CONTENTS
Page
I. Objectives and Strategies. . . . . . . . . . . . . . . . I-1
II. Content and Organization of the Report . . . . . . . . . II-1
III. Legal Requirements . . . . . . . . . . . . . . . . . . . III-1
IV. Marketing and Partnership Program. . . . . . . . . . . . IV-1
V. Questionnaire Content, Data Collection Forms, and
Sampling Plan. . . . . . . . . . . . . . . . . . . . . . V-1
Questionnaire Content. . . . . . . . . . . . . . . . . . V-1
Mailout Questionnaire Forms. . . . . . . . . . . . . . . V-6
Field Data Collection Forms. . . . . . . . . . . . . . . V-9
Sampling Plan for the Long-Form Questionnaire. . . . . . V-10
VI. Address List Development and Review/Update . . . . . . . VI-1
Address List Development . . . . . . . . . . . . . . . . VI-1
Program for Address List Supplementation . . . . . . . . VI-4
Local Update of Census Addresses Program . . . . . . . . VI-5
VII. Geographic Data Base Development - TIGER . . . . . . . . VII-1
VIII. Field Office Infrastructure and Staffing . . . . . . . . VIII-1
Field Office Infrastructure. . . . . . . . . . . . . . . VIII-1
Field Office Staffing. . . . . . . . . . . . . . . . . . VIII-4
IX. Field Data Collection. . . . . . . . . . . . . . . . . . IX-1
Planning Database. . . . . . . . . . . . . . . . . . . . IX-1
Basic Enumeration Strategy . . . . . . . . . . . . . . . IX-3
Telephone Assistance and the Internet. . . . . . . . . . IX-4
Be Counted National Campaign . . . . . . . . . . . . . . IX-5
Procedures to Enumerate Special Populations. . . . . . . IX-7
Enumeration of People With No Usual Residence. . . . . . IX-10
Special Data Collection Methods for Targeted Areas . . . IX-12
Vacant Housing Unit Followup . . . . . . . . . . . . . . IX-14
Large Household Followup . . . . . . . . . . . . . . . . IX-15
Unduplication of Multiple Responses. . . . . . . . . . . IX-16
Coverage Edit. . . . . . . . . . . . . . . . . . . . . . IX-18
Nonresponse Followup Strategy. . . . . . . . . . . . . . IX-19
Quality Check Survey . . . . . . . . . . . . . . . . . . IX-21
Potential Effect of Nonresponse Followup and
Quality Check Sampling Options . . . . . . . . . . . . . IX-24
Demographic Analysis . . . . . . . . . . . . . . . . . . IX-26
X. American Indian and Alaska Native Areas
and Hawaiian Homelands . . . . . . . . . . . . . . . . . X-1
XI. Telecommunications Support and
Automated Data Processing. . . . . . . . . . . . . . . . XI-1
Telecommunications Support . . . . . . . . . . . . . . . XI-1
Data Capture System. . . . . . . . . . . . . . . . . . . XI-3
Automated Data Processing System . . . . . . . . . . . . XI-5
XII. Dissemination and Products . . . . . . . . . . . . . . . XII-1
Tabulation and Dissemination Program . . . . . . . . . . XII-1
P.L. 94-171 Redistricting Data Program . . . . . . . . . XII-3
Geographic Products. . . . . . . . . . . . . . . . . . . XII-5
XIII. Testing, Dress Rehearsal, Evaluation, and Research . . . XIII-1
Census 2000 Tests and Research . . . . . . . . . . . . . XIII-1
United States Census 2000 Dress Rehearsal. . . . . . . . XIII-3
Quality Assurance. . . . . . . . . . . . . . . . . . . . XIII-6
Census 2000 Evaluation Program . . . . . . . . . . . . . XIII-8
Research and Experimentation Program . . . . . . . . . . XIII-9
Administrative Records . . . . . . . . . . . . . . . . .XIII-10
2010 Census Planning . . . . . . . . . . . . . . . . . .XIII-12
XIV. Puerto Rico. . . . . . . . . . . . . . . . . . . . . . . XIV-1
XV. Island Areas . . . . . . . . . . . . . . . . . . . . . . XV-1
Appendixes
A. Glossary of Abbreviations and Acronyms
B. Key Census Bureau Telephone Contacts
---------------------------------------------------------------------
|The Census 2000 Operational Plan is subject to change based upon |
|Congressional funding and questionnaire content changes, the results |
|of our testing and research, the advice generated from our ongoing |
|consultation process with stakeholders, and on what occurs as we |
|begin operationalizing and implementing the plan's specific elements.|
---------------------------------------------------------------------
The next census of the United States' population and housing will be conducted
as of April 1, 2000. Reflecting a long tradition, Census 2000 will be the
22nd decennial enumeration in an unbroken chain that our Nation has undertaken.
In many significant respects, however, Census 2000 will deviate sharply from
tradition.
As this Nation moves forward into a new century, the decennial census also
must move forward. The Census Bureau has developed a plan for conducting
Census 2000, incorporating many new features that address the two concerns
that many people had about the 1990 census: that it cost too much and that
it did not include everyone. The Census 2000 operational plan redesigns
the census process in bold and fundamental ways.
From apportioning the U.S. House of Representatives to providing the data used
by communities, businesses, and Americans everywhere, the decennial census
is the cornerstone of our knowledge about our Nation. The census is the only
data gathering operation in the United States that is mandated by the
Constitution and the only one that produces a broad array of information on
the American people and their housing at the smallest geographic levels.
Objectives of Census 2000
The goal of every census is to be the best census ever. So it is with
Census 2000. The Census 2000 operational plan contains strategies to
improve the completeness and reduce the cost of the census. The
following objectives are fundamental to our efforts:
o Make every effort to count every household and person--from simpler,
user-friendly forms to the better design of field operations
o Maintain an open process that diverse groups and interests can understand
and support
o Eliminate the differential undercount of racial and ethnic groups
o Produce a "one-number" census that is right the first time
Four Strategies for Fundamental Change
The Census Bureau's operational plans for Census 2000 are built around
four fundamental strategies for change:
Strategy One: Build Partnerships at Every Stage of the Process
The Census Bureau cannot accomplish its goals alone. So for Census 2000,
we are reaching out and forming partnerships to help us accomplish our
objectives. We need to think in terms of every activity being done by a
"best in class" provider. This means the Census Bureau must build:
o Partnerships with state, local, and tribal governments. These
governments know their local conditions and circumstances better than
the Census Bureau. They can help us correct our maps and address
lists, and tell us where to put unaddressed questionnaires, called
"Be Counted" forms, in locations where people will find them. They
also can alert us to problems and advise us of opportunities to
publicize Census 2000. The law now allows us to share our address
lists and related information--while maintaining their
confidentiality--with these governments and get their input.
o Partnerships with community groups. These groups know their
constituents better than either the Census Bureau or any other
governmental office. The groups can alert us to the best ways to
communicate with their constituents to ensure they are included.
o Partnership with the U.S. Postal Service (USPS). For Census 2000, the
Census Bureau will use address information provided by the USPS to
enhance our comprehensive address file. This effort will reduce the
number of costly field checks. The USPS also will deliver census
questionnaires to over 80 percent of the addresses nationwide.
o Partnership through privatization. To be "world class" in every
stage of the census, we will award contracts to private sector
partners, including:
o Advertising and promotion. We intend to use private companies to
manage our efforts to promote the census more visibly and effectively.
o Facilities management. We are considering contracts with data
processing companies to manage the facilities where completed census
forms are translated into computer files.
Strategy Two: Keep It Simple
The simpler and easier Census 2000 is, the greater the response, and the
more accurate and less expensive it will be. Simplicity is the goal for
every part of the process. For example:
o User-friendly forms. Our modern, powerful computer systems will
allow us to use forms that are easier to read and complete.
Moreover, because everyone is deluged with junk mail, Census 2000
questionnaires must be attractive, motivating (by explaining the
benefits and mandatory nature of the census), easy to understand,
and simple to fill out. The Census 2000 forms will stand out because
they will carry a well-publicized "census" identity. Private designers
are working with us to simplify the forms and implement the user-friendly
features demonstrated in our testing and research to increase response.
o Multiple contacts. We also have learned from our testing and
research that repeated contacts and reminders pay big dividends
in response. So for Census 2000, we will implement a multiple
mail contact strategy. The first contact with each address will
be a letter that alerts the recipient to the census and its benefits.
A few days later, a census questionnaire will arrive, noting that
"your response is required by law." Shortly thereafter, a postcard
will arrive thanking those who have participated and reminding
others to do so. Finally, another questionnaire will be sent
out, with a final message encouraging nonrespondents to reply.
o More ways to respond. Our first priority is to deliver a
census questionnaire at each address. For the first time,
however, we also will put unaddressed Be Counted questionnaires
in post offices and libraries, at stores and malls, in schools and
community centers, and at other places where people frequent.
There will be a well-publicized toll-free telephone number. We
also plan to mail census forms in another language to households
in areas where a significant number of people speak the other
language, along with an English-language questionnaire.
o Other simplified procedures
o A new method will ensure that Census 2000 finds people--such as those
with no usual residence--at shelters, soup kitchens, and other places
where they obtain services.
o Special targeted methods will improve the count for population
groups and in areas that historically have had large undercounts.
One example is enlisting community leaders to designate
neighborhoods where English is not the primary language for the
Census Bureau to include as part of the targeted mailing.
Strategy Three: Use Technology Intelligently
Dramatic advances in computing will allow Census 2000 to be simpler, less
costly, and more accurate. For example:
o Digital "capture" of forms. In Census 2000, for the first time,
we will scan most of the completed questionnaires directly into
computers that read handwriting. The completed forms will be read
directly into computer files that will be used later for tabulation.
o "Matching" software. Sophisticated computer software will allow
us to spot multiple responses from the same household. For example,
if one spouse returns a regular questionnaire by mail while the
other fills out a Be Counted questionnaire, we will be able to
determine that both records come from the same household.
o "Point and click" data tabulation. Data seekers will be able to
find the information they want from Census 2000. "Point and click"
computing from our new DADS system will allow them to select the
specific information they want, instead of thumbing through census
reports that may or may not have the answer they are looking for.
Strategy Four: Use Statistical Methods
Sampling and statistical estimation have been an integral part of
the census process since 1940. At one time, the census asked every
household for all the census information; now, most census questions
are asked of a sample of households.
In 1990, respondents who did not return their census forms by mail cost
the Census Bureau more than those who did, as temporary census workers
were needed to conduct personal visits with nonresponding households.
Statisticians agree that incorporating widely accepted statistical
methods into Census 2000 will produce better results at less cost.
So for Census 2000, we will make every attempt to find everyone. Some
households, however, will neither mail in nor phone in their response.
So we will use personal visits to obtain responses from the remaining
addresses, to ensure that we directly contact at least 90 percent in
each census tract.
Using sampling to gather information on nonrespondents will ensure that
Census 2000 is built on a solid core of responses. It will ensure that
we can complete our personal visits with no loss of accuracy but with
substantial savings of time and money. Sampling will allow us to make
scientific estimates of the population for the final 10 percent of
the housing units.
Our experience in the last six decennial censuses has demonstrated that
having responses from 100 percent of the housing units does not ensure
inclusion of 100 percent of the population. People are left out for many
reasons, and our objective is to account for everyone.
To check the quality of our work and to reach our goal of accounting
for 100 percent of the population, we will take a second independent
sample--of about 750,000 housing units--of the total population and
conduct a second interview. We will use the information from the second
interview as the basis for quality checking all our results: the results
from the mail returns, the Be Counted program, telephone interviews and
possibly the Internet, and personal followup visits. This quality check
survey will lead to a "one-number" census and will eliminate the need for
subsequent adjustment of the decennial count. We will use demographic
analysis to validate the results.
By using both kinds of sampling--that is, sampling for nonresponse and
quality check sampling--the accuracy of Census 2000 will be very high
for all states, congressional districts, and other populous areas.
There always will be some uncertainty surrounding population totals for
some smaller areas, such as census blocks, census tracts, or small
communities. Unlike previous censuses, for Census 2000, we will know
the level of uncertainty for all areas. The totals for historically
undercounted areas will be much better than with traditional methods.
Effective Management Tools
The Census Bureau has instituted several management initiatives--such as the
following--to facilitate a more effective and efficient planning process for
Census 2000 as well as its actual implementation:
o A sophisticated electronic Management Information System, with a Master
Activity Schedule component and a Cost and Progress component, will provide
information on scheduled dates, performing organization, budget, cost to
date, and current progress for Census 2000 operations. This system provides
decision support functions, such as critical path analysis, and what-if
analysis.
o The Commerce Administrative Management System is a modern electronic
financial management system which, among other features, provides up-to-date
financial data available for on-line query as well as paperless processing
for purchase orders and payments.
o The Census 2000 Cost Model provides an automated means to estimate
staffing and budget requirements for Census 2000 based on a well-defined set
of activities specific to the major components of census operations. The
Cost Model is the sole tool used to prepare the cost estimates for the
budget process. It also is used to answer inquiries from Congress, the
Department of Commerce, the Office of Management and Budget, senior
managers at the Census Bureau, and our stakeholders.
Using the various innovative and cost-saving methods that center around the
four strategies for conducting Census 2000, as well as the improved
management of census operations, cost modeling techniques have estimated
the cost of Census 2000 to be much less than if the 1990 census design were
repeated in 2000.
This report presents the major elements of the operational plan for Census
2000. For each element, it details the objective, major features, and
milestone dates.
Figure II-1 depicts the Census 2000 process. Early census activities must
provide a broad and firm foundation for effectively supporting successive
steps in the census process. In the end, the goal of Census 2000 is to
produce statistics that will serve our Nation well.
The organization of this report generally follows the steps in the census
process.
The milestone dates shown in each section are presented in terms of quarters
of the calendar year. The quarters and the months they include are as follows:
Q1 = January, February, and March
Q2 = April, May, and June
Q3 = July, August, and September
Q4 = October, November, and December
Questions or comments relating to specific aspects of the operational plan may
be directed to the appropriate person listed in Appendix B, "Key Census Bureau
Telephone Contacts."
-------------
| |
|Data |
|Dissemination |
| |
------------- -------------- ------------
| | | |
|Data |Data | Quality |
|Capture |Processing | Check |
| | | |
------------- -------------- ------------
| |Nonresponse | Special |
|Basic |Followup | Place |
|Enumeration |Enumeration | Enumeration|
| | | |
-------------- -------------- -------------- ------------ -------------
|Address List | |Field Office/ | Public-Use |Marketing |
|and Geographic|Content |Data Capture | Form |Partnership |
|Data Base |Development |Center | Development|Outreach |
|Development | |Infrastructure| |Promotion |
-------------- -------------- -------------- ------------ -------------
The decennial census is mandated by the U.S. Constitution (Article I,
Section 2) to provide the population counts needed to apportion the seats
in the U.S. House of Representatives among the states. However, the
Constitution does not prescribe how the decennial census should be conducted.
In Title 13, U.S. Code, the Congress gave the Secretary of Commerce (and by
delegation, the Director of the Census Bureau) discretion to enact
decennial census plans, subject to executive and congressional review.
The planning and conduct of Census 2000 must comply with a number of legal
requirements, some of which specify deadlines, as described in the following:
o The geographic scope of who we enumerate in a decennial census is specified
in Title 13 as covering the 50 states, the District of Columbia, the
Commonwealth of Puerto Rico, the Virgin Islands of the United States, the
Commonwealth of the Northern Mariana Islands, and Guam, and any other
areas as may be determined by the Secretary of Commerce. In Census 2000,
we also will enumerate the Pacific Island Area of American Samoa.
o The Secretary of Commerce, acting under the authority of Title 13, is
required to report to the Congress twice regarding the content of the
census questionnaires: first at least 3 years before Census Day (by
April 1, 1997) on the subjects proposed for inclusion, and again at least
2 years before the census (by April 1, 1998) on the proposed specific
question wording. Accordingly, on March 28, 1997, the Census Bureau
submitted to the Secretary of Commerce for transmission to the Congress
the list of subjects proposed for inclusion in Census 2000.
o All subjects submitted to Congress had a strong legislative justification
for being included. They were either specifically mandated or strongly
implied by Federal law. Congress has enacted laws requiring the use of
census data to determine how much Federal funding to allocate to states,
cities, school districts, and other governmental units to administer a
wide variety of important programs.
o The Office of Management and Budget (OMB) Statistical Policy Directive
No. 15 currently requires that race and ethnic data gathered by Federal
agencies be collected for the following set of categories: American Indian
or Alaskan Native, Asian or Pacific Islander, Black, White, and Hispanic.
