Mr. Chairman, Mrs. Maloney, and members of the Committee:

Thank you for this opportunity to testify today on plans for Census 2000 in Puerto Rico. I want to introduce Ms. Teresa Angueira, who is Assistant Division Chief in our Decennial Management Division and is in charge of planning for Census 2000 in Puerto Rico. I may be calling on Ms. Angueira for technical answers during the questioning.

Puerto Rico has been included in every decennial census since 1910; the U.S. government also conducted a special census in Puerto Rico in 1899. Title 13 of the U.S. Code, which sets out the basic law under which we conduct the census, specifies that in addition to the 50 states and the District of Columbia, the census should include the Commonwealth of Puerto Rico, the U.S. Virgin Islands, Guam, the Commonwealth of the Northern Mariana Islands, and "such other possessions and areas over which the United States exercises jurisdiction, control, or sovereignty."

Census 2000 plans are well along, and on schedule. I was honored to have the opportunity to visit Puerto Rico in early August, at which time I had a luncheon meeting with Congressman Carlos Romero-Barceló, met with Governor Pedro Rosselló and participated in an important Census 2000 Town Hall meeting of mayors, government officials, and data users.

The essential thing to remember about Census 2000 operations for Puerto Rico is that they will be comparable to those in the states. We will conduct the full range of operations to ensure a complete and accurate enumeration, tailoring them as necessary, in consultation with our partners in Puerto Rico, to meet local needs.

Mr. Chairman, I will first address the issues concerning tabulations raised in your letter of invitation and in the Report language proposed by Representative José Serrano.

To cite the Report language:

"The Committee also expects the Census Bureau to include Puerto Rico in any electronic access to census data as a 'state equivalent' in the same manner as the District of Columbia is included. Similarly, any extracts of census data like the 1990 Public-Use Microdata Sample should treat Puerto Rico in the same manner as the District of Columbia."

Electronic access to Census 2000 data will be through a newly developed system called the American FactFinder. This system is available now on the Census Bureau's Internet homepage. Not only will Puerto Rico be treated as a 'state equivalent,' we are developing a special Spanish-language "interface" for Puerto Rico. As for Public-Use Microdata Samples, we did have comparable samples for Puerto Rico in 1990 and would expect to again in 2000. But I should caution that tabulation plans are not final, and we do not yet have details about the microdata samples, for the 50 states, for the District of Columbia or for Puerto Rico. We have every expectation of treating Puerto Rico in the same manner as the District of Columbia.

To further cite the Report language:

"The Committee expects the Census Bureau to include Puerto Rico in any national totals in these national summary data products."

The issue of including Puerto Rico in the "national totals" would differ from what we announced in Census 2000 Decision Memorandum No. 64. The Decision Memorandum, issued in November 1998, stated that we would include data for Puerto Rico in the national summary data products for Census 2000. This was in keeping with my response to a letter from Congressman Romero-Barceló also in November 1998 and to a commitment made in May 1996 by then Secretary of Commerce Mickey Kantor to Governor Rosselló. This means that, unlike 1990 when data for Puerto Rico were included only in one national summary table showing total population, in 2000 census data for Puerto Rico will be included in every national summary report, including those presenting data on population and housing characteristics.

The best way to understand the difference between the Decision Memorandum and the Report language is to compare 1990 with the plan for 2000, and then the plan for 2000 with illustrative tabulations responsive to the Report language.

I refer first to Attachment 1 of my written statement, which has two pages. The first page shows national summary totals for certain characteristics data as presented in 1990. There is no line for Puerto Rico on this page. Compare this with the second page of Attachment 1, which illustrates the plan for 2000 as set forth in Decision Memorandum No. 64. On this page you will see a separate line for Puerto Rico. This is the current plan: to display Puerto Rico counts on all national summary tables which report population and housing characteristics. The plan does not preclude the inclusion of Puerto Rico counts in the "national totals."

Attachment 2, which also has two pages, displays what we believe would be required by the Report language that Puerto Rico counts be included in "national totals." Page 1 of this attachment shows the only table produced for the 1990 Census where totals for Puerto Rico were included on the same page as those for the United States, as traditionally defined. You will see a line at the top that says "Total." This is the total population count for all areas and groups covered in the 1990 Census, including, of course, Puerto Rico. Below that is a line indicating "United States" and then separate totals for each Outlying Area and three categories of U.S. population abroad. For purposes of this testimony I interpret the line "United States" and not the line "Total" to be equivalent to the term "national totals" used in the Report language.

