as Veterans’ (VA) payments, unemployment compensation, child support or alimony.
Do NOT include lump sum payments such as money
from an inheritance or the sale of a home.

__ Yes --->

$
.00
TOTAL AMOUNT for past
__ No
12 MONTHS
g. Retirement, survivor, or disability pensions.
Do NOT include Social Security.

__ Yes --->

$
.00
TOTAL AMOUNT for past
__ No
12 MONTHS
f. Any public assistance or welfare payments
from the state or local welfare office.

__ Yes --->

e. Supplemental Security Income (SSI).

40 (Continued)

1999-2002
QUESTION W ORDING

$
.00
TOTAL AMOUNT for past
12 MONTHS

Question number changed

__ None OR $

.00
__ Loss
TOTAL AMOUNT for
past 12 MONTHS

42 What was this person’s total income during the
PAST 12 MONTHS? Add entries in questions 40a to
40h; subtract any losses. If net income was a loss, enter
the amount and mark (X) the “Loss” box next to the
dollar amount

Question number changed

__ No

__ Yes --->

$
.00
TOTAL AMOUNT for past
__ No
12 MONTHS
g. Retirement, survivor, or disability pensions.
Do NOT include Social Security.

__ Yes --->

$
.00
TOTAL AMOUNT for past
__ No
12 MONTHS
f. Any public assistance or welfare payments
from the state or local welfare office.

__ Yes --->

e. Supplemental Security Income (SSI).

41 (Continued)

2003-2005
QUESTION W ORDING

Appendix B.1 Pre-Notice Letter (U.S. Version)

UNITED STATES DEPARTMENT OF COMMERCE
Economics and Statistics Administration

U.S. Census Bureau
Washington, DC 20233-0001
OFFICE OF THE DIRECTOR

Dear Resident:
In a few days your household will receive a questionnaire in the mail for a very
important national survey, the American Community Survey. When the questionnaire
arrives, please fill it out and mail it back promptly. The U.S. Census Bureau is
conducting this survey and chose your address, not you personally, as part of a
randomly selected sample.
The American Community Survey collects information about various topics like
education, housing, and jobs. Information from this survey is used by federal, state,
local, and tribal governments to meet the needs of communities across America. For
example, community leaders use this information to decide where schools,
highways, hospitals, and other services are needed. The survey also is used to
develop programs to reduce traffic congestion, provide job training, and plan for the
healthcare needs of the elderly.
If you have access to the Internet and want to learn more about the American

Community Survey, please visit the Census Bureau’s Web site:

www.census.gov/acs/www.

Thank you in advance for your help.
Sincerely,

Charles Louis Kincannon

Director, U.S. Census Bureau

ACS-12(L)S (1-2003)

USCENSUSBUREAU
www.census.gov

Helping You Make Informed Decisions

B-1

Appendix B.2 Pre-Notice Letter (Puerto Rico Version)

UNITED STATES DEPARTMENT OF COMMERCE
Economics and Statistics Administration

U.S. Census Bureau
Washington, DC 20233-0001
OFFICE OF THE DIRECTOR

Dear Resident:
In a few days your household will receive a questionnaire in the mail for a very
important survey–the Puerto Rico Community Survey. When the questionnaire arrives,
please fill it out and mail it back promptly. The U.S. Census Bureau is conducting this
survey and chose your address, not you personally, as part of a randomly selected
sample.
The Puerto Rico Community Survey collects information about various topics like
education, housing, and jobs. Information from this survey is used by federal, state,
and municipio governments to meet the needs of communities in Puerto Rico. For
example, the government uses this information to decide where schools, highways,
hospitals, and other services are needed. The survey also is used to develop
programs to reduce traffic congestion, provide job training, and plan for the healthcare
needs of the elderly.
If you have access to the Internet and want to learn more about the Puerto Rico
Community Survey, please visit the Census Bureau’s Web site:
www.census.gov/acs/www.
Thank you in advance for your help.
Sincerely,

Charles Louis Kincannon
Director

ACS-12(L)PR (12-2003)

USCENSUSBUREAU
Helping You Make Informed Decisions

www.census.gov

B-2

USCENSUSBUREAU

YOUR RESPONSE IS
REQUIRED BY LAW

The American Community Survey
Form Enclosed

ACS-46(2003) (1-2004)

OFFICIAL BUSINESS
Penalty for Private Use $300

U.S. CENSUS BUREAU
Jeffersonville, IN 47132-0001

Economics and Statistics Administration

U.S. DEPARTMENT OF COMMERCE

Appendix B.3 Outgoing Envelope (U.S. Version)

B-3

AN EQUAL OPPORTUNITY EMPLOYER

PRSRT
FIRST-CLASS MAIL
POSTAGE & FEES PAID
U.S. Census Bureau
Permit No. G-58

USCENSUSBUREAU

ACS-46(2005)PR (1-2004)