The Directive allows for the collection of additional categories, so long
as they can be grouped into the basic categories. The OMB is reviewing
this Directive for possible changes in the categories and will issue a
decision in the Fall of 1997 as to whether the categories will or will
not change. If the racial classification standards are revised, OMB
will provide guidelines. The decision will be crucial in determining
the content of the race and ethnic questions in Census 2000.
o Public Law (P.L.) 94-311 requires the use of Spanish-language forms and
Spanish-speaking interviewers in areas having significant concentrations of
Hispanic populations. In Census 2000, for the first time, we will include
a Spanish-language census questionnaire along with an English-language form
in the mailout package for these areas.
o Before the census forms go to print, the OMB is required by law to review
the questions to ensure they meet the data needs of the Executive Branch
departments and agencies responsible for implementing programs mandated
and authorized by the Congress. In addition, under the Paperwork Reduction
Act (Title 44), the OMB must see that the time burden for a household to
respond to the questionnaire is held to a minimum.
o P.L. 103-430 requires that the United States Postal Service provide its
address information to the Census Bureau to improve the Master Address
File (MAF).
o Established in response to the requirements of P.L. 103-430, the LUCA
program (Local Update of Census Addresses) provides an opportunity for local
and tribal officials to designate a liaison to review the address information
in the MAF for their jurisdiction and the geographic information in the
Census Bureau's geographic database (TIGER) to improve their completeness
and accuracy.
o As specified in Title 13, Census Day for Census 2000 is April 1, as it has
been for each decennial enumeration since 1930. All census questions
generally are to be answered with reference to April 1, regardless of
when the questionnaire is filled out.
o Title 13 guarantees the confidentiality of respondents' answers to the
census forms. In fact, the Census Bureau takes extraordinary steps
throughout the entire census process to assure the confidentiality of
census information. All Census Bureau employees must take an oath of
confidentiality. As required by Title 13, the Census Bureau maintains
tight security over completed questionnaires. Furthermore,
disclosure-avoidance programs during the data tabulation phase make
certain that individual persons or housing units cannot be identified,
either from paper or electronic tabulations.
o Title 44 specifies that individual census information from the decennial
census cannot be made public for 72 years.
o Under the terms of Title 13, the Secretary of Commerce must deliver state
population counts to the President within 9 months of Census Day (by
December 31, 2000). These counts are used to reapportion the seats in
the U.S. House of Representatives.
o P.L. 94-171 requires that the Census Bureau provide selected census
tabulations to the states by April 1 of the year following the census year.
The states use these tabulations to redraw the areas included in each
Congressional district for purposes of reapportionment. The states also
use these tabulations to redraw the areas included in other districts
used for redistricting plans for state and local elections.
o Under the Voting Rights Act, the Census Bureau is required to provide to
the states race and ethnic data for small geographic areas to be used
for the redistricting process specified in P.L. 94-171. The race and
ethnic categories required are those mandated by OMB Statistical
Policy Directive No. 15 (currently under review).
OBJECTIVEThe Census 2000 Marketing and Partnership Program will be, for the first time, an integrated communications effort. The objectives of the campaign are to increase awareness of Census 2000 and boost response rates. MAJOR FEATURES Prior to Census Day, the Marketing and Partnership Program will be geared toward building awareness that the census is approaching and how it will benefit the community. During the mailout/mailback period, the campaign focus will shift to motivating people to return their questionnaires promptly to increase the initial mail response rate. The marketing program also will encourage cooperation with census enumerators during the followup operation with nonresponding households and will let people know the census is "not over" during the quality check survey. The comprehensive marketing and partnership strategy includes: o Partnerships and community outreach o Paid advertising o Special methods to encourage response o Traditional public relations o Special events Partnerships and Community Outreach The Census Bureau has begun forming partnerships with other Federal agencies, state, local and tribal governments, as well as with community-based organizations and businesses. The Bureau recognizes the unique local knowledge, experience, and expertise these partners can bring to planning and taking an accurate census. Partnerships have each member performing those activities for which it is best qualified, assuring the most effective expenditure of staff and financial resources. To establish and maintain continuing liaison and partnership with government and non-government entities, we will hire three types of Census Bureau partnership specialists throughout the country: government, media, and community specialists. Twelve government partnership specialists--one for each Census Bureau regional office--already have been hired. The Census 2000 Publicity Office will coordinate the full range of Census 2000 programs with governmental and nongovernmental partners to ensure that we do not make unnecessary or overlapping requests of those willing to work in partnership. Census 2000 will provide numerous opportunities for government and nongovernment entities to participate in partnership activities. Examples of these activities include: o The Census Bureau has formed a partnership with the U.S. Postal Service (USPS) to use its address information to enhance the Census 2000 Master Address File. Working with the USPS provides the best nationwide source of address updates, which will help reduce the number of households missed in Census 2000. o Under the LUCA program, partnerships with local and tribal governments are being formed to provide valuable assistance in reviewing and updating the Master Address File. o Partnerships with governments and organizations will support Census 2000 promotional activities by issuing public statements of endorsement, holding press conferences, placing census articles in newsletters, including census messages in employee paychecks, sponsoring census promotional events, and posting census promotional material in agency facilities. o Partnerships with local organizations will aid in recruiting candidates to apply for census office and field enumerator positions. Temporary jobs will be available in all geographic areas. o Partnerships with community-based organizations and local and tribal governments will identify strategic and high visibility locations in the community to place unaddressed questionnaires, called "Be Counted" forms, for people to pick up and complete. These partners also will recommend locations for Walk-in Questionnaire Assistance Centers and for office space to test and train temporary census workers. o During the enumeration period, partnership and outreach efforts will remind people of ways to respond if they did not receive a questionnaire. Mail response rates will be available so that outreach/partnerships efforts can target slow mail response areas. Paid Advertising For the first time, a paid advertising campaign will be used for a decennial census. In designing the Census 2000 paid campaign, we have benefitted greatly from the advice given by the U.S. military, the Postal Service, and private communications contractors, as well as from our own survey and focus group research. The major components will include: o A national media campaign, including TV (both broadcast and cable), radio, and print media, will be aimed at increasing mail response. o Using national, regional, and local media outlets, a contractor will design and implement a flexible advertising effort directed at increasing mail response among targeted audiences, especially traditionally undercounted populations. The local effort will use, for example, community news outlets, posters, flyers, and mass transit advertising. Special Methods to Encourage Response o Integrated mailing package. For the first time, the mailing package design--including the questionnaires, envelopes, motivational slogans, and logo--will be wholly compatible and integrated into the design of the rest of the marketing plan. o Direct mail campaign and mail strategy. The Census Bureau will publicize its full-scale direct mail campaign that is being designed to increase opportunities for people to be included in the census at the lowest possible cost. In addition, to increase questionnaire mail response, a new strategy will focus on multiple mail contacts with respondents, including mailing respondents an advance notice letter, an initial questionnaire, a thank you/reminder postcard, and a replacement questionnaire. o Other ways to respond. Also for the first time, special unaddressed questionnaires, called "Be Counted" forms, will be available at community centers, post offices, schools, libraries, convenience stores, Walk-in Questionnaire Assistance Centers, and other public locations where people frequent. Moreover, a well-publicized toll-free telephone number and possibly an Internet address will assist those who request to respond to the census by these methods. Traditional Public Relations For Census 2000, the public relations effort will be decentralized with media specialists assigned directly to local census offices to cultivate press contacts and respond to media inquiries. Special Events A variety of special events--such as parades, athletic events, public service television documentaries--will be co-sponsored by state, local, and tribal governments and by community organizations and businesses. The events will emphasize the importance of participating in the census and will motivate people to respond. MILESTONES Q4 1996 Hired 12 government partnership specialists (one per region) Q4 1996 Began forming partnerships with local and tribal governments for geographic programs Q1 1997 Began forming partnerships with national/umbrella governmental and nongovernmental organizations Q1 1998 Award contract for paid advertising campaign Q1 1998 Hire media, community, and remaining government partnership specialists Q1 1998 Begin forming partnerships with local media, community organizations, and businesses Q2 1998 Implement prototype advertising campaign for Census 2000 Dress Rehearsal
OBJECTIVEThe goal in selecting Census 2000 questionnaire content is to meet the many statutory data requirements of Federal agencies, as well as the needs of state, local, and tribal governments to administer governmental programs. Given the many critical uses of census data, it is essential that the Census 2000 questionnaires contain those topics that will produce data our Nation will need as it enters the 21st Century. Nevertheless, the Census Bureau must balance the many demands for census information against the length of the questionnaires and the burden on the respondents to complete them. MAJOR FEATURES The content determination process is used to select the questions to be asked on the census forms. The objective of the Census 2000 content process is to develop questions that are easy to understand and answer by all segments of the population and thus yield the highest and most valid response. The major components and general timing of Census 2000 content development activities are depicted in Figure V-1. o Shortly after the 1990 census, the Census Bureau reviewed and evaluated the 1990 census questions through a content reinterview. o The Census Bureau then organized an extensive review and consultation program to determine which subjects should be included in Census 2000. o We assessed the legislative and geographic requirements of census data for both the Federal and non-Federal sectors. Federal agencies were asked to identify all legal mandates and programs requiring census data. Non-Federal requirements were obtained by means of a survey directed to a broad spectrum of data users such as state, local, and tribal governments; ethnic and community organizations; the business sector; academic researchers and librarians; religious groups; and the general public. o We also maintained regular contact with our standing advisory committees, expert panels, professional associations, housing data-user groups, and community and ethnic organizations to ensure that all segments of the data-user community would be kept informed throughout the content development process.
------------------------- |Evaluation of 1990 Census| | Questions | ------------------------- | ------------|------------ |Assessment of Legislative| | and Geographic | | Requirements | ------------------------- | ___________________________|___________________________ | | ----------|----------- -----------|----------- | Federal Agencies | | 1994 Survey of Census | |(Federal Requirements)| | Needs of Non-Federal | | | |Data Users (Non-Federal| ---------------------- | Requirements) | | ----------------------- | | _______________________________________________________ | ------------|------------ | Other Sources of | | Recommendations for | | Content | ------------------------- ____________________________|____________________________ | | | ---------|------------ ------------|------------ ---------|------- | Advisory Committees | |Professional Associations| | General Public | | and Expert Panels | | | | | ---------------------- ------------------------- ----------------- | | | |____________________________|____________________________| | ------------|------------ | Formulate Content for | | Research and Testing | ------------------------- ____________________________|_________________________ | | | ---------|------------ ------------|------------ ---------|---------- | Focus Group Studies | | 1996 National | |1996 Race and Ethnic| |and Cognitive Research| | Content Survey | | Targeted Test | ---------------------- ------------------------- -------------------- | | | ____________________________|_________________________ | ------------------------- | Analysis of Research and| | Test Results | ------------------------- | ------------|------------ | Submit 2000 Subjects to | | Congress | | April 1, 1997 | ------------------------- | ------------|------------ | Submit 2000 Question | | Wording to Congress | | April 1, 1998 | ------------------------- | ------------|------------ | Census 2000 | | Dress Rehearsal | | April 4, 1998 | ------------------------- | ------------|------------ | Clearance of 2000 Forms | | by Office of Management | | and Budget - Fall 1998 | ------------------------- | ------------|------------ | Census 2000 | | April 1, 2000 | ------------------------- o Two census tests were conducted to evaluate the questionnaire content proposed for Census 2000. The most extensive of these was the 1996 National Content Survey (formally known as the U.S. Census 2000 Test), which was designed to test new and revised question wording, formatting, and sequencing. The 1996 Race and Ethnic Targeted Test (formally known as the 1996 Census Survey) examined several major possible changes to the race and ethnic questions for Census 2000. During this time, the Census Bureau also conducted a wide range of focus group studies and cognitive research to elicit information about questionnaire content and design. o On March 28, 1997, as required by law, the Census Bureau submitted a list of subjects planned for inclusion in Census 2000 (see Table V-1) to the Secretary of Commerce for transmission to Congress. As the chart below shows, both the short and long forms proposed for Census 2000 have fewer subjects than their 1990 census counterparts. _______ Proposed 2000 Short Form| |7 subjects |_______|____ 1990 Short Form | |12 subjects |____________|_____________________ Proposed 2000 Long Form | |34 subjects |__________________________________|____ 1990 Long Form | |38 subjects |_______________________________________| | o The law requires that the actual questions for Census 2000 be submitted to the Congress by April 1, 1998. Thus, a determination must be made before that time about the particular wording, format, and sequence of individual questions. To make these critical decisions, the Census Bureau will be guided by: o Any budget and content constraints imposed by the Congress o The results of our Census 2000 research and testing program o The advice generated from our ongoing consultation process with stakeholders o The legislative requirements for data from the questions
POPULATION HOUSINGName Tenure (whether home is owned or Sex rented) Age Relationship Hispanic origin Race SAMPLE SUBJECTS POPULATION HOUSING Social characteristics: Physical characteristics Marital status Units in structure Place of birth, citizenship, and Number of rooms year of entry Number of bedrooms Education-school enrollment and Plumbing and kitchen facilities educational attainment Year structure built Ancestry * Year moved into unit Residence 5 years ago (migration) House heating fuel Language spoken at home Telephone Veteran status Vehicles available Disability Farm residence Grandparents as caregivers ** Financial characteristics Economic characteristics: Value of home Labor force status (current) Monthly rent (including congregate Place of work and journey to work housing) Work status last year Shelter costs Industry, occupation, and class of worker ==================================== Income (previous year) NOTE: Two additional subjects will be added if the Federal Government's ----------------------------------------- official measure of poverty is *Ancestry may be combined as one question redefined before Census 2000: with race and ethnicity and included as a 1) Support expenditures and health 100-percent subject, if government-wide coverage and standards for the classification of such 2) Household noncash benefits. data are revised by the Federal Government later in 1997. ==================================== **New subject for Census 2000. 1990 CENSUS SUBJECTS DROPPED FOR CENSUS 2000 POPULATION HOUSING Children ever born (fertility) Source of water Year last worked Sewage disposal Condominium status o The questions we plan to ask in the Census 2000 Dress Rehearsal will be, to the greatest extent possible, the same as those we subsequently include in Census 2000. In this way, we can develop prototypes of the products planned for Census 2000, solicit comments from our stakeholders, and fine-tune the products for the census. o The Census Bureau also is required by law to submit the recommended questions to the Office of Management and Budget (OMB), which has the responsibility of ensuring that the questions meet essential data needs and that respondent burden (the time it takes for the average household to fill out a questionnaire) is held to a minimum. o In response to a request from Congress, the Census Bureau is working with the OMB to investigate the possibility of having other Federal agencies share the cost of collecting, processing, and disseminating data from the Census 2000 long form. MILESTONES 3/28/1997 Submitted subjects planned for Census 2000 to the Congress 4/1/1998 Submit questions planned for Census 2000 to the Congress Q3 1998 Submit questions planned for Census 2000 to OMB Q4 1998 Prepare and award printing contracts Q1 1999 Test printing runs for technical review Q3 1999 Begin questionnaire printing
OBJECTIVEThe goals in developing the Census 2000 mailout questionnaires are to increase mail response and the accuracy of the information collected. The Census Bureau intends to do this by: o Designing forms that are more respondent-friendly, and o Increasing the number of mail contacts with respondents MAJOR FEATURES In Census 2000, the questionnaire mailout/mailback system again will be the primary means of census-taking. Cities, towns, suburban areas, and rural areas where city style mailing addresses (house number and street name) exist generally will comprise the mailout/mailback areas. Respondent-Friendly Design The Census Bureau has been working with private sector designers to produce more streamlined forms that are easy to read and understand, show people why they are asked the questions, and are simple to fill out and mail back. One key innovation is that the design of the complete mailing package--including the outgoing and return envelopes, cover letter, questionnaire, motivational slogans, and logo--will be compatible and integrated with the rest of the marketing and communications effort. The following user-friendly design features have been shown in our testing and research program to improve response and are being incorporated into the design of the Dress Rehearsal and Census 2000 forms: o A larger, easier-to-read font o Graphic icons distributed throughout the forms to illustrate the benefits of the census to the individual and community o Strong visual contrast--using color and shading--between the questions and answer boxes to make it easier to identify the correct space to answer o All questions for each household member grouped together in one space instead of in the row-column answer format with the questions placed vertically down the left-hand side of the page and the names of household members place horizontally across the top o Navigational aids such as arrows to guide the respondent through the questionnaire o Putting the respondent instructions directly on the form instead of in a separate guide In redesigning the forms, the Census Bureau also is incorporating the specifications required for printing, postal delivery, and electronic image data capture. Types of Mailout Questionnaires Census 2000 will include two types of questionnaires for mailout: o A "short" form will be delivered to approximately 83 percent of all housing units. It will include the basic population and housing questions pertaining to each household member (up to 5 people) and housing unit. A continuation roster will allow the respondent to list up to 12 household members. o A "long" form will be delivered to a sample--approximately 17 percent--of all housing units. It will include the short-form questions as well as additional questions on the characteristics of each household member (up to 5 people) and the housing unit. Obtaining these detailed, more comprehensive data on a sample basis is less costly than obtaining the same information from all housing units. A traditional roster will allow the respondent to list up to 12 household members. Mailout Questionnaires in Other Languages Questionnaires in English will be mailed to every housing unit. For the first time in a decennial census, specific neighborhoods having a significant concentration of Spanish-speaking households will be targeted for delivery of Spanish-language questionnaires, both short and long form, along with the English-language questionnaires. While the Census Bureau has made Spanish-language questionnaires available in the past, they have never before been included in the initial mailout package. This procedure has been shown in our census tests to improve the mail response of the Spanish-speaking population by making it more convenient and easier to respond. We also are researching the feasibility of expanding this approach for other languages, such as those spoken by the Asian community. An English-language questionnaire would be included in the mailing package being delivered with the other-language questionnaire. Multiple Mailing Strategy The Census Bureau is investigating the operational issues concerned with conducting a new mailing strategy for Census 2000. This strategy--which has been demonstrated in our testing and research to boost response--increases the number of mail contacts we have with respondents. The multiple mail contacts consist of: o An advance notice letter to every mailout address that alerts households the census form is being sent to them soon o An initial questionnaire to every mailout address o A postcard to every mailout address that serves as a thank you for respondents who have mailed back their questionnaire or as a reminder to those who have not o A replacement questionnaire A full-scale multiple mailing strategy, using first-class postage for all mailing pieces, will yield maximum mail response and increase the likelihood of delivery to the correct address. In areas lacking city style addresses, the U.S. Postal Service will deliver an advance notice letter and thank you/reminder postcard to every "Postal Patron" so these people will be alerted to the census. MILESTONES Q4 1998 Government Printing Office awards contracts for printing short form and long form Q3 1999 Begin questionnaire printing Q3 1999 Begin addressing questionnaires for mailing
OBJECTIVEWhile the mailout questionnaires will account for the bulk of Census 2000 data collection, the Census Bureau is developing many other forms to ensure that everyone has the opportunity to participate in Census 2000. These special forms will be used to enumerate people who live in a residence other than the usual house, apartment, or mobile home, or to increase the participation of people who might otherwise go uncounted in the census. MAJOR FEATURES o Several types of questionnaires--containing only population questions for one person--will be used to enumerate specific segments of the population. These forms will be used to count people in living arrangements requiring special operations, such as military bases, ships, college dormitories, nursing homes, shelters, and prisons. Long-form versions are being developed for some of the forms; many will be translated into Spanish. o Short- and long-form "simplified enumerator questionnaires" are being developed that are worded to conform to a personal interview method of data collection. These forms will be used as basic data collection instruments by field enumerators in areas that do not receive a mailback questionnaire (for example, in list/enumerate areas). They also will be used by enumerators in mailout/mailback areas and in update/leave areas during personal visits to households that did not mail back their questionnaire. o A short form is being prepared for the Be Counted National campaign for people who did not receive a questionnaire or believe they were not included on a census form. These unaddressed Be Counted questionnaires will be printed in numerous languages and placed at locations where people frequent, such as in libraries, post offices, community centers, Walk-in Questionnaire Assistance Centers, and convenience stores. The responses on these forms will be checked against census records to eliminate duplications. MILESTONES Q3 1999 Begin questionnaire printing
OBJECTIVE
Since the 1960 census, the bulk of decennial census data has been collected
from a sample of housing units. Likewise in Census 2000, the Census Bureau
will deliver the long-form questionnaire to a sample of housing units. The
use of sampling will allow the Census Bureau to meet the objectives of
reducing cost and maintaining the level of respondent burden comparable
to the 1990 census.