Page 2 of the attachment reconfigures the 1990 table to follow the proposed Report language. As you can see, we no longer have a separate line for Puerto Rico under Outlying Areas. And the 1990 U.S. total, instead of being 248.7 million is now 252.2 million because it includes Puerto Rico's population. Had we presented the 1990 counts in this manner, for the last nine years every ratio, proportion, or other statistic in the country that takes the total U. S. population as its denominator would have been different.

Under Decision Memorandum No. 64, it is the Census Bureau's plan to include Puerto Rico in national summary data products, but not in "national totals."

Though it is not our present intent, it is feasible to include Puerto Rico in the national totals. The decision, however, would have to be made prior to April 30, 2000, in order to incorporate it into our tabulation planning process. After that date it would be difficult to implement a change from our present plans for the Census 2000 tabulation.

I hope that my testimony to this point answers the questions about tabulation posed for this hearing. I would also like to take a few minutes to explain the logic behind Decision Memorandum No. 64.

The Census Bureau and other federal statistical agencies historically have defined the "United States total" as the aggregate of the states and the District of Columbia. The Census Bureau would not normally deviate from historic practice in such a major way without prior and thorough consultation with relevant stakeholders, including other members of the statistical system.

To tabulate Puerto Rico in the "United States" total for the 2000 census would introduce a statistical redefinition of the United States. There are many consequences to be considered. For example, let me refer you back to Attachment 2, where the first page is a 1990 census table that displays land and water area as well as population. In the mockup of this table, page 2 of the Attachment, we have included Puerto Rico's area measurements as well as its population in the "United States" line. The Census Bureau is unsure whether that is the intention of the proposed Report language. If not, the population measures and the area measures would have inconsistent denominators. If it is the intent, every measure of population density, which itself is a denominator for other statistical uses, would change compared to 1990.

This is an example of how Census Bureau products could be affected. There also are far-reaching consequences of data compatibility with products from other Federal statistical programs. Different statistical definitions of the "United States" from one agency to another would affect comparability in all of our statistics -- demographic, economic, health, education, agricultural, justice, etc. For example, a recently released Bureau of Labor Statistics report on "Employment and Average Annual Pay for Large Counties, 1997" includes data for San Juan, Puerto Rico. The tabular presentation also includes a line labeled "United States" and the BLS is careful to note, and I quote, "Calculations for the United States do not include data from Puerto Rico."

For reasons of statistical consistency, the Census Bureau would hesitate unilaterally to establish a new denominator. A statistical system requires common definitions if it is to be a "system." There are few definitions more basic than what constitutes the national total population. Any fundamental change in this definition should be fully explored with stakeholders within and outside the federal government.

It is the recommendation of the Census Bureau that the policy set forth in Decision Memorandum No. 64 -- which will result in Puerto Rico being shown in all Census 2000 national summary tables of population and housing characteristics, but not included in the "national totals" -- be controlling pending broad consultation and more intensive review of the implications of altering the basic denominator for nearly every statistic generated over the next decade.

I next want to briefly describe how we will conduct Census 2000 in Puerto Rico, with references to major differences from the 1990 census. This will respond to issues 2 and 3 in your letter of invitation.

Questionnaire Content

In past censuses, the questionnaires (short and long form) for Puerto Rico have been similar to those in the states, but with differences in keeping with local needs and conditions. For example, in 1990, the stateside questionnaires, but not those for Puerto Rico, contained questions on race and Hispanic origin. In October 1997, the Planning Board, on behalf of the Population and Housing Interagency Committee that it chaired, requested that the questionnaire content for Puerto Rico be the same as that of the 50 states. We have accepted that recommendation with only minor modifications required in a few questions and response categories on the long form.

As required by Title 13, on March 30 of last year, the Census Bureau informed the U.S. Congress of the planned questions for Census 2000. Title 13 requires that we tell the Congress 2 years before Census Day, the exact questions that we plan to ask. As we reported to the Congress, Census 2000 will include two questionnaires: a short form and a long form. The short form will contain a core set of seven questions that will be asked of the entire population and about every housing unit. The long form will include both the core set of questions and 45 additional questions that will be asked of a sample of the population and housing units. The short form will be delivered to 5 out of every 6 housing units, and the long form will be delivered to about 1 out of every 6 of the Puerto Rico's housing units. While the questionnaires for Puerto Rico did not have to be submitted to Congress, they were approved by the U.S. Office of Management and Budget, which approves all Federal data collections. The contracts for printing questionnaires have been awarded and printing has begun.