OFFICIAL BUSINESS
Penalty for Private Use $300

U.S. CENSUS BUREAU
Jeffersonville, IN 47132-0001

Economics and Statistics Administration

U.S. DEPARTMENT OF COMMERCE
AN EQUAL OPPORTUNITY EMPLOYER

B-4

YOUR RESPONSE IS REQUIRED BY LAW

The Puerto Rico Community Survey Form Enclosed

SU RESPUESTA ES REQUERIDA POR LEY

Se Incluye Cuestionario de la Encuesta sobre la Comunidad de Puerto Rico

Appendix B.4 Outgoing Envelope (Puerto Rico Version)
PRSRT
FIRST-CLASS MAIL
POSTAGE & FEES PAID
U.S. Census Bureau
Permit No. G-58

Appendix B.5 Initial Mail Package Cover Letter (U.S. Version)

UNITED STATES DEPARTMENT OF COMMERCE
Economics and Statistics Administration

U.S. Census Bureau
Washington, DC 20233-0001
OFFICE OF THE DIRECTOR

Dear Resident:
I recently sent a letter to your household about the American Community Survey.
Enclosed is a questionnaire and information about the survey. Please complete the
questionnaire and mail it back as soon as possible in the postage-paid envelope.
This survey collects critical up-to-date information used to meet the needs of
communities across the United States. For example, results from this survey are
used to decide where new schools, hospitals, and fire stations are needed. This
information also helps communities plan for the kinds of emergency situations that
might affect you and your neighbors, such as floods and other natural disasters.
The U.S. Census Bureau chose your address, not you personally, as part of a
randomly selected sample. You are required by U.S. law to respond to this survey.
The Census Bureau is required by U.S. law to keep your answers confidential. The
enclosed brochure answers frequently asked questions about the survey.
If you need help filling out the questionnaire, please use the enclosed guide or
call our toll-free number (1–800–354–7271).
Thank you.
Sincerely,

Charles Louis Kincannon
Director, U.S. Census Bureau
Enclosures

ACS-13(L)SM (12-2002)

USCENSUSBUREAU
www.census.gov

Helping You Make Informed Decisions

B-5

Appendix B.6 Initial Mail Package Cover Letter (Puerto Rico Version)

UNITED STATES DEPARTMENT OF COMMERCE
Economics and Statistics Administration

U.S. Census Bureau
Washington, DC 20233-0001
OFFICE OF THE DIRECTOR

Dear Resident:
I recently sent a letter to your household about the Puerto Rico Community Survey.
Enclosed is a questionnaire and information about the survey. Please complete the
questionnaire and mail it back as soon as possible in the enclosed postage-paid envelope.
This survey collects critical up-to-date information used to meet the needs of communities
across Puerto Rico. For example, results from this survey are used to decide where new
schools, hospitals, and fire stations are needed. This information also helps communities
plan for the kinds of emergency situations that might affect you and your neighbors, such
as floods and other natural disasters.
The U.S. Census Bureau chose your address, not you personally, as part of a randomly
selected sample. You are required by U.S. law to respond to this survey. The Census
Bureau is required by U.S. law to keep your answers confidential. The enclosed brochure
answers frequently asked questions about the survey.
If you need help filling out the questionnaire, please use the enclosed guide or call our
toll-free number (1-800-717-7381).
Thank you in advance for your cooperation.
Sincerely,

Charles Louis Kincannon
Director
Enclosures

ACS-13(L)PR (3-2004)

USCENSUSBUREAU
Helping You Make Informed Decisions

www.census.gov

B-6

Appendix B.7 ACS Questionnaire (U.S. Version)

U.S. DEPARTMENT OF COMMERCE
Economics and Statistics Administration
U.S. CENSUS BUREAU

THE

American Community Survey

Start Here

This form asks for three types of information:
• basic information about the people who are living or staying at
the address on the mailing label above
• specific information about this house, apartment, or mobile
home
• more detailed information about each person living or staying
here

People are our most important resource.
This Census Bureau survey collects
information about education,
employment, income, and housing—
information your community uses to
plan and fund programs. Your
response is important, and we
keep your answers confidential.

This booklet shows the
content of the
American Community Survey
questionnaire.

➜

What is your name? Please PRINT the name of the person who
is filling out this form. Include the telephone number so we can
contact you if there is a question, and today’s date.

Last Name

First Name
If you need help or have questions
about completing this form, please call
1-800-354-7271. The telephone call is free.

Area Code + Number

Telephone Device for the Deaf (TDD):
Call 1–800–582–8330. The telephone call is free.
Date (Month/Day/Year)

¿NECESITA AYUDA? Si usted habla español
y necesita ayuda para completar su cuestionario,
llame sin cargo alguno al 1–877–833–5625.

➜

For more information about the American
Community Survey, visit our web site at:
http://www.census.gov/acs/www/

How many people are living or staying at this address?
Number of people

➜
USCENSUSBUREAU

Please turn to the next page to continue.