MAJOR FEATURES
Decennial census data collected on a variety of socioeconomic and housing
subjects are required by Federal agencies for implementing programs defined in
legislation. In addition, these data are used by, state, local, and tribal
governments, as well as the private sector for planning and developing social
and economic policy and for a myriad of other uses.
To collect these valuable data, Census 2000 will implement a variable rate
sampling scheme. Use of variable sampling rates will allow for more
efficient allocation of the sample and will maintain the accuracy and
reliability of census data at small geographic levels (block groups, census
tracts, and small communities), while reducing respondent burden.
The variable rate sampling scheme for the Census 2000 long form probably will
be similar to the 1990 census scheme and basically will be as follows:
o The overall sampling rate will be about 1-in-6 households or 17 percent.
o The sampling rate in general purpose governmental units with populations
fewer than 2,500 will be 1-in-2 households. School districts will be
considered general purpose governmental units. Also included will be
the American Indian and Alaska Native areas identified for the census
that have American Indian or Alaska Native populations of fewer than
2,500.
o The sampling rate in other governmental units will be 1-in-6 or 1-in-8
households based on our precensus counts of housing units for census
tracts, controlling on the fixed overall sampling rate of 17 percent.
MILESTONES
Q1 1998 Determine final long-form sampling methodology
Q3 1999 Determine long-form sample to implement variable-rate
sampling methodology
OBJECTIVEThe Census Bureau intends to construct a complete listing of living quarters to control the collection and tabulation of Census 2000 data. While we are pursuing a myriad of partnerships to create and update our address list, we realize that partners will not come forward for all areas. Nevertheless, in building the address file, we must ensure that all areas have adequate coverage. MAJOR FEATURES To enumerate and tabulate Census 2000, the Census Bureau must identify all living quarters and locate these living quarters with respect to the geographic entities for which we report data. We accomplish this by creating and maintaining a Master Address File (MAF) that identifies all living quarters and spatially locates those addresses using our geographic database called TIGER. The building and maintenance of the MAF and TIGER involve partnerships with other Federal agencies, with state, local, and tribal governments, regional and metropolitan planning agencies, the private sector, and nongovernmental organizations. The inventory of all living quarters includes addresses for each housing unit and each group quarters. Except where the address list is created at the time of enumeration (such as in list/enumerate areas), each listing must have a complete address that can be used for mailing a census questionnaire and/or a location description that can be used by an enumerator to locate the living quarters. Each listing must be linked to the TIGER data base. We are creating and maintaining the MAF through a series of operations, described in the following: In areas where the U.S. Postal Service (USPS) delivers mail to city style addresses, we: o Create the MAF by combining addresses from the 1990 census Address Control File with those addresses in the USPS Delivery Sequence File (DSF). The DSF is a national file of individual delivery point addresses. As part of a cooperative agreement, the USPS provides the Census Bureau with updated DSFs on a regular basis. o Locate these addresses in the TIGER data base. Where we cannot locate an address, the location is researched and resolved through an office operation or through assistance from local partners. As a result of this research, we identify new features and correct and add address ranges to the TIGER data base. Since we will use mailout/mailback enumeration methodology in areas with mail delivery to city style addresses, we need to determine where mail is not delivered to city style addresses. In addition, there are some areas that are very remote or sparsely populated that we want to send an enumerator to only once. We identify these areas for different enumeration methodologies using information on the types of mail delivery, the types of addresses, and our regional office knowledge of the area. We create the address list in these areas during: o Address listing operations. In areas where the addresses are predominately non-city style, the Census Bureau will create an address list through a door-to-door canvassing operation and capture of the spatial locations. We will enumerate these areas by having enumerators deliver addressed census questionnaires during an update/leave or an update/enumerate operation. Where there is no mailing address for the listing or the mailing address is not a city style address, the listing will include a location description. o List/enumerate operations. Census enumerators will create the address list at the time of enumeration while canvassing their assignment area and picking up or completing unaddressed questionnaires that the USPS previously had delivered to each household. The completed address listings and their spatial locations will be captured at this time. A separate operation will build an inventory of all special places. We will interview an official at each special place using a Facility Questionnaire. The responses to the questionnaire will identify each group quarters and any housing units associated with the special place. We will classify each group quarters and any housing units according to whether they will be enumerated as part of special place enumeration or through regular enumeration. We will add those group quarters and housing units to the MAF and link them to the TIGER data base. In city style address areas, we also will have some targeted field operations to help verify addresses, identify new addresses, and determine the location of addresses that are not in the TIGER data base. We will incorporate local knowledge updates to build the MAF through the Program for Address List Supplementation (PALS) and the Local Update of Census Addresses (LUCA) program. As another quality check, we will be comparing the number of addresses in the MAF with independent housing benchmarks for aggregated areas (state, county, census tract). MILESTONES Q4 1997-Q1 1998 Determine areas for different enumeration methodology Q2 1998 Begin targeted field operations Q2 1998 Begin address listing Q4 1998 Complete MAF/TIGER integration for areas with mail delivery to city style addresses Q3 1999 Deliver first nationwide MAF for Census 2000
OBJECTIVETo allow all state, local, and tribal governments, regional and metropolitan planning agencies, and related nongovernmental organizations to submit locally developed and maintained address lists to the Census Bureau. MAJOR FEATURES The Program for Address List Supplementation (PALS) is a partnership program that allows the Census Bureau to take advantage of local knowledge to enhance its Master Address File (MAF). The participants will benefit by more complete and accurate Census 2000 data for their area. PALS is made possible by the Census Address List Improvement Act of 1994 (P.L. 103-430). The program operates as follows: o The program is limited to areas with city style addresses (house number and street name). o The local participants are requested to submit electronic address lists to the Census Bureau. o The Census Bureau will match the local list both to the MAF and TIGER data base. We will provide disposition lists that show each locally provided address and the results of our matching. o The Census Bureau has identified several frequently occurring problems in the local address lists it has received to date. Among these are the inclusion of: 1) incorrect or inconsistently labeled "within structure" unit identifiers; 2) alternate names, spellings, abbreviations, or formats; and 3) "group quarters unit identifiers" (for instance, dormitory, residential hotel, or institutional room numbers). Census Bureau staff are examining these situations to determine the most effective way of using local address lists to enhance the MAF. MILESTONES 9/1996 Extended invitation to submit lists of city style addresses Through Q4 1999 Provide final PALS response
OBJECTIVETo allow local and tribal governments to designate a liaison to review the portion of the Master Address File (MAF) that covers their jurisdiction. MAJOR FEATURES The Local Update of Census Addresses (LUCA) program is a partnership program that will allow the Census Bureau to take advantage of local knowledge in developing its MAF. The participants will benefit by more complete and accurate Census 2000 data for their area. LUCA is made possible by the Census Address List Improvement Act of 1994 (P.L. 103-430) which, for the first time, authorizes designated representatives of local and tribal governments to review and validate the accuracy of the MAF. The program will operate as follows: o The local or tribal government will designate a liaison to review the portion of the MAF for that jurisdiction. The liaison will be subject to the same restrictions on revealing census information as other census workers. The address lists are confidential under Title 13, and participants must sign an oath promising to protect the confidentially of the lists. o Prior to the census and after the initial targeted field operations or address listing have been completed, the Census Bureau will send the liaison a listing from the MAF (in either paper or electronic form), the accompanying maps for their jurisdiction, and a tally of MAF records for each census block. o The liaison will provide input regarding the completeness of the MAF. Census Bureau staff currently are re-examining the ways in which liaisons can best provide this information. o P.L. 103-430 allows the local participants to appeal final Census Bureau decisions. MILESTONES Q4 1997 Extend initial invitation to participate Q1 1998 Conduct first round local workshops Q3 1998-Q2 1999 Provide LUCA materials Q4 1999 Conduct second round local workshops 1/31/2000 Complete LUCA appeals
OBJECTIVE
To provide the necessary information to associate each living quarters in
Census 2000 to a spatial location, each location to a specific geographic
area, and each geographic area to a name or number, and geographic codes
and attributes.
MAJOR FEATURES
The geographic data base for the census--TIGER®* (Topologically
Integrated Geographic Encoding and Referencing)--provides the geographic
structure for the control of the data collection, tabulation, and
dissemination operations.
The geographic data base constantly changes. Not only are new streets built,
but some streets cease to exist, and the path of some existing streets moves.
The names and address ranges associated with these streets change, too. Not
only is the inventory of geographic entities different from year to year, but
also the boundaries, names, and related attributes for the entities may
change.
To ensure that the information in the TIGER data base is complete and correct,
the Census Bureau works in partnership with other Federal agencies, state,
local, and tribal governments, and others in the public and private sectors.
We maintain the TIGER data base through internal programs and partnership
activities.
The Census Bureau obtains updates to the feature network, including
associated address ranges, through the following operations:
o Census MAF building activities. These include office resolution of
addresses that cannot be automatically spatially located, the various
targeted field operations, and the address listing operations.
o Partnership MAF building activities. These include PALS, the LUCA program,
and local participation in the resolution of addresses that need to be
spatially located.
o Digital files. Local areas with suitable digital files provide them to
the Census Bureau in lieu of a manual updating process. We transfer
the update information from the local file to the TIGER data base by
an automated process.
-----------------------------------
*TIGER® is a registered trademark of the U.S. Bureau of the
Census. For ease of presentation, the trademark symbols for TIGER and
TIGER-related products are omitted from the text.
o Local governments. In response to a preview of the census map of their
jurisdiction, local governments may furnish the Census Bureau with updated
street features and their names.
We obtain updates to the boundaries, names, and attributes of the various
geographic entities for which we tabulate data through various methods.
These include:
o A voluntary survey designed to collect an accurate inventory of all local
general-purpose governmental units and to obtain up-to-date information on
boundary and annexation changes.
o A program that provides the highest elected official of each general-
purpose governmental unit with a copy of the census map that shows the
jurisdiction boundaries we plan to use to tabulate Census 2000. The local
official will review and validate these boundaries. This program replaces
the local review of boundaries portion of the 1990 Postcensus Local Review
Program.
o A program that provides local participants the opportunity to delineate
Census 2000 participant statistical areas (block groups, census county
divisions, census designated places, and census tracts).
o Additional programs that offer participants the opportunity to identify
other areas for which the Census Bureau will tabulate data (for example,
traffic analysis zones).
The Census 2000 boundaries for general-purpose governments and other legal
entities are those legally in effect on January 1, 2000. Where the boundaries
for any geographic entity intersect a feature, we assign any addresses
associated with that feature to the geographic entity by an automated process
using information in the TIGER data base and information provided by local
governments. The LUCA and boundary collection and validation programs also
provide participants with the opportunity to review the assignment of addresses
in their jurisdiction.
The information from all these programs and operations is inserted into the
TIGER data base where it is processed and undergoes various checks for
consistency and accuracy. As required by specific census operations, the
Census Bureau creates extracts from the TIGER data base to produce the
necessary map products and geographic files.
MILESTONES
Q1 1998 Collect January 1, 1998 boundaries
Q4 1998 Establish block structure for use in LUCA program
Q1 1999 Collect January 1, 1999 boundaries
Q1 2000 Conduct final boundary validation
Q4 2000 Establish tabulation geographic structure
OBJECTIVETo develop and set up an effective and cost-efficient infrastructure that facilitates the complete and accurate enumeration in Census 2000 and the capture and processing of census information in time to meet Constitutional and legislative requirements. MAJOR FEATURES Since the decennial census requires a massive mobilization of human and physical resources, the Census Bureau will develop an extensive temporary office infrastructure across the country to conduct Census 2000. Physical Infrastructure The function of each major infrastructure component is as follows: Regional Census Centers (RCCs) - 12 stateside Centers (and an Area Office in Puerto Rico) o Manage all census field data collection operations, address listing, and address list enhancement for city style address areas through a network of Census Field Offices (CFOs) and Local Census Offices (LCOs) o Coordinate Local Update of Census Addresses (LUCA) activities o Produce maps o Conduct geographic activities such as geocoding, TIGER data base updates, and working with local participants on the participant statistical programs and the P.L. 94-171 Redistricting Data Program o Conduct recruiting o Manage payroll and personnel administrative system Census Field Offices (CFOs) - (425-450 Offices) o Perform address listing or address list enhancement for city style address areas o Conduct local recruiting o Perform clerical review of completed field work Local Census Offices (LCOs) - 475 Offices (466 stateside, 9 in Puerto Rico) o Produce enumerator maps and assignments o Conduct local recruiting o Conduct outreach and promotion (for example, the Be Counted campaign) o Conduct group quarters/special place/service-based enumeration operations o Coordinate military enumeration (if applicable) o Conduct update/leave operation o Conduct list/enumerate operation o Conduct followup enumeration (nonresponse followup, coverage edit, large household followup) o Manage field staff payroll and personnel administrative system Data Capture Center (DCC) in Jeffersonville, Indiana o Process address listing o Check in mail returns o Edit questionnaires o Conduct unduplication of multiple questionnaires o Code questionnaires o Process quality check survey o Manage payroll and personnel administrative system Other Data Capture Centers (DCCs) - 3 Centers o Check in mail returns o Edit questionnaires o Conduct unduplication of multiple questionnaires o Conduct data capture o Manage payroll and personnel administrative system Establishing the above offices will involve the following activities: o Lease office space o Obtain furniture, equipment, and supplies o Procure and install computer hardware and software (except for CFOs) o Establish voice and data line connections Commercial Phone Centers (CPCs) - 2 Centers o Conduct telephone questionnaire assistance o Handle requests to respond to census by telephone Administrative Infrastructure In addition to physical infrastructure, there are administrative structure requirements that must be in place to recruit, manage, and pay the census workers who are hired. These functions consist of the following: o Develop system of competitive pay rates o Develop position descriptions o Validate selection aid test o Implement Decennial Automated Name Check (DANC) system for criminal history screening of potential census workers o Implement Preappointment Management System (PAMS) to handle hiring of temporary census workers o Develop bonus/incentive pay system that supports producing a quality product, staff retention, and high productivity o Implement Automated Decennial Administrative Management System (ADAMS) to manage payroll system for temporary census workers MILESTONES Q4 1997-Q1 1998 Open RCCs Q1 1998 Open urban CFOs Q2-Q3 1998 Open rural CFOs Q3 1999 Open LCOs Q4 1998-Q3 1999 Open DCCs Q1-Q2 2000 Open CPCs
OBJECTIVETo recruit and train a sufficient number of temporary census workers to complete Census 2000 operations on schedule. MAJOR FEATURES The decennial census is the largest peacetime activity undertaken by the Federal Government. The Census Bureau expects to hire about 500,000 temporary census workers in the field to conduct Census 2000. Attaining this goal will require the recruiting and testing of millions of persons. Positions include a wide range of activities such as local census office managers, enumerators, partnership specialists, media specialists, and clerks. This effort will require a very large recruiting effort throughout the country. Every job applicant will have to take a written test and meet certain other requirements before being hired as a census worker. The Census Bureau will use the Decennial Applicant Name Check (DANC) system to screen all applicants for criminal histories. Qualified applicants who are selected will be required to take the oath of office and sign an affidavit of nondisclosure in which they agree they will "not disclose any information contained in the schedules [questionnaires], lists, or statements obtained for or prepared by the Bureau of the Census, to any person or persons either during or after employment." In recognition of the changing composition of the labor force and the increasing difficulty in hiring a sufficiently large number of temporary census workers, especially enumerators, to conduct the census, the Census Bureau is studying different and innovative methods of setting pay and incentives for persons to work on Census 2000. In addition, the Census Bureau is attempting to expand the labor pool from which it can recruit by negotiating with other Federal and state agencies that manage retirement and income transfer programs (Federal civilian and military retirement, Aid to Families with Dependent Children, Public and Indian Housing program, and so on) to reduce any barriers and encourage recipients of the various programs to work for the Census Bureau. MILESTONES Q1 1997 Begin DANC system Q1 1998 Begin recruiting for census field offices
OBJECTIVETo determine where to implement a variety of special enumeration methods in Census 2000 and to assist in a wide range of planning activities. MAJOR FEATURES The main function of the Census 2000 Planning Database is to serve as a single repository of data from a variety of sources that users can easily access to perform the desired activity. Specifically, this database will be used to: o Share information with local governments as part of the partnership agreement o Aid in planning recruitment of temporary census workers o Assist with marketing and outreach program o Identify hard-to-enumerate areas o Identify areas where special targeted enumeration methods should be used, such as team enumeration; Walk-in Questionnaire Assistance Centers; mailout of Spanish-language and other-language questionnaires; and targeted distribution of Be Counted questionnaires in English, Spanish, and numerous other languages o Analyze mail response rate patterns and trends o Conduct research to improve the stratification used in sampling for nonresponse followup and the quality check survey o Provide access to historical census data for benchmarking purposes o Identify areas for coverage edit and large household followup operations The majority of the database will consist of geocoded demographic and socioeconomic data. These data will be used to develop tabulations at various geographic levels. Once the tabulations have been calculated and stored in the system, a set of algorithms or formulas will be run against the tabulations, and these results will be stored in the system. The algorithms are designed to determine which methods will be applied to specific programs and geographies. Each Regional Census Center (RCC) will have access to the information on the database for its area. The RCCs also will be able to review the areas identified for special methods by the algorithms. The RCCs may solicit input from local governments as part of the partnership program. Users will have the ability to conduct "what-if" exercises through the system menus. They may alter algorithm conditions and/or thresholds within the algorithms to see the impact on the data. Users also can download extracts from the database that can be printed or used for mapping variables and algorithm results. MILESTONES 4/1996-9/1996 Defined functional requirements of planning database Q1 1998 Complete system design Q2 1998 Complete system programming and testing Q2 1998 Begin implementing functions of the system
OBJECTIVETo obtain a questionnaire from every housing unit in Census 2000. MAJOR FEATURES To ensure that we obtain a completed questionnaire from every housing unit, the Census Bureau must first make sure that a questionnaire is delivered to every housing unit. We will accomplish this by using one of the following three basic data collection methods: o Mailout/mailback. The Census Bureau will use U.S. Postal Service (USPS) letter carriers to deliver questionnaires to the vast majority of housing units that have city style addresses (house number and street name). o Update/leave. In areas where there is no USPS delivery to city style addresses, enumerators will hand-deliver addressed census questionnaires to each housing unit for the householder to complete and mail back. They also will update and make any necessary corrections and/or additions to the Master Address File as they deliver the questionnaires. o List/enumerate. In very remote or sparsely-populated areas, enumerators will visit each housing unit and pick up or complete unaddressed short-form questionnaires that the USPS previously delivered to each unit. The enumerators will ask additional long-form questions of a sample of units. They also will develop an address list for the area and spot the housing unit's location on a map. Additional data collection strategies noted in the following sections cover enumeration of special population groups (for example, persons in group quarters and on military bases), people with no usual residence, and nonresponse followup of housing units that did not return a questionnaire. MILESTONES 3/13-3/15 2000 Deliver initial mailout/mailback questionnaires 3/3-3/30 2000 Conduct update/leave operation 3/31-5/01 2000 Conduct list/enumerate operation
OBJECTIVETo provide respondents with convenient access to obtain assistance in the completion of their Census 2000 questionnaires or to respond directly to the census. MAJOR FEATURES o The Census Bureau will implement an extensive Telephone Questionnaire Assistance (TQA) operation to answer general questions about the census so that the respondent can complete the census questionnaire and mail it back. This service will have a well-publicized national, toll-free number and will use an automated touch-tone system that can handle a large number of calls concurrently. Assistance will be available in English, Spanish, and other languages. There also will be a toll-free telephone device (TDD) for the hearing impaired. o In addition, TQA will offer a means for callers to respond directly to the census. Operators will record answers to the census over the telephone, if this service is requested during telephone assistance. Assistance will be available in English, Spanish, and other languages. o If the Census Bureau can resolve confidentiality concerns and certain technical issues, an Internet address will give the opportunity of responding electronically. MILESTONES Q2 1999 Award contract(s) for TQA Q2-Q3 2000 Conduct TQA
OBJECTIVEThe Be Counted National Campaign will provide a means for people to be included in Census 2000 who may not have received a census questionnaire or believe they were not included on one. The Census Bureau will place particular emphasis on developing ways to include population groups that historically have been undercounted. MAJOR FEATURES Be Counted Questionnaire The Be Counted questionnaire is being designed to be respondent friendly and easy to understand and complete by anyone who picks it up. This would include people with a usual residence who did not receive a questionnaire at their address, people who believe the returned questionnaire for their address excluded them, people who require questionnaires in different languages, migrants or seasonal farmworkers, those who have no usual residence, and so forth. These forms will contain short-form questions along with several additional items needed to process and match the forms to the census results. Questionnaires Available in Many Locations and in Numerous Languages Be Counted questionnaires will be readily accessible at numerous public locations, such as Walk-in Questionnaire Assistance Centers, post offices, libraries, community centers, departments of motor vehicles, convenience stores, and other places where people frequent. The promotion campaign and distribution of the Be Counted forms will begin just after the delivery of the census questionnaires and will end just before the nonresponse followup operation begins. Be Counted forms will be printed in numerous languages in addition to English and Spanish. We are currently researching the languages we plan to use for these forms. The findings from this research will be discussed with our professional advisory committees, our partners in the marketing and partnership program, and other stakeholders. MILESTONES Q2-Q3 1998 Submit printing specs for Be Counted questionnaires for bid Q3 1999 Begin printing Be Counted questionnaires 3/15-4/12/2000 Conduct Be Counted Campaign
OBJECTIVEIn Census 2000, the Census Bureau will implement a comprehensive set of procedures to enumerate people who do not live in traditional housing units. These include people who live in group quarters situations (for example, nursing homes, group homes, and colleges), people without housing, people who live at migrant and seasonal farmworker camps, and people living on military installations and ships. Special procedures also will be applied to those who live in unique areas of the country like remote Alaska and American Indian reservations. MAJOR FEATURES Special Place Facility Questionnaire Operation To enumerate people at these special locations, we must identify the places. Identification of these places will occur on a flow basis, by means of a procedure called the Special Place Facility Questionnaire Operation. This operation will update existing information for our inventory of special places and group quarters, identify additional group quarters, identify contact persons at each location, assign a group quarters type code, determine availability of administrative records, identify any housing units at and/or associated with the special place or group quarters, and collect other administrative information. Most information will be collected during an automated Computer-Assisted Telephone Interview (CATI), with some cases being completed by personal visit using a paper census questionnaire. Group Quarters Enumeration Staff in our local census offices will conduct the group quarters enumeration. Starting in January 2000, census workers will make advance visits to the group quarters to meet with facility staff and discuss the upcoming census enumeration. These visits are very beneficial because they promote and encourage participation in the census, and identify any difficulties that might be encountered during the enumeration. In April 2000, enumerators will enumerate people in each group quarters by listing all the residents, distributing questionnaire packets. When needed, enumerators will provide assistance in completing the questionnaires. Enumeration results will be checked and verified using a quality control checklist to ensure enumeration at the facility was complete. A small number of facilities, such as jails and prisons, will self-enumerate their facility. These facilities will use regular census procedures to conduct the enumeration, and the facility staff become special sworn census employees to protect the confidentiality of the census information. Transient Night (T-Night) Operation Transient night, commonly referred to as T-Night, is an operation designed to count persons of a highly transient nature. T-Night will take place on the day before Census Day (on Friday, March 31). T-Night enumerators will visit and interview people occupying campgrounds at racetracks, recreational vehicle (RV) campgrounds or RV parks, commercial or public campgrounds, fairs and carnivals, and marinas. Every person enumerated during T-Night will have the opportunity to report a usual residence. On T-Night, enumerators will visit each assigned T-Night place, meet with a contact person at the site to explain the purpose of the visit, offer the Privacy Act notice, answer any questions, and verify information about the site. Then the enumerator will interview each person at the assigned location. In the 1990 census, there were approximately 13,000 T-Night sites nationwide. Remote Alaska Enumeration Several methods will be used to enumerate the varied types of areas in the State of Alaska. The two largest cities, Anchorage and Fairbanks (and their vicinities), will be designated as mailout/mailback areas due to their concentrated populations and existence of city style addresses. The rest of Alaska will be enumerated by the list/enumerate method, which also is being used in the sparsely populated areas of the lower 48 states. The unusual feature of Census 2000 enumeration in Alaska will be the treatment of outlying or remote areas. Most of these settlements, located throughout the state, are accessible only by small-engine airplane, snowmobile, four-wheel-drive vehicle, dogsled, or a combination of these. Roads rarely exist to link the widely scattered settlements. These settlements range in population from a few people to several hundred persons, with a few larger places of 2,000 people or more. The timing of the mailout/mailback enumeration will be the same as in other states. However, enumeration of the remote areas will begin earlier, in mid-February, but all census questions will be asked in relation to Census Day (April 1). The special timing will permit travel to these areas during the period when conditions will be most favorable. For example, the ground and rivers still will be frozen so that planes can fly in and out, and the residents will still be at home. Once the spring thaw (or "breakup" as it is known locally) begins, travel to some of these areas will be difficult or impossible, and the people will leave home to fish and hunt. Enumerators will have to finish their work before then, or they will miss a large part of the population. Military/Maritime Enumeration People living on military installations and on maritime vessels will be enumerated during Census 2000. To enumerate people residing on military installations, the Census Bureau will work with the Department of Defense and U.S. Coast Guard to identify housing units and other living quarters on the installations. Different enumeration methodologies, such as mailing census questionnaires to housing units on installations and enumerating people at their work station, will be used. The Census Bureau will work with the U.S. Maritime Administration and others to identify maritime vessels in operation at the time of the census and mail enumeration materials to those vessels for completion. MILESTONES Q3 1998-Q2 2000 Conduct Special Place Facility Questionnaire Operation Q1 2000 Conduct advance visits to special places/group quarters Q1 2000 Conduct local knowledge update of group quarters locations Q1 2000 Conduct enumeration in remote Alaska 3/31/2000 Conduct T-night operation 4/3-5/6/2000 Enumerate people at group quarters and conduct coverage improvement validation
OBJECTIVETo enumerate people with no usual residence in Census 2000. This operation, called Service-Based Enumeration (SBE), is designed to improve the count of an area by including people that use services who might not be included through other enumeration methods. MAJOR FEATURES The following SBE procedures are designed to include people who might otherwise be missed in the census by enumerating them at selected service locations, such as shelters and soup kitchens, and at targeted non-sheltered outdoor locations. SBE will not provide a count of homeless persons or of service users. o The Census Bureau will work with local governments and community-based organizations to identify the list of service locations open at census time. o Using simplified enumeration procedures and forms, the Census Bureau will conduct a one-time enumeration at shelters, soup kitchens, and regularly scheduled mobile food vans that provide services primarily for people without housing. o The Census Bureau will use statistical techniques to improve the enumeration of people without housing. o The Census Bureau plans to have enumerators visit targeted non-sheltered outdoor locations where people without housing congregate. These sites will be identified by local officials, advocacy groups, and community-based organizations. o The Census Bureau will work with local officials and community-based organizations to identify camps and other locations where migrant and seasonal farmworkers can be found at the time of the census. In addition to being enumerated at the designated service locations, people with no usual residence will be able to pick up Be Counted questionnaires at other selected service locations, such as clothing distribution centers, drop-in centers, and health care clinics serving people without housing. We also will publicize a national toll-free telephone number and encourage people to respond that way if they do not have access to a census questionnaire. MILESTONES Q2-Q4 1999 Develop list of service locations 4/3 and 4/4/2000 Conduct enumeration of people at service locations 4/10 and 4/11/2000 Conduct followup visit at sample of shelters and soup kitchens if needed for estimation methodology Q1-Q2 2000 Distribute Be Counted questionnaires at service locations