ACS-1(INFO)(2005)

FORM
(5-20-2004)

ACS-1(INFO)(2005), Page 1, Base (Black)

B-7

OMB No. 0607-0810

ACS-1(INFO)(2005), Page 1, GREEN Pantone 354 (20% and 100%)

List of Residents
READ THESE
INSTRUCTIONS
FIRST

1 What

is this
person’s
sex?

2 What is this person’s

age and what is this
person’s date of birth?
Print numbers in boxes.

X Person 1

Age (in years)

Last Name (Please print)

Month Day

Age (in years)

Last Name (Please print)
Male
First Name

Female

Month Day

Year of birth

Age (in years)

Male

First Name

Year of birth

Month Day

Last Name (Please print)

Age (in years)
Male

Month Day

Father or mother
Grandchild
In-law
Other relative

Husband or wife

Roomer, boarder

Son or daughter
Brother or sister

Housemate,
roommate
Unmarried partner
Foster child
Other nonrelative

Father or mother
Grandchild
In-law
Other relative

Husband or wife

Roomer, boarder

Son or daughter
Brother or sister

Housemate,
roommate
Unmarried partner
Foster child
Other nonrelative

Year of birth

Grandchild
In-law
Other relative

Relationship of Person 5 to Person 1.
Age (in years)

Last Name (Please print)
Male
Female

First Name

➜

Housemate,
roommate
Unmarried partner
Foster child
Other nonrelative

Brother or sister

Father or mother

Female

Person 5

➜

Roomer, boarder

Son or daughter

Relationship of Person 4 to Person 1.

Person 4

First Name

Female

Husband or wife

Relationship of Person 3 to Person 1.

Last Name (Please print)

IF YOU ARE NOT SURE
WHOM TO LIST, CALL
1–800–354–7271.

(Person 1 is the person living or staying
here in whose name this house or
apartment is owned, being bought, or
rented. If there is no such person, start
with the name of any adult living or
staying here.)
Relationship of Person 2 to Person 1.

Person 2

Person 3
If this place is a
vacation home or a
temporary residence
where no one in this
household stays for more
than 2 months, do not
list any names in the List
of Residents. Complete
only pages 4, 5, and 6
and return the form.

Year of birth

• DO NOT LIST anyone who
is living somewhere else
for more than 2 months,
such as a college student
living away.

Female

• LIST anyone else staying
here who does not have
another usual place to
stay.

First Name

• LIST everyone who is
living or staying here for
more than 2 months.

to Person 1?

Person 1
Male

Please fill out this form
as soon as possible after
receiving it in the mail.

3 How is this person related

Month Day

Year of birth

Husband or wife

Roomer, boarder

Son or daughter
Brother or sister

Housemate,
roommate
Unmarried partner
Foster child
Other nonrelative

Father or mother
Grandchild
In-law
Other relative

If there are more than
five people, list them
here. We may call you
for more information
about them.

Person 6

Person 7

Person 8

Last Name (Please print)

After you’ve created
the List of Residents,
answer the questions
across the top of the
page for the first five
people on the list.

First Name

2
ACS-1(INFO)(2005), Page 2, GREEN Pantone 354 (10%, 20% and 100%)

ACS-1(INFO)(2005), Page 2, Base (Black)

B-8

person’s
marital
status?

NOTE: Please answer BOTH Questions 5 and 6.

5 Is this person Spanish/

6 What is this person’s race? Mark (X) one or more races to indicate what this

Hispanic/Latino?
Mark (X) the "No" box if
not Spanish/Hispanic/Latino.

person considers himself/herself to be.

White

Now married
Widowed
Divorced
Separated
Never married

No, not Spanish/Hispanic/Latino
Yes, Mexican, Mexican Am.,
Chicano
Yes, Puerto Rican
Yes, Cuban
Yes, other Spanish/Hispanic/
Latino — Print group.

Now married
Widowed
Divorced
Separated
Never married

No, not Spanish/Hispanic/Latino
Yes, Mexican, Mexican Am.,
Chicano
Yes, Puerto Rican
Yes, Cuban
Yes, other Spanish/Hispanic/
Latino — Print group.

White

Now married
Widowed
Divorced
Separated
Never married

No, not Spanish/Hispanic/Latino
Yes, Mexican, Mexican Am.,
Chicano
Yes, Puerto Rican
Yes, Cuban
Yes, other Spanish/Hispanic/
Latino — Print group.

White

Now married
Widowed
Divorced
Separated
Never married

No, not Spanish/Hispanic/Latino
Yes, Mexican, Mexican Am.,
Chicano
Yes, Puerto Rican
Yes, Cuban
Yes, other Spanish/Hispanic/
Latino — Print group.

White

Now married
Widowed
Divorced
Separated
Never married

No, not Spanish/Hispanic/Latino
Yes, Mexican, Mexican Am.,
Chicano

Black or African American
American Indian or Alaska
Native – Print name of enrolled
or principal tribe.

Asian Indian
Chinese
Filipino
Japanese
Korean
Vietnamese
Other Asian –
Print race.

Native Hawaiian
Guamanian or Chamorro
Samoan
Other Pacific Islander – Print race below.
Some other race – Print race below.