OBJECTIVETo overcome barriers to successful enumeration in Census 2000 by implementing special data collection methods in targeted areas. DESCRIPTION Many targeted methods will be used in Census 2000: o The Census 2000 Planning Database will help to identify the most appropriate areas in which to use targeted methods designed to overcome difficult enumeration barriers. o Information will be shared with officials of local and tribal governments. They can then help identify, in advance, areas that likely will require additional "get out the count" efforts. o A team or crew of enumerators will go into a targeted area and conduct the enumeration in a short period of time. Team enumeration will be used in areas where field conditions may interfere with the timely completion of the enumeration. These conditions may be high concentrations of multiunit buildings, enumerator safety concerns, low enumerator production rates, and so forth. o A Be Counted National campaign will make unaddressed Be Counted questionnaires available in sites such as post offices, libraries, community centers, stores, Walk-in Questionnaire Assistance Centers, and other publicly accessible locations. o Mail response rates and maps will be provided to local and tribal officials periodically during the census enumeration. They will work in partnership with census staff to identify unexpectedly low response areas. Targeted enumeration efforts and additional outreach and publicity activities then will be implemented. o In partnership with local and tribal governments and community-based organizations, the local census offices will establish Walk-in Questionnaire Assistance Centers in their communities (for example, in non-English speaking areas) to assist respondents in completing their questionnaires. These centers will be established in community centers, large apartment buildings, and so forth. o Assistance in various foreign languages will be provided for people who are not able to respond to questionnaires in English or Spanish. MILESTONES Q2 1998 Begin implementing planning database Q1 2000 Open Walk-in Questionnaire Assistance Centers 3/15-4/12/2000 Conduct Be Counted Campaign
OBJECTIVETo verify the accuracy of vacant housing unit information provided to the Census Bureau by the U.S. Postal Service (USPS). MAJOR FEATURES o A 10-percent sample of housing units designated as vacant by the USPS will be selected for followup by census enumerators to determine if those units were truly vacant on Census Day. This operation will ensure the integrity of the vacancy information provided by the USPS. o The operation also will gather information about the characteristics of those vacant housing units. Additional vacant units will be encountered by enumerators during nonresponse followup operations among vacant housing units not designated as such by the USPS. o As a final check, the consistency of the vacancy rates with independent estimates and historical data will be assessed. MILESTONES Q2 2000 Conduct vacant housing unit followup
OBJECTIVETo obtain Census 2000 data for housing units with more than five persons. MAJOR FEATURES Both short- and long-form census questionnaires will allow for up to five persons to provide census information about themselves. If the person filling out the questionnaire indicates in the roster that there are six or more persons in the housing unit, the Census Bureau will conduct a followup interview to obtain information for the additional residents. The followup will be conducted by Computer-Assisted Telephone Interviewing (CATI) and personal visit for households that cannot be contacted by telephone. MILESTONES Q2 2000 Conduct large household followup
OBJECTIVETo eliminate duplicate addresses and questionnaires in Census 2000. MAJOR FEATURES One of the main goals of Census 2000 is to make it simpler for people to be counted. In addition to our standard enumeration methodology for an area (for example, mailout/mailback data collection), census forms will be made readily available in public places and provided in multiple languages. Responses to the census also will be accepted over the telephone and possibly on the Internet. Providing these response options will make it easier for persons to be counted but may increase the possibility that multiple responses will be submitted for a given person and household. A complete, accurate address list, high speed data capture capabilities, along with automated matching technologies, will be the keys to avoiding the duplication of people and residences. The control of the enumeration of Census 2000 will be based on an address list called the Master Address File (MAF). Every housing unit in the census will have a unique identifier. Every response to the census will be data captured and then linked to the MAF using powerful matching computer programs to assign the identifier. Once an address is linked to the MAF, we will be able to determine when multiple returns for a housing unit have been submitted. Matching tools again will be applied to identify and correct instances of duplicate counting of individuals. Unduplication of multiple responses in past censuses would have required a massive clerical operation since only a small subset of person names was data captured. An automated matching capability was not feasible without the names of each of the persons on census forms. It would have been necessary to clerically compare the information on the individual forms and then feed the results into the computer. Since the 1990 census, we have embarked on a path that will ensure timely, complete capture of all census responses. This includes the possible electronic capture of responses provided on the Internet and the telephone call-ins of census responses, in addition to the commitment to use imaging technology with electronic optical mark and intelligent character recognition for the data capture of information from the paper forms. High speed capture will allow the Census Bureau to accelerate the process of capturing the names and demographic characteristics of all persons on the paper forms. The advances in computer technology in the areas of computer storage, retrieval, and matching, along with image capture and recognition, have now given the Census Bureau the flexibility to provide multiple response options without incurring undue risk to the accuracy of the resulting census data. As a further safeguard of the quality of the census enumeration, we will develop statistical procedures to identify areas from which we receive unusually large numbers of unaddressed Be Counted questionnaires and will verify the validity of the census responses. MILESTONES Q2-Q3 2000 Conduct unduplication of multiple responses
OBJECTIVETo improve the coverage of persons in housing units in Census 2000. MAJOR FEATURES In Census 2000, a coverage edit will be performed to review Census 2000 questionnaires for potential missing people. After the Census Bureau receives the questionnaires, they will be checked to see if there is any discrepancy between the number of persons reported as members of that household and the number of persons for whom census information was provided on the form. For example, the respondent may have indicated that five persons lived in the housing unit, but there was information for only two persons. For these discrepancies, telephone clerks will call the household to resolve the problem. There will be no personal visit followup for households with reporting discrepancies that we cannot reach by telephone. The coverage edit should not be confused with the quality check survey (Integrated Coverage Measurement), the edit and followup of large households, or the content edit (for missing or incomplete responses to population or housing items). (The content edit in Census 2000 will be completed solely by computer with no field followup.) MILESTONES Q2 2000 Identify housing units requiring coverage edit followup Q2 2000 Conduct coverage edit followup operation
OBJECTIVEThe Census Bureau will make every effort to secure a response in Census 2000 from every resident and every housing unit. Following that, and to ensure an acceptable response among all census tracts, the Census Bureau will use statistical sampling to account for a portion of the population in each census tract. MAJOR FEATURES o During the initial response period (the period immediately before and immediately after Census Day), the Census Bureau will issue reminder publicity urging people to return their questionnaires, fill out Be Counted questionnaires, or use the telephone and possibly the Internet to provide their census responses. o On a daily basis, the Census Bureau will determine the response rate for every census tract, which is a neighborhood of about 4,000 people. The response rate is defined as: Mail + Telephone + Other Responses *100% Questionnaires Mailed or Delivered o For any census tract in which this rate is less than 100 percent after the initial response period, enumerators will perform nonresponse followup (NRFU) (that is, contact the respondent and complete a census questionnaire) under the direct sampling strategy. o In the direct sampling strategy, the Census Bureau will select a sample of nonresponding addresses in each census tract at the end of the initial response period. The sample will vary from census tract to census tract based upon the particular response level and will be designed to achieve at least a 90-percent total response rate in each census tract. o Enumerators will perform NRFU for each of the selected sample addresses. The addresses will be visited by an enumerator who will complete a questionnaire by personal interview. o The Census Bureau will not use sampling to complete the enumeration on American Indian reservations, or in Alaska Native Village statistical areas, the U.S. Virgin Islands, or the Pacific Island Areas. MILESTONES Q1 1997 Determined NRFU plan for Census 2000 Q2 2000 Conduct NRFU
OBJECTIVE
To produce a "one-number" census estimate of the U.S. population in Census
2000 that will improve accuracy, reduce costs, and eliminate confusion and
controversy caused by having more than one number measuring the same
population.
MAJOR FEATURES
In Census 2000, the Census Bureau will conduct a "one-number" census of
population and housing. That is, there will be just one set of official
census results produced by the legal deadlines. In recent past censuses,
the population counts were represented by two sets of numbers: the number of
people actually counted and the number of people estimated to be living in
the United States, after compensating for enumeration errors.
The quality check of Census 2000 results, known as the Integrated Coverage
Measurement (ICM) survey, will be composed of three phases:
o Housing Unit Phase
o Quality Check Person Interview Phase
o Person Matching Phase
Housing Unit Phase
During this phase, housing units within the sample blocks will be listed
independently of the census and later matched to the census inventory of
housing units. After reconciling the differences, a list of housing units
that are confirmed to have existed within the sample blocks on Census Day
will be prepared for conducting quality check person interviews.
Quality Check Person Interview Phase
In this phase, the interviewer will collect information about the current
residents and anyone who has moved out of the sample block between Census Day
and the time of the interview. The interviewer will ask questions about
alternate residences to establish where people lived on Census Day according
to census residence rules.
Person Matching Phase
In this phase, the people counted in the quality check survey will be
compared with those enumerated in the census. Using estimation procedures,
the Census Bureau will produce estimates of people missed or duplicated in
the census enumeration. These estimates then will be used to update the
final census data files to produce the one-number census results.
MILESTONES
Q3 1999 Select quality check sample
Q3 1999-Q1 2000 Conduct housing unit listing phase
Q1 2000 Conduct housing unit matching and followup phase
Q2-Q3 2000 Conduct quality check person interviewing phase
Q2 2000 Conduct person matching and followup phase
Q3 2000 Missing data and estimation
Q4 2000 Create transparent data files
Q4 2000-Q1 2001 Create P.L. 94-171 redistricting data products
CENSUS QUALITY CHECK SURVEY
------------------------------ ---------------------------------
| Produce list of addresses in | | List addresses in quality check |
| all blocks for the CENSUS | | sample blocks |
------------------------------ ---------------------------------
| |
| |
------------------------------ ---------------------------------
| Enumerate U.S. residents in | | Conduct quality check person |
| the Census | | interviews for sample addresses |
------------------------------ ---------------------------------
| |
| |
| -------------------------------- |
| | Compare results-determine how | |
--| well we counted individuals in |--
| the quality check sample blocks|
--------------------------------
|
|
|
--------------------------------
| Use quality check survey |
| results to produce census |
| results that account for |
| people missed or counted more |
| than once during the census |
| enumeration to produce a |
| ONE-NUMBER CENSUS |
--------------------------------
The Census 2000 plan calls for the innovative use of statistical sampling for
conducting two major census operations. These operations are sampling for
nonresponse followup (NRFU) and for a quality check called Integrated Coverage
Measurement (ICM). These two operations have different goals, but they
complement each other.
Sampling for nonresponse will be used to complete the census enumeration.
This is a major departure from the process used in previous censuses. Rather
than visiting all households that do not provide a response during the
initial response period, census enumerators will visit and conduct interviews
for a representative sample of those households. Information collected
during this operation will be used to estimate the characteristics of the
households not included in the sample.
After completion of census NRFU operations, we will conduct a follow-up
survey of a representative sample of housing units across the nation. This
operation is referred to as the census quality check or ICM survey. The
ICM sample is selected from a list of addresses created independently from
the Master Address File. This survey is designed to identify people missed
in enumerated housing units and in missed housing units, as well as
identifying people that were counted in the wrong place or more than once
during the initial census enumeration.
The final population estimates are the result of combining information
collected from responses to the census, including mail returns and other
opportunities to respond (such as on the Be Counted questionnaires, by
phone, or possibly the Internet) with results from the census NRFU and
quality check operations.
Because the nonresponse and quality check operations are sample-based, they
contribute error to the census population estimates. This error is referred
to as sampling error. Conversely, the quality check operations should
result in a reduction of the largest source of error in previous censuses,
the coverage error. The coverage error occurs differentially across
geographic areas and among different population groups. This error is more
commonly referred to as the undercount.
Table IX-1 provides a comparison of the 1990 census undercount rates with results
from an empirical simulation designed to project the effect of the use of
nonresponse and quality check sampling methods on the quality of total population
estimates and estimates for major racial and ethnic groups for the United States.
The two statistics provided for each estimate are as follows:
o Coefficients of variation (CV) for Census 2000 estimates, which account
for the error contribution from the nonresponse and quality check
sampling operations. A coefficient of variation expresses the error
(sampling) as a percent of the population estimate.
o Undercount rates for the 1990 census, as measured by the 1990 Post-
Enumeration Survey. An undercount rate expresses the error (nonsampling)
as a percent of the population estimate.
For example, Table IX-1 shows that the 1990 census missed 5.0 percent of the
Hispanic population in the United States. The implementation of the Census
2000 plan will introduce a sampling error of 0.8 percent for the Hispanic
population but will remove the coverage error for a net improvement of
about 4.2 percent.
------------------------------------------------------
| | Census 2000 | 1990 Census |
------------------------------------------------------
| Race | Projected | Undercount |
| | Sampling error | Rate |
| | (coefficient of | |
| | variation) | |
------------------------------------------------------
| United States, | 0.1% | 1.6% |
| total | | |
------------------------------------------------------
| White, non- | 0.1% | 0.7% |
| Hispanic | | |
------------------------------------------------------
| Black | 0.6% | 4.4% |
------------------------------------------------------
| Asian and | 1.4% | 2.3% |
| Pacific Islander| | |
------------------------------------------------------
| American | 1.4% | 4.5% |
| Indian | | |
------------------------------------------------------
| Hispanic origin | 0.8% | 5.0% |
| (may be of any | | |
| race) | | |
------------------------------------------------------
OBJECTIVE
To use independent estimates to validate the quality check estimates and the
"one-number" census results in Census 2000.
MAJOR FEATURES
Demographic analysis (DA) represents a macro-level approach to measuring
coverage. The demographic approach differs fundamentally from the quality
check estimates, which represent a micro-level approach (case-by-case
matching).
Demographic estimates of net undercount are derived by comparing census
results to estimates based largely on aggregate administrative data. The
national estimates for the population below age 65 are derived by the basic
demographic accounting equation:
Population = Births - Deaths + Immigrants - Emigrants
Aggregate medicare data are used to estimate the population 65 years and
over. To produce estimates below the national level, the equation is
modified to allow for domestic migration. Since administrative records
are utilized, the DA estimates are derived independently of the census
being evaluated.
Use of Demographic Analysis in Census 2000 - Demographic analysis
will provide checks at two distinct points in the Census 2000 process.
First, DA will be compared with preliminary census results before the
quality check operation to provide an early assessment of coverage
differentials at the national and state levels. Second, the DA
estimates will be used to validate the quality check estimates and
ensure the demographic consistency of the final one-number census
results.
Below the state level, we will use independent population and housing
estimates, aggregate administrative records, and other analytic tools as
benchmarks to assess the quality of the evolving census results. In addition
to broad coverage checks of population and housing, this review can identify
content problems and possible anomalies due to geocoding and other
nonsampling errors.
MILESTONES
Q4 1998 Implement features of DA in Dress Rehearsal
Q4 1999 Evaluate quality/completeness of MAF
Q2-Q3 2000 Evaluate quality/completeness of "pre-quality check"
census results
Q4 2000 Validate quality check estimates and one-number census
results
OBJECTIVETo conduct the best possible enumeration in Census 2000 of American Indian and Alaska Native Areas (AIANA) and Hawaiian Homelands. MAJOR FEATURES The Census Bureau will base its strategy for enumerating the populations in AIANA and in Hawaiian Homelands on building partnerships for: o Address list development o Geographic programs o Outreach and promotion o Field operations o Data collection methodologies o Data processing and dissemination Address List Development In areas where the U.S. Postal Service (USPS) delivers mail to city style addresses, we will use the USPS Delivery Sequence File to build the address list and use the mailout/mailback methodology for enumeration. In other areas, we will conduct an address listing operation prior to the census and use the update/leave with respondent mailback of the questionnaire methodology. In more remote areas, we will use the list/enumerate methodology. Tribal governments will have an opportunity to review the address list for their jurisdiction as part of the LUCA program. Geographic Programs Table X-1 shows the wide diversity of programs the Census Bureau will offer American Indian tribes, Alaska Native areas, and Hawaiian Homelands to review and define geographic areas. Outreach and Promotion Census Bureau staff and tribal liaisons will compile a listing of all electronic and print media within the marketing area for paid promotion dissemination. In addition, the Census Bureau will seek the help of tribal liaisons and Complete Count Committees (if the tribes form them) to assist with outreach (census awareness and education) and the promotional campaign, using national or local materials. Field Operations The Census Bureau plans to obtain assistance with the following activities: o Provide local office space for testing and training o Participate in training local census office staff (such as in cultural awareness) o Assist in recruiting strategies for filling census jobs and identifying local referrals for assistance o Assist in identification of vacant housing units o Assist and advise census field staff about potential problem situations o Attend and participate in periodic census/tribal staff meetings o Identify sites for Questionnaire Assistance Centers o Identify locations for distributing Be Counted questionnaires o Use tribal administrative records for a portion of the quality check survey Data Collection Methodologies The Census Bureau will work with tribal officials to select the appropriate data collection methodology or combination of methodologies for each area. These methodologies will range from the use of mailing lists with the respondent either receiving a questionnaire in the mail (mailout/mailback) or an enumerator leaving the form for the respondent to return by mail (update/leave) to the enumerator listing the housing unit and conducting the census interview in one visit (list/enumerate). Data Processing and Dissemination Census 2000 data collected in the AIANA and the Hawaiian Homelands will be processed and disseminated in the same way as information collected for the rest of the Nation. MILESTONES - See Table X-1.
Table X-1. Census 2000 Geographic Programs for
|
|||
| Geographic Program | Type of Area | Purpose of Program | Milestones |
|---|---|---|---|
| Tribal Review Program | Federally recognized tribes with a land base | Provide boundary and feature updates | Q1 1997- Q1 2000 |
| Block Definition Project | Federally recognized tribes with a land base and Federally recognized tribes in Oklahoma without a land base | Identify Census 2000 block boundaries | Q1-Q4 1997 |
| Participant Statistical Areas | Federally recognized tribes with a land base and Federally recognized tribes in Oklahoma without a land base | Define statistical areas such as census tracts, block groups, census designated places, and census county divisions | Q2 1997- Q1 2000 |
| Tribal Subdivision Program (Proposed) | Federally recognized tribes with a land base | Designate special subdivisions (NEW) | Q1-Q3 1998 |
| Tribal Jurisdiction Statistical Area Program | Federally recognized tribes in Oklahoma without a land base | Delineate an identifiable land area as a tribal jurisdiction statistical area | Q4 1997- Q2 1998 |
| Tribal Designated Statistical Area Program (Proposed) | Federally recognized tribes outside Oklahoma without a land base | Delineate an identifiable land area as a tribal designated statistical area | Q1-Q3 1998 |
| Alaska Native Regional Corporation Program | Alaska Native areas | Alaska Native Regional Corporations review and update boundaries | Q4 1997- Q2 1998 |
| Alaska Native Village Statistical Area Program | Alaska Native areas | Alaska Native Regional Corporations delineate, review, and update boundaries for these areas | Q4 1997- Q2 1998 |
| State Reservation Program | State recognized tribes with a land base | State government liaison can review and update boundaries | Q1-Q3 1998 |
| State Designated American Indian Statistical Area Program (Proposed) | State recognized tribes without a land base | Replaces 1990 Tribal Designated Statistical Area program for state recognized tribes | Q1-Q3 1998 |
| Hawaiian Homelands | Areas recognized by the Department of Hawaii Homelands | New program to identify and include Hawaiian Homelands in TIGER data base | Q3 1996- Q1 2000 |
OBJECTIVETo provide the infrastructure necessary to support the Census 2000 telecommunications requirements. MAJOR FEATURES The planned Census 2000 telecommunications network will encompass communication links between the following facilities: o Census Bureau Headquarters in Suitland, Maryland o 12 Regional Offices (ROs) o Bowie, Maryland Computer Center o 12 Regional Census Centers (RCCs) o Jeffersonville, Indiana Data Capture Center (DCC) o Additional (up to 3) Data Capture Centers (DCCs) o 2 Census Bureau Telephone Centers (Hagerstown, Maryland, and Tucson, Arizona) o Approximately 475 Local Census Offices (LCOs) In addition, we intend to establish communication links to the planned opening of commercial telephone centers to support Telephone Questionnaire Assistance. Figure XI-1 shows the Wide Area Network (WAN) diagram for Census 2000. We plan to use Asynchronous Transfer Mode (ATM) as our communications link (via frame relay or another type of dedicated link) between Headquarters, Bowie Computer Center, and the DCCs. The ROs, RCCs, and LCOs will be linked to the frame relay cloud via leased T1 communication lines. MILESTONES Telecommunications requirements: Currently operational Regional Offices Q2-Q3 1997 Bowie Computer Center Q1 1998 Regional Census Centers Q4 1999 Jeffersonville Data Capture Center Q2-Q3 1999 Remaining Data Capture Centers Q1 2000 Local Census Offices
-------------------
|Jeffersonville, IN |
|Data Capture Center|
-------------------
|
|T1 or T3 ATM
|
---- ----- Quantity 3
_______| HQ |_____________________| DCCs |each in different city
| ---- T1 or T3 ATM -----
T3 or OC1 ATM| |
| | T1
--------------- |_____
| Bowie, MD | |
|Computer Center| |
--------------- ============= T1 128k PVC to each
|____________| Frame Relay |__________ 128/256kbps -----
T1 ============= /_________________| ROs |
| | -----
| | |
| | |
| | 128k|
| | PVC |
T1 128k PVC to each| | T1 1024k PVC |
256k bps| | to each ------
| |______________________________| RCCs |
| ------
|
------ Quantity 475
| LCOs |each in different city
------
OBJECTIVETo utilize the best available data capture methodology in Census 2000. MAJOR FEATURES The Census 2000 data capture methodology must utilize the best available technology that will accommodate the use of respondent-friendly questionnaires. The Census Bureau also will continue efforts to identify components of the data capture process that may be best performed and provided by private-sector partners. The Census Bureau will be able to take advantage of available commercial off-the-shelf hardware and software representing technological advancements in information technology and systems without limiting itself to creating in-house solutions. The following are the most significant features of the Data Capture System 2000: (DCC 2000) o Fewer centers will be responsible for data capture and data processing functions than for the 1990 census. o A full electronic data capture and processing system will record an image of every questionnaire. o Mail-return questionnaires will be sorted automatically to ensure timely conversion and capture of critical information needed before nonresponse followup activity begins. o Optical mark recognition (OMR) will be used for all check-box data items. o Intelligent character recognition (ICR) will be used to capture write-in character-based data items. o Key-from-image will capture and/or resolve difficult ICR cases. o Quality assurance will be conducted on all data keying activities. o Paper questionnaires will be handled only at the beginning of the data capture process: during check-in, forms preparation, and scanning. All subsequent operations will be accomplished using the electronic image and captured data, reducing the logistical and staffing requirements for handling large volumes of paper questionnaires. MILESTONES Q1 1997 Award contract Q2 1997 Draft system development plan Q2 1997 Draft operations and facilities plan Q3 1997-Q3 1998 Demonstrate plan Q3 1998 Complete final system development plan/operations plan Q3 1998 Complete pre-production system design Q4 1998-Q3 1999 Open data capture centers/install equipment 3/17/2000 Begin data capture of Census 2000 forms
OBJECTIVETo develop an effective and efficient system for controlling, managing, and processing Census 2000 data. MAJOR FEATURES The Census 2000 Data Processing System will be a complex network of operational controls and processing routines intended to store and service the decennial control and data requirements. It will include the necessary interactions with the Master Address File (MAF), Decennial Field Interface (DFI), Data Capture System 2000 (DCS 2000), and telephone questionnaire assistance (TQA), not only to control, accept, and store the data but also to provide the necessary computer processing to produce a one-number census. The Census 2000 Data Processing System is divided into three operational phases of precensus, census, and post-census activities. o Precensus activities will be those required for converting the MAF into the decennial control data base that remains linked to both TIGER and the MAF. These activities will include form sampling (long or short), targeting identifications (for example, with foreign language questionnaires), and preparing the address files for printing on the questionnaires. Control information (both geographic and address related) will be provided for guiding both field canvassing and address capture processing through post-capture processing. o Activities concurrent with census data collection/capture will be those necessary to coordinate the check in and storage from the multiple sources of collection (DCS 2000, Be Counted questionnaires, telephone, and possibly the Internet), to define the responding/nonresponding universes, and to provide enumeration controls and workload to the field. Included in these activities will be the loading and updating of the central data bases for the storage of all census responses provided through the enumeration and data capture processes. o Post-census activities will be those necessary to prepare data from the original responses for release. These activities will include unduplicating multiple responses, editing and imputation, coding of write-in response data (such as race, language, industry and occupation, place of work/migration), estimation, tabulation recoding, and data disclosure avoidance. In addition, the detailed data files will be prepared from information collected on the long-form questionnaires. These activities will include editing the responses, applying statistical techniques to account for missing data, and applying weights to these sample records. The files will be provided to DADS for data dissemination MILESTONES Q3 1999 Receive the Census 2000 MAF Q3 1999 Send initial address files to printing contractors Q1 2000 Provide update/list address files to the DFI Q1-Q2 2000 Accept responses from DCS 2000, telephone, and possibly the Internet Q2 2000 Define nonresponse followup universe and samples Q2-Q4 2000 Code write-in response data Q3 2000 Provide raw data for quality check person matching Q3 2000 Process 100-percent edits and imputations Q3-Q4 2000 Process 100-percent estimations, disclosure avoidance, and tabulation recoding Q4 2000-Q1 2001 Provide 100-percent estimated and edited files to DADS Q4 2000-Q3 2001 Prepare detailed data files from information collected on long-form questionnaires
OBJECTIVEThe Tabulation and Dissemination Program for Census 2000 will be significantly different from those of previous censuses. By taking advantage of new technology, the Census Bureau will be able to meet customer demand for faster and more flexible access to census data. MAJOR FEATURES Census 2000 data will be tabulated and disseminated using the Data and Access Dissemination System (DADS). Still in the developmental stages, DADS will provide an interactive electronic system that will allow data users to access prepackaged data products, data documentation, and on-line help, as well as build custom data products on-line and off-line. Figure XII-1 depicts the various DADS products. Certain data products--such as those including summary, profile data--will be disseminated in traditional media as well as through DADS. The options and issues related to determining the types of data products and their medium of dissemination have been discussed with various segments of the data user community; these consultations will continue until the final decisions are made. The Census Bureau has solicited the advice and recommendations of data users throughout the planning, design, and testing stages of DADS. DADS will be accessible to the widest possible array of users through the Internet, Intranet, and all available intermediaries, including the nearly 1,800 Data Centers and affiliates, the 1,400 Federal Depository libraries and other libraries, universities, private organizations, and so forth. DADS is being designed with the capability to: o Provide access to Census 2000 data such as the type of information shown in the 1990 census P.L. 94-171 redistricting files, excerpts of Summary Tape Files 1 through 4, and the Equal Employment Opportunity Special File, as well as Public-Use Microdata Samples (PUMS). Data users will have online viewing, downloading, and ordering capabilities. o Create customized products, including various display formats such as tables, charts, graphs, and maps based on Census Bureau or user-defined geographic areas. o Furnish metadata that provides documentation and explanatory information for data subjects and geographic areas. o Provide users with an on-line help feature for using the system and accessing census data, as well as instructions on how to seek further assistance. There are many issues concerning Census 2000 data dissemination that we must resolve in-house and with the data user community. We plan to work with data users during the next couple of years to answer four broad questions (as well as some more detailed issues) that will help finalize the overall design of the 2000 Tabulation and Dissemination Program. The questions are as follows: 1. Since data can be downloaded by data users from DADS onto other media (for example, CD-ROM, diskette, and paper), what proportion of Census Bureau resources should be used to generate CD-ROMs and printed reports? (Should any other media be considered?) 2. Since some segments of the data user community have less access to computers, how should their data needs be met? 3. What types of geographic products should accompany each data product? 4. What priority do historical data have compared with releasing information from Census 2000? (Are population and housing unit counts needed from previous censuses?) Also, data users will be asked for recommendations on the types of data to be predefined in DADS and included in various data products. Our ongoing channels of communication will continue to inform on the progress of DADS, such as through meetings and workshops with specific groups and organizations, and articles in census publications. MILESTONES Q3 1996 Released DADS Prototype 1, "proof of concept" for basic design, technology, and functionality of DADS Q4 1997 Release DADS Prototype 2, with expanded functions such as data product creation and on-line help Q4 1998 DADS begins creating Dress Rehearsal products Q1-Q2 1999 DADS begins release of Dress Rehearsal product prototypes for Census 2000 (redistricting data first, other 100-percent and sample data to follow) Q1 2001 DADS begins creating and releasing Census 2000 redistricting data products 3/31/2001 DADS completes release of redistricting data to states
OBJECTIVETo satisfy the requirements for (Public Law) P.L. 94-171, the Census Bureau established the Census 2000 Redistricting Data Program. This program offers the redistricting officials in each state the opportunity to provide the information used by the Census Bureau to create the geography for tabulating their redistricting data. By using the state-provided geographic information, the Census Bureau can furnish redistricting data and related geographic products that enable the states to complete Federal and state redistricting according to their specific state deadlines. MAJOR FEATURES The Census 2000 Redistricting Data Program consists of: Phase 1, the Block Boundary Suggestion Project, offers state redistricting officials the opportunity to identify visible features that they suggest be held as Census 2000 block boundaries. The Census Bureau will identify these boundaries in its TIGER®* data base and, once agreed upon, hold them as block boundaries for tabulation of Census 2000. Phase 2, the Voting District Project, is the phase where state redistricting officials may submit the boundaries and geographic codes of the voting districts (election precincts) and state legislative districts using whole census blocks. The Census Bureau will insert these boundaries into the TIGER data base. Phase 3, Release of Census 2000 Redistricting Data, is the dissemination of Census 2000 data and accompanying geographic products to the governor and majority and minority legislative leaders responsible for redistricting in each state. States that provided voting districts will receive their data tabulated by voting district. States that provided state legislative districts will receive their data tabulated by state legislative district. The products, in paper and electronic form as appropriate, include: o Data files for standard tabulation areas (for example, county, city, census tract), census block, and--when provided by the state--voting districts and state legislative districts: broken down by major race groups and Hispanic origin, for the total population and housing units and for persons 18 years and over *TIGER® is a registered trademark of the U.S. Bureau of the Census. For ease of presentation, the trademark symbols for TIGER and TIGER-related products are omitted from the text. o TIGER/Line files (including voting districts and state legislative districts when provided by the state) o County Block Maps (displaying voting districts and state legislative districts when provided by the state) o Voting District Outline Maps (displaying state legislative districts when provided by the state) o Census Tract Outline Maps o Congressional District Atlas, 108th Congress of the United States, in digital form and as a printed report, including a series of maps and tables from the TIGER data base for the districts of the 108th Congress resulting from the Census 2000 reapportionment o Wall map of the United States by the Congressional Districts of the 108th Congress The Census Bureau is required by P.L. 94-171 to deliver redistricting data/ geographic products to the states within one year after Census Day. Individual states have their own timing requirements for the completion of state and Federal redistricting. Priority processing of census data and geographic products will need to be given to those states that must complete redistricting early. MILESTONES Q1 1998 Complete Phase 1, Block Boundary Suggestion Project Q1 1999 Begin Phase 2, Voting District Project 3/31/2001 Complete Phase 3, release of P.L. 94-171 redistricting data/geographic products to states Q1 2003 Release Congressional District Atlas, 108th Congress
OBJECTIVEThe requirements for Census 2000 are not just to collect, tabulate, and disseminate data, but to relate these data to geographic entities. Data for smaller geographic areas are necessary to meet the requirements of redistricting, the Home Mortgage Disclosure Act, and numerous other Federal, state, and local programs. So that data users may understand and effectively use census data, the Census Bureau provides geographic products and appropriate tools to identify the names, boundaries, codes, and other attributes of the geographic entities. MAJOR FEATURES The names, boundaries, and attributes of the geographic entities for which we tabulate data are identified through a variety of means, including internal Census Bureau operations and participant programs with state, local, and tribal governments and regional and metropolitan planning agencies. We also work with these participants to update the features--including their names and address ranges--shown in our geographic data base. We incorporate the information relating to the geographic entities and features into the TIGER®* data base. We prepare extracts from TIGER for use in conducting the census and in tabulating and disseminating census data. (Special geographic extracts from the TIGER data base will support DADS.) The geographic products planned for Census 2000 are: o Maps, in digital and hardcopy form. These will include detailed maps (such as the County Block Maps), outline maps (such as the County Subdivision Outline Maps), and thematic maps (such as percent of population by county for a specific racial category). We also will generate supporting maps based on a specific data product. o TIGER Extracts, in digital form (hardcopy may be available). These will include such products as the TIGER/Line files, the TIGER/Census Tract Street Index, cartographic boundary files, comparability files, and the TIGER/Geographic Identification Code Scheme. We will generate appropriate supporting TIGER extracts based on specific data products and to support noncensus programs (for example, the TIGER/Census Tract Street Index for the Home Mortgage Disclosure Act). *TIGER® is a registered trademark of the U.S. Bureau of the Census. For ease of presentation, the trademark symbols for TIGER and TIGER-related products are omitted from the text. The geographic entities--we report in data products--vary. Some geographic entities and related products remain constant across data products. Other geographic entities are relevant at different times and in different products. For example: o Voting districts are a very important geographic entity for P.L.94-171 geographic products but are not included in the "regular" census products. We prepare special maps and geographic products showing these areas. o While 100-percent data (from the questions asked of all persons) are available for census blocks, sample data (from the long form) are not. Similarly, some data files will not include certain geographic entities or geographic entities below a certain population (for example, places with populations of fewer than 10,000). o We determine the urban and rural population (including the delineation of urbanized areas) based on the results of the decennial census and cannot identify the urban and rural population until the total population has been determined. At the same time, data for urban and rural components traditionally have been included in the 100-percent data tabulations. o Reapportionment from Census 2000 will be reflected in the redistricting for the 108th Congress. Initial data files for Census 2000 will contain the districts of the 106th Congress. MILESTONES 3/31/2001 Release products for Redistricting Data Program to states Q1 2001 Release TIGER/Line files Q2 2001 Release county block maps, census tract outline maps, county subdivision outline maps Q3-Q4 2001 Release products for districts of 108th Congress Q4 2001 Release special tabulations products--census (for example, traffic analysis zones) Q4 2001 Release special requests products--data users
(Not shown on Internet Version.)
OBJECTIVE
The Census Bureau is engaged in an ambitious testing and research program
designed to develop new approaches and techniques for possible implementation
in Census 2000. The ultimate goal of this program is to boost participation
in the census, thereby improving coverage and reducing costs.
MAJOR FEATURES
o Even before the 1990 census data products were completely released,
the Census Bureau initiated many tests and research projects (such as
focus group and cognitive studies), spanning from 1992 through 1994.
Earlier testing was needed to allow time to study fundamental reforms
in census design. The following is a summary of these efforts:
- The earliest tests and research focused on ways to increase the
willingness and ability of respondents to respond by mail to the
census by using: (1) questionnaire design and mailing package
formats that were easier for respondents to understand and complete;
and (2) notifications for alerting and reminding respondents to
complete the forms.
- Various testing and research studies were conducted concerning
technologies to speed data collection and on ways to give people
greater flexibility in how they respond to the census. Other
research assessed current and emerging data capture technologies
(for example, electronic imaging and optical mark recognition) that
would offer the potential for processing cost reductions.
- Research on automated address list maintenance focused on supporting the
development of a continuously updated Master Address File linked to the
TIGER data base.
- Considerable testing and research were conducted to study administrative
records including: examining respondent attitudes about using records held
by other agencies, assessing public reaction to collecting social security
numbers, investigating various state and local administrative records
systems, creating a data base of sources of administrative records,
and developing effective methods to use when importing, standardizing,
and matching files.
o Then at mid-decade, the 1995 Census Test gave further testing to many
procedures and features that had been shown to be successful in the
earlier tests and research, along with additional test objectives.
This test evaluated 15 activities, among them a multiple mail strategy
with respondent-friendly forms, new methodology to count persons with
no usual residence, techniques of sampling for nonresponse, mail strategy
of Spanish-language forms to targeted areas, and the collection of
long-form (sample) data using various length forms (to see how response
rates were affected by form length).
o Ongoing cost modeling research developed statistical models using data
available from the various census tests being conducted. This research
experimented with the prediction of many variables, such as census
response, data collection workloads, and staff sizing of district offices.
o The 1996 National Content Survey (formally known as the U.S. Census 2000
Test) was the principal vehicle for testing and evaluating subject content
for Census 2000. It also provided information on questionnaire design
and mailing strategy, and techniques to improve coverage.
o The 1996 Race and Ethnic Targeted Test (formally known as the 1996 Census
Survey) was the principal vehicle for testing and evaluating several
major alternatives for asking the race and ethnic questions. It studied
how the proposed alternatives affected the distribution and quality of
responses compared with the 1990 questions.
o The 1996 Community Census tested various features of the quality check
(Integrated Coverage Measurement Program) on two American Indian
reservations and in an urban site. Some of these features included
techniques for measuring coverage in housing units and noninstitutional
group quarters, use of administrative records for coverage improvement,
and experimental questions designed to enhance address listing procedures.
OBJECTIVE
The purpose of the United States Census 2000 Dress Rehearsal is to test all
the various operations, procedures, and questions that are planned for
Census 2000 under as near census-like conditions as possible.
MAJOR FEATURES
The Dress Rehearsal will provide for operational testing of Regional Census
Center Local Census Office, and Data Capture Center procedures and systems
planned for use in Census 2000. We will employ the full array of methods,
techniques, materials, workflows, equipment applications, and promotion and
outreach programs intended for use in Census 2000.
New procedures being considered for Census 2000--such as respondent-friendly
forms readily available in many locations, multiple mail contacts with each
household, digital capture of forms, and statistical estimation
techniques--have all been tested individually in earlier operations.
The Dress Rehearsal will provide a census-like environment to test
simultaneously those procedures planned for use in Census 2000. The Dress
Rehearsal also may include some procedures and systems that have not been
tested operationally in any prior field or processing activity because they
are needed to meet new requirements.
The Census Bureau plans to launch an unprecedented partnership effort for the
Dress Rehearsal and Census 2000. For the Dress Rehearsal, the Bureau intends
to work closely with state, local, and tribal governments, community
organizations, and others to conduct a wide range of census activities. For
example, the Bureau will ask local and tribal governments to review the
Master Address File to ensure its completeness and accuracy. Other
partnership efforts will include working closely with local and tribal
governments in implementing the promotion and outreach program, and to
facilitate the availability of census questionnaires. The Dress Rehearsal
communities will work with the Bureau as exclusive partners in the final
evaluation of planned Census 2000 operations.
From the Dress Rehearsal, we will produce prototype redistricting data
products required by P.L. 94-171, and minimal standard printed and
machine-readable 100-percent and sample data products.
Sites
The Census Bureau has selected three sites for the Census 2000 Dress
Rehearsal. The sites are the city of Sacramento, California; 11 counties in
an area near and including the city of Columbia, South Carolina; and the
Menominee American Indian Reservation, Wisconsin. The combination of a
large urban site, a small city-suburban-rural site, and an American Indian
Reservation site will provide a comprehensive testing environment for
refining planned Census 2000 methodology. These three sites reflect
characteristics that we believe will provide a good operational test of
Census 2000 procedures and systems.
o Sacramento, California
This site consists of the city of Sacramento, which had a 1990 census
resident population of 369,365 and 153,362 total housing units. The
Census Bureau's official 1994 estimate showed an increase in the city's
population to 373,964.
Sacramento contains great racial and ethnic diversity, including significant
African American, Hispanic, and Asian and Pacific Islander populations. This
diversity will allow us to test planned Census 2000 methods designed to
reduce the differential in the count and produce an accurate census for all
components of the population. Sacramento is also a primary media market,
which will allow us to implement a prototype of the Census 2000 advertising
program. The site, representing the size of a typical urban local census
office in Census 2000, will provide an understanding of the effectiveness of
census operations and systems in this environment.
o Columbia, South Carolina
This site contains the city of Columbia in its entirety, including a small
portion in Lexington County; the town of Irmo in its entirety, which is in
Richland and Lexington Counties: and the following contiguous counties in
north central South Carolina:
Chester Kershaw Newberry
Chesterfield Lancaster Richland
Darlington Lee Union
Fairfield Marlboro
The 1990 census found that the counties comprising the Columbia site had a
resident population of 650,035 and 251,874 total housing units. Our
official 1996 estimates showed an increase for the 11 counties to 671,234
persons and 290,095 housing units.
The Columbia site exhibits the characteristics of a small
city-suburban-rural area that contains living situations and socioeconomic
characteristics that we do not find in a predominately urban environment.
This site provides our only opportunity to test procedures for developing
a Master Address File in an area containing both city style addresses
(house number and street name) and non-city style addresses (rural route
or box number). The site offers a mix of difficult and presumably
easy-to-enumerate areas in a suburban and rural setting. In addition, the
relatively high proportion of African Americans in this site allows
testing of methods designed to reduce the differential in the count for this
population group and produce an accurate census for all components of the
population.
o Menominee American Indian Reservation, Wisconsin
The Menominee American Indian Reservation is located in northeastern
Wisconsin. Based on the 1990 census, the Menominee Reservation had a
resident population of 3,397 and 1,176 total housing units. Menominee
County had a 1990 resident population of 3,890 and 1,742 housing units.
(The Menominee Reservation and Menominee County share the same exterior
boundary, but pockets of interior land totaling over 2 square miles that
are part of the county are excluded from the reservation. The Dress
Rehearsal will include the entire county, however.) The Census Bureau's
official 1996 estimates for Menominee County showed an increase to
4,609 persons and 1,899 housing units.
Conducting the Dress Rehearsal on an American Indian reservation allows the
Census Bureau to test proposed Census 2000 methodologies for reducing the
differential in the count among this component of the population. The
Menominee Reservation has a high proportion of American Indians living on it
and was recommended by the Census Advisory Committee on the American
Indian and Alaska Native Populations.
MILESTONES
Q2 1997 Develop and update address lists
Q3 1997 Local officials review and correct address lists
Q4 1997 Open local census offices
4/4/1998 Census Day
Q2 1998 Complete census data collection
Q3 1998 Complete quality check data collection
Q4 1998 Complete census and quality check processing
Q4 1998 Release site counts
3/31/1999 Release prototype P.L. 94-171 redistricting products
Q4 1999 Complete Dress Rehearsal evaluations
OBJECTIVE
To detect and correct peformance errors that can significantly affect
coverage and data quality.
MAJOR FEATURES
Census 2000 Quality Assurance (QA) activities will cover critical precensus,
data collection, and data processing operations. QA plans will be developed
for the following activities:
o Field geocoding
o Field address listing, validation, and/or map spotting
o Printing of public use forms
o Input materials
o Map production
o Assembly kits - materials used by enumerators in the field
o People-assisted data collection operations
o Personal field interviewing
o Computer Assisted Personal Interviewing
o Computer Assisted Telephone Interviewing
o Data Capture modes
o Intelligent character recognition
o Optical mark recognition
o Keying from paper
o Keying from imaging
o Scanning
o Digitizing
o Clerical or automated coding of responses
Each of these operations is designed and implemented to meet decennial
objectives. The QA is tailored not only to eliminate significant nonsampling
errors, but also to be integrated efficiently into the operation workflow.
MILESTONES
Q1 1998-Q1 1999 QA on geographic support activities
(geocoding, digitizing, and map production)
Q2 1998-Q3 1999 QA on address listing and validation
Q3 1999 QA on printing public use forms
Q2 2000 QA on people-assisted data collection
Q2 2000 QA on data capture operation
OBJECTIVES
To obtain information about the quality of Census 2000 data and to provide
information for future census planning.
MAJOR FEATURES
Evaluations of key components of Census 2000 will be planned before and
implemented during the Census 2000 process. Evaluation results will be
released in the form of a Report Card on Census 2000. The Census Bureau
intends to release certain components of the Report Card by
December 31, 2000.
The components of Census 2000 to be evaluated will fall into three broad
categories:
o Quality check evaluations
o Coverage improvement evaluations
o General evaluations
MILESTONES
Q2 1997 Define components of Census 2000 Dress Rehearsal Report Card.
Q3 1997 Define components of Census 2000 Report Card.
Q4 1997 Solicit study proposals for the various components and other
ideas.
Q3 1998 Develop draft plan for production of Census 2000 Report
Card and make available for wide internal/external review and
comment.
Q4 1998 Release prototype of Census 2000 Report Card for comment.
Q2 1999 Finalize plan for Census 2000 Report Card. This plan will
reflect what we learned from the Census 2000 Dress Rehearsal
Report Card as well as from our many internal/external
consultations.
Q4 1999- Implement evaluation studies.
Q4 2000
Q1 2001 Issue Census 2000 Report Card with or shortly after release of
the Census 2000 counts by December 31, 2000.
Beyond Q1 Some evaluation studies may be prepared and issued.
2001
OBJECTIVETo conduct a program of research and experimentation during the Census 2000 cycle that will provide information for planning the 2010 census. MAJOR FEATURES The Census Bureau is conducting research on different aspects of decennial census activities. These research activities will be coordinated and managed in a comprehensive research program. The Census Bureau will conduct experiments during Census 2000 to test alternative methods that may be considered in planning the 2010 census. One experiment will test an administrative records census; other experiments will be determined. The process for managing the experimental program will involve the following: o Develop criteria for selection of experiments o Solicit ideas for experiments o Review proposals and select experiments based on pre-identified criteria and resources o Ensure that implementation of experiments is coordinated with all participating Census Bureau divisions o Monitor budget and schedules for experiments o Review results and coordinate the documentation of results into a Census 2000 results memorandum series MILESTONES Q2-Q3 1997 Define criteria for research and experimentation program Q4 1997 Solicit ideas for research and experimentation Q2 1998 Identify experiments to be included Q4 1999-Q4 2000 Implement experiments Q4 2000-Q4 2002 Document results of research and experiments
OBJECTIVE
To explore the feasibility of using administrative records in decennial
censuses.
MAJOR FEATURES
The Census Bureau is evaluating the feasibility of using administrative
records to supply missing responses for households that return their census
form but fail to answer all questions.
In addition, the Census Bureau plans to include an experiment in Census 2000
in parallel with standard methods to provide a basis for analysis and future
decision making with regard to an expanded use of administrative records in
the 2010 census.
Developing an administrative records file for use in Census 2000 involves the
following:
o Identify and acquire administrative record files from selected national
programs that contribute to coverage and to demographic characteristics
o Develop methods to evaluate the quality of each administrative record
file
o Develop state and national level administrative record data bases
o Generate administrative record household files for the Census 2000 Dress
Rehearsal and Census 2000
o Conduct experiments in Census 2000 to support planning for the 2010 census
An administrative records research agenda has been established to identify
relevant issues and the corresponding research projects that are required.
In addition, the Census Bureau is conducting privacy research to gauge public
acceptance of administrative records use.
MILESTONES
Q2 1997 Conducted privacy group meetings on use of administrative records
Q3 1997 Evaluate 1996 Community Census use of administrative records
Q1 1998 Develop state-level administrative records data base (Dress
Rehearsal)
Q3-Q4 1998 Evaluate Dress Rehearsal use of administrative records
Q4 1998 Decide on use of administrative records in Census 2000
Q4 1998 Experimental administrative records national data base
OBJECTIVETo identify research and experiments in Census 2000 that will facilitate planning of the 2010 census. MAJOR FEATURES Demographic and social changes in the United States will make the year 2010 differ from 2000 even more than 2000 differed from 1990. For example, many of the baby boomers will be out of the work force, the continuing telecommunications revolution will have rounded the corner with a generation of children brought up with computers, several minority groups will have grown considerably as a proportion of the total population, and the World War II generation that relied on social security and medicare will be replaced by those who know they cannot rely entirely on such entitlements. The demographic changes and probable reduction of Federally-run programs will influence the data requirements and the manner in which the census can be taken in 2010. Therefore, as belts continue to tighten, and society and technology continue to change, early planning for the 2010 census may be even more important than it was for Census 2000. Census 2000 provides a unique opportunity to implement research for new methodologies envisioned for the future. Conducting experiments during the decennial census eliminates many variables present in "test census" environments. Thus, the first objective for planning the 2010 census is to identify research and experiments for Census 2000 implementation to gather the data necessary to continue the "business process reengineering" of the decennial census. The results of this effort will become the basis for continued planning and ultimate selection of the best 2010 census design. Much current research is already part of the research planning for the 2010 census. For instance, technological advancements and improvements in public and private sector administrative records make it likely that the Census Bureau will be able to increase their use in future censuses. Continued research into the accuracy of sampling techniques, measures of the total accuracy of various census designs, ongoing integration of the decennial census with the American Community Survey (being developed as a replacement for data collected in the sample long form of the 2010 census), and other opportunities for procedural efficiencies are planned. Activities such as these will yield the basis of a likely continuum of possible designs for 2010. Early planning for the 2010 census includes the following features: o Identification of a continuum of potential designs o Implementation of key research contracts to inform tests for Census 2000 o Identification of key experiments and evaluations for implementation in Census 2000 o Participation of staff actively involved in Census 2000 for full integration of concepts o Early input from stakeholders MILESTONES Q4 1997 Begin identifying continuum of potential designs Q4 1997 Develop selection criteria for Census 2000 tests Q4 1997 Propose experiments to test in Census 2000 Q4 1998 Begin external advisory process for 2010 census Q1 1999 Finalize plans for administrative records census test design Q1 1999 Finalize total error test design Q1 1999 Begin implementation activities for Census 2000 experiments Q4 2000 Begin documenting empirical evidence for proposed 2010 census designs Q4 2001 Define post-Census 2000 experiments and research for 2010 census
OBJECTIVESCensus 2000 operations in Puerto Rico will be comparable in scope to stateside activities. The Census Bureau is working in partnership with the government of Puerto Rico--as represented by the Puerto Rico Planning Board (PRPB)--on the program objectives to ensure that Census 2000 meets the legal requirements set forth in Title 13, U.S. Code, as well as the specific data needs of Puerto Rico. MAJOR FEATURES Census 2000 operations in Puerto Rico will be built around the same four fundamental strategies to be used stateside: Strategy One: Build Partnerships at Every Stage of the Process o The Census Bureau will develop and sign a Memorandum of Agreement (MOA) with the government of Puerto Rico that will outline the mutual roles and responsibilities of each party in the conduct of Census 2000 on the Island. o In consultation with the PRPB and its local Interagency Committee, census questionnaire content will be developed that meets the legislative and program needs of Puerto Rico. o An advertising and promotion campaign designed to build awareness of the census and boost participation will be developed for Puerto Rico that will take into account its specific needs. o The Census Bureau will conduct an address listing operation in Puerto Rico in 1998. This will allow for the full implementation of the Local Update of Census Addresses (LUCA) program and will serve as the basis for use of the update/leave method of data collection on the Island. During the update/leave operation, field enumerators update the address list and map and leave a census questionnaire at each housing unit for the residents to complete and mail back. Strategy Two: Keep It Simple o Using the findings from our census testing and research, the Census Bureau will design user-friendly questionnaires that are simpler and easier for respondents to understand and complete. Forms will be available in both Spanish and English. o Census questionnaires and other forms will be made more readily available to respondents and will be placed at post offices, community centers, and other convenient places. Strategy Three: Use Technology Intelligently o Using the update/leave methodology for data collection for the first time in Puerto Rico will give respondents the opportunity to complete the census questionnaires themselves and return them by mail. This will allow the Census Bureau to make use of the same technological advances that will be used stateside. o The Census Bureau will make greater use of the telephone as a data collection tool, in addition to its use in providing assistance to respondents with questions about Census 2000. o Data users will have access to Census 2000 data and products through the Internet, using the Data Access and Dissemination System (DADS). DADS will give users the flexibility to extract and tabulate census data quickly to meet their specific data needs. Strategy Four: Use Statistical Methods o The Census Bureau will use personal visits as well as the telephone to obtain response from households that do not return a census questionnaire. o On a daily basis, the Census Bureau will determine the response rate for every census tract, which is a neighborhood of about 4,000 people. The response rate is defined as: Mail + Telephone + Other Responses *100% Questionnaires Mailed or Delivered o For any census tract in which this rate is less than 100 percent after the initial response period, enumerators will perform nonresponse followup (NRFU) (that is, contact the respondent and complete a census questionnaire) under the direct sampling strategy. o In the direct sampling strategy, the Census Bureau will select a sample of nonresponding addresses in each census tract at the end of the initial response period. The sample will vary from census tract to census tract based upon the tract's response level and will be designed to achieve at least a 90-percent total response rate in each tract. o Enumerators will perform NRFU for each of the selected sample addresses. The addresses will be visited by an enumerator who will complete a questionnaire by personal interview. o A quality check survey will be conducted shortly after the regular enumeration to determine if people and housing units were missed or counted more than once. This survey is designed to eliminate the undercount experienced in the 1990 census and will result in a "one-number" census that accurately reflects the population of Puerto Rico. MILESTONES Q3 1997 Finalize Census 2000 plan for Puerto Rico Q3 1997 Complete Phase 1 of Block Boundary Definition Program Q1 1998 Finalize Memorandum of Agreement Q3 1998 Complete questionnaire content determination process Q3 1998 Begin address listing activities Q3 1999 Conduct LUCA program Q1 2001 Release total counts for Puerto Rico 3/31/2001 Deliver P.L. 94-171 redistricting counts
OBJECTIVESCensus 2000 operations in American Samoa, the Commonwealth of the Northern Mariana Islands, Guam, and the U.S. Virgin Islands (collectively referred to as the Island Areas) will be conducted by the Census Bureau in partnership with the governments of each Island Area. These partnerships will work on the program objectives to ensure that Census 2000 meets the legal requirements set forth in Title 13, U.S. Code, as well as the specific data needs of each Island Area. MAJOR FEATURES Census 2000 operations in the Island Areas will be built around three of the four fundamental strategies to be used stateside (no sampling operations will be conducted in these Areas): Strategy One: Build Partnerships at Every Stage of the Process o The Census Bureau will develop and sign a Memorandum of Agreement (MOA) with the Governor of each Island Area that will outline the mutual roles and responsibilities of each party in the conduct of Census 2000 for each Island Area. o In consultation with the local Interagency Committee established by each Island Area, the Census Bureau will develop the census questionnaire content that meets the legislative and programmatic needs of each Area. o An advertising and promotion campaign designed to build awareness of the census and boost participation will be developed for each Island Area that will take into account its specific needs. o The Census Bureau will conduct the data collection of the Island Areas using the list/enumerate method. This decision was based on recommendations from Island Area representatives and an analysis of the various data collection methodologies. During the list/enumerate operation, field enumerators list the housing units and show their spatial location on a map and enumerate the residents in one visit. Strategy Two: Keep It Simple o Using the findings from our census testing and research, the Census Bureau will design respondent-friendly questionnaires and forms that are simpler and easier for the enumerators to administer and for the respondents to understand and answer. Forms will be available in English and in other languages widely spoken in the Island Areas. o Census questionnaires and forms will be made more readily available to respondents and will be placed at post offices, community centers, and other convenient places. Strategy Three: Use Technology Intelligently o The Census Bureau will make greater use of the telephone as a data collection tool, in addition to its use in providing assistance to respondents with questions about Census 2000. o Data users will have access to Census 2000 data and products using the Data Access and Dissemination System (DADS). DADS will give users the flexibility to extract and tabulate census data quickly to meet their specific data needs. MILESTONES Q4 1997 Finalize Census 2000 plan for each Island Area Q3 1998 Finalize Memorandum of Agreement for each Island Area Q4 1998 Complete questionnaire content determination process 3/31/2001 Release total counts for each Island Area
ACF (Address Control File) The residential address list used in the
1990 census to label questionnaires, control the mail response
check-in operation, and determine the nonresponse followup
workload.
AIANA (American Indian and Alaska Native area) A Census Bureau term
referring to these entity types: American Indian reservation,
American Indian subreservation area, American Indian trust
land, state designated American Indian statistical area,
tribal jurisdictional statistical area, tribal designated
statistical area, tribal subdivision, Alaska Native Regional
Corporation, Alaska Native village, and Alaska Native village
statistical area.
CATI (Computer Assisted Telephone Interviewing) A method of data
collection using telephone interviews in which the questions
to be asked are displayed on a computer screen and responses
are entered directly into the computer.
CCD (Census county division) A subdivision of a county that is a
relatively permanent statistical area established
cooperatively by the Census Bureau and local government
authorities. Used for presenting decennial census statistics
in those States that do not have well-defined and stable minor
civil divisions that serve as local governments.
CD-ROM (Compact Disk - Read Only Memory) An optical disk that is
created by a mastering process and used for storing large
amounts of data. Unlike standard computer disks and diskettes,
CD-ROMs can be used only to read stored data, not to update or
change its content.
CFO (Census Field Office) One of 425-450 temporary Census Bureau
offices to be established in Census 2000 to manage address
listing field work and conduct local recruiting.
CV (Coefficient of variation) The ratio of the standard error
(square root of the variance) to the value being estimated,
usually expressed in terms of a percentage (also known as the
relative standard deviation). The lower the CV, the higher
the relative reliability of the estimate.
DADS (Data Access and Dissemination System) A generalized
electronic system for all access and dissemination of Census
Bureau data. This interactive electronic system will be
designed to allow efficient and cost-effective access to data
generated by the various areas of the Census Bureau. The
DADS system will serve as the vehicle for accessing and
disseminating data from Census 2000 and from the American
Community Survey.
DANC (Decennial Applicant Name Check) This automated system will
be used to screen all applicants backgrounds for criminal
histories to facilitate the selection, hiring, promotion, and
payrolling of qualified and suitable applicants for the
conduct of Census 2000.
DCC (Data Capture Center) One of four decentralized Census Bureau
facilities (one permanent, three temporary) that will check in
Census 2000 questionnaires returned by mail, create images of
all questionnaire pages, and convert data to computer readable
format using OMR, ICR, and data keying technologies. The DCCs
also will perform other computer processing activities,
including automated questionnaire edits, work flow management,
and data storage. Called "processing office" (PO) in previous
censuses.
DCS 2000 (Decennial Census 2000) The data capture system that will be
used to capture information from census forms. This system
will incorporate the following activities: processing more than
120 million incoming forms; digitally capturing and processing
billions of bits of information on the forms; converting
automatically the image of the form to text-based data; and
editing/repairing data that the system is unable to decipher
automatically.
DFI (Decennial Field Interface) The collection of systems that
will be used in census field offices to control and manage
the census data collection effort. It includes, among others,
the operations control, payroll and personnel, map production,
and management information systems.
DSF (Delivery Sequence File) A computerized file containing all
delivery point addresses serviced by the USPS. The USPS
updates the DSF continuously as its letter carriers identify
addresses for new delivery points or changes in the status
of existing addresses.
GQ (Group quarters) A place where people live that is not the
typical household-type living arrangement. The Census Bureau
classifies all persons not living in households as living in
group quarters. There are two types of group quarters:
institutional (for example, correctional facilities, nursing
homes, and mental hospitals) and noninstitutional (for example,
college dormitories, military bases and ships, hotels, motels,
rooming houses, group homes, missions, shelters, and
flophouses).
HH (Hawaiian Homelands) Areas created as a result of the Hawaiian
Homes Commission Act of 1920 to provide agricultural, pastoral
and residential land for native Hawaiians.
HU (Housing unit) A house, an apartment, a mobile home, a group
of rooms, or a single room that has its own kitchen facilities,
a separate entrance, and is occupied as a separate living
quarters or, if vacant, intended for occupancy as a separate
living quarters.
ICR (Intelligent character recognition) Technology that uses an
optical scanner and computer software to "read" human
handwriting. Sometimes referred to as "optical character
recognition" (OCR).
LCO (Local Census Office) One of approximately 475 temporary
Census Bureau offices to be established for Census 2000
data collection purposes. Called "district office" (DO) in
previous censuses.
L/E (List/enumerate) A method of data collection in which
enumerators list each residential address and enumerate
the household in one visit.
LUCA (Local Update of Census Addresses) A Census 2000 program,
established in response to requirements of P.L. 103-430,
that provides an opportunity for local and tribal governments
to review and update individual address information in the
MAF and associated geographic information in the TIGER data
base to improve the completeness and accuracy of both computer
files.
MAF (Master Address File) A computer file based on a combination
of the addresses in the 1990 ACF and current versions of the
DSF, supplemented by address information provided by state,
local, and tribal governments. The MAF is being updated
throughout the decade to provide a basis for producing address
labels needed to deliver Census 2000 questionnaires, keep
track of which forms have been returned and which need
followup, serve as the sampling frame for the Census Bureau's
periodic demographic surveys, and support other Census Bureau
statistical programs.
MCD (Minor civil division) A primary government and/or
administrative subdivision of a county, such as a township,
precinct, or magisterial district.
MO/MB (Mailout/mailback) A method of data collection in which the
USPS delivers addressed questionnaires to residents who
are asked to complete and mail back the questionnaire to the
appropriate Census Bureau office.
NRFU (Nonresponse followup) A census followup operation in which
temporary field staff, known as enumerators, visit addresses
from which no questionnaire was returned by mail, from which a
telephone response was not received, or for which no
administrative records could be located.
OMR (Optical mark recognition) Technology that uses an optical
scanner and computer software to scan a page, recognize the
presence of marks in predesignated areas, and assign a value
to the mark depending on its specific location on a page.
PALS (Program for Address List Supplementation) A program providing
governmental units and regional and metropolitan agencies
the opportunity to submit lists of individual addresses for
their community to the Census Bureau for use in building
the MAF. This program, enabled by the census Address List
Improvement Act of 1994 (P.L. 103-430), will help ensure
the completeness and accuracy of the MAF and the TIGER
data base.
P.L. 94-171 (Public Law 94-171) The public law that requires the Census
Bureau to provide selected decennial census data tabulations
to the states by April 1 of the year following the census
enumeration. These tabulations are used by the states to
redefine the areas included in each Congressional district
and in other districts used for state and local elections, a
process called redistricting.
P.L. 103-430 (Public Law 103-430) The public law that amends Title 13,
United States Code, to allow designated local and tribal
officials access to the address information in the MAF
to verify its accuracy and completeness. This law also
requires the USPS to provide address information it compiles
to the Census Bureau to improve the MAF.
PUMS (Public Use Microdata Sample) Computerized files containing a
small sample of individual long-form census records showing the
population and housing characteristics of the people included
on those forms.
QC (Quality check) A coverage measurement methodology that will
be used to determine the number of people and housing units
missed or counted more than once in Census 2000. This
information is combined with the enumeration results before
producing a single set of official census results (the
one-number census estimates).
RCC (Regional Census Center) One of 12 temporary Census Bureau
offices established to manage LCO activities and to conduct
geographic programs and support operations such as automated
map production.
RO (Regional Office) One of 12 permanent Census Bureau offices
established in 12 cities throughout the country to implement
outreach and promotion activities during the census period and
to conduct survey enumeration and other decentralized work
of the Census Bureau.
SBE (Service-based enumeration) An operation designed to
enumerate people at places where they might receive services,
such as shelters, soup kitchens, and other selected locations.
This operation targets the types of services that primarily
serve people who have no usual residence.
SP (Special place) A residence where people live or stay other
than the usual house, apartment, or mobile home. Examples
are colleges and universities, boarding and rooming houses,
marinas, nursing homes, hospitals, and prisons.
STF (Summary Tape File) A series of 1990 census summary
tabulations of complete count and sample population and
housing data available for public use on computer tape and
CD-ROM.
TIGER (Topologically Integrated Geographic Encoding and Referencing)
A computer data base that contains a digital representation
of all census-required map features (streets, roads, rivers,
railroads, lakes, and so forth), the related attributes for
each, and the geographic identification codes for all entities
used by the Census Bureau to tabulate data for the United
States, Puerto Rico, and Island Areas. The TIGER data base
records the interrelationships among these features,
attributes, and geographic codes and provides a resource for
the production of maps, entity headers for tabulations, and
automated assignment of addresses to a geographic location in
a process known as "geocoding."
T-NIGHT (Transient night) An enumeration procedure conducted to
enumerate people occupying campgrounds at racetracks,
recreational vehicle (RV) campgrounds or RV parks, commercial
or public campgrounds, fairs and carnivals, and marinas.
TQA (Telephone Questionnaire Assistance) A toll-free service that
will be provided by a commercial phone center to answer
questions about Census 2000 or the census questionnaire.
U/L (Update/leave) A method of data collection in which
enumerators personally deliver a census questionnaire to a
household to be completed and returned by mail and at the same
time update the address list.
USPS (United States Postal Service) The organization responsible
for delivering the mail questionnaires in Census 2000, and
the producer of the DSF.
Program AreaContact Person Telephone Number Marketing Michael Freeman 301-457-2984 Partnerships Brenda August 301-457-4033 Content Determination Louisa Miller 301-457-2073 Forms, Printing and Mailing Carolyn Hay 301-457-4008 Address List Development Linda Franz 301-457-1014 Geographic Services Robert LaMacchia 301-457-1022 Office Infrastructure Harold Hayes 301-763-8111 Automated Collection Howard Prouse 301-457-1933 Personal Visit/Be Counted Robert Rinaldi 301-457-2014 Special Populations Annetta Clark-Smith 301-457-2378 Telephone Questionnaire John Marshall 301-457-1943 Assistance/Internet Data Capture Alan Berlinger 301-457-1737 Data Processing Maureen Lynch 301-457-4092 Statistical Design Ann Vacca 301-457-4304 Quality Check Operations David Whitford 301-457-4035 Dissemination and Products Enrique Gomez 301-457-3007 (DADS) Evaluation Tommy Wright 301-457-1030 Research/Experimentation Florence Abramson 301-457-4222 2010 Census Planning Jay Keller 301-457-4040 Puerto Rico/Island Areas Lourdes Flaim 301-457-4041 General Customer Services. . . . . . . . . . . . . . . . . . . . . . . 301-457-4100 Census Locator . . . . . . . . . . . . . . . . . . . . . . . . 301-457-1713 Census Bureau Website. . . . . . . . . . . . . . . . . . . www.census.gov/ Census Bureau Regional Offices (Information Services, Data Product Information) Atlanta, GA . . . . . . . . . . . . . . . . . . . . . 404-730-3833/3964 (TDD) Boston, MA . . . . . . . . . . . . . . . . . . . . . . 617-424-0510/0565 (TDD) Charlotte, NC . . . . . . . . . . . . . . . . . . . . 704-344-6144/6548 (TDD) Chicago, IL . . . . . . . . . . . . . . . . . . . . . 708-562-1740/1791 (TDD) Dallas, TX . . . . . . . . . . . . . . . . . . . . . . 214-640-4470/4434 (TDD) Denver, CO . . . . . . . . . . . . . . . . . . . . . . 303-969-7750/6769 (TDD) Detroit, MI . . . . . . . . . . . . . . . . . . . . . 313-259-1875/5169 (TDD) Kansas City, KS . . . . . . . . . . . . . . . . . . . 913-551-6711/5839 (TDD) Los Angeles, CA . . . . . . . . . . . . . . . . . . . 818-904-6339/6249 (TDD) New York, NY . . . . . . . . . . . . . . . . . . . . . 212-264-4730/3863 (TDD) Philadelphia, PA . . . . . . . . . . . . . . . . . . . 215-597-8313/8864 (TDD) Seattle, WA . . . . . . . . . . . . . . . . . . . . . 206-728-5314/5321 (TDD) Regional Office Liaison at Headquarters. . . . . . . . . . . . . .301-457-2032
Return:"The Plan for Census 2000" (summary) page