Using the findings from our census testing and research, we are designing user-friendly questionnaires that are simpler and easier for respondents to understand and complete. Forms will be available in both Spanish and English. As I will explain later, questionnaires will be delivered to each housing unit by census takers and the householders will be asked to respond in the mail. Multiple contacts will be made with each housing unit through alert and reminder postcards that will be sent before and after the questionnaires. Additionally, census questionnaires and other forms will be available to any respondents who did not receive one at their residence. They will be placed at walk-in assistance centers and other convenient places that people frequent. Householders can also respond by telephone if they choose.

Data Collection

With regard to data collection, the biggest difference between Census 2000 and previous censuses is that for the first time residents of Puerto Rico will be allowed to respond by mail. In previous censuses, the Census Bureau sent census takers to each housing unit to conduct a personal interview. For Census 2000, in March of next year, we will send census takers to each housing unit to update the census address list and to leave a questionnaire. We will ask householders to fill out the questionnaire and mail it back to the census office for their area. We will follow up on every housing unit for which we do not receive a questionnaire to complete the enumeration. We have tested this method in the United States and have used it for the last several censuses. It will allow us to conduct a more accurate and cost-effective population count for Puerto Rico.

The increase in limited-access communities and a population that mirrors stateside in the prevalence of two-income working families made it imperative that the Census Bureau implement a new data collection method. The new procedures will improve coverage as we will "cover the ground" more than once and it will also allow us to build an address list for future census efforts. Another benefit is that the new procedures allow for self-response. Census Bureau studies have shown that self-response results in higher quality data than that obtained through enumerator interviews.

In order to conduct a census by mail, we must first create a master list of all residential addresses/location descriptions in Puerto Rico. In February 1998, we opened the Puerto Rico area office in San Juan. Between June and October 1998, we opened about 24 small census field offices to complete address listing work. The address listing was completed between October 1998 and January 1999. This summer, as part of the Local Update of Census Addresses (LUCA), we gave listings to municipio officials for review. This is a new program mandated by the Congress that was not available last time. The review period for the LUCA concluded in early July. This fall, we give the municipio officials the results of the recanvass work done to check any discrepancies noted. We have closed the address listing offices, but this summer, we opened 9 local census offices to conduct data collection operations next year. These offices have hired management staff.

Accuracy and Coverage Evaluation

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 housing units in the states as well as in Puerto Rico. This is the Accuracy and Coverage Evaluation survey. We will check the results from the mail-in, from telephone interviews, and from the personal visits. This quality check will, for the first time, correct for the undercount and lead to a census that reflects the total population of Puerto Rico.

Partnerships

Finally, I want to mention the importance of building partnerships to ensure a successful census. The Census Bureau cannot accomplish its goals alone. It needs to reach out early and consistently to find "best in class" partners to help it accomplish its objectives.

As we have for each census since 1960, we are working jointly with the Puerto Rico Planning Board in an effort to make Census 2000 an accurate census that collects the data Puerto Rico will need for the first decade of the 21st century. In February 1998, the Honorable Norma Burgos, then Chairwoman of the Planning Board, finalized the Memorandum of Agreement, which had earlier been signed by then Director of the U.S. Census Bureau Martha Farnsworth Riche. The Memorandum of Agreement amends the original 1958 Agreement between the two parties and lays out our respective roles and responsibilities. The Memorandum of Agreement was preceded by several years of joint efforts between the Planning Board and the Census Bureau and these efforts continue.

As our census partner, the Planning Board also helps by disseminating data as lead agency for the Census Data Center Program, providing technical assistance to the municipios, and coordinating municipio participation in geographic programs.

Municipio governments know the local conditions and circumstances. They helped us correct our maps and address list, such as in the LUCA program I mentioned earlier, and will tell us where to place questionnaire assistance centers. They can alert us to problems and advise us of opportunities to publicize Census 2000. They can also play a big role in encouraging participation in the census. We have mailed materials to the nearly 80 municipios encouraging them to form complete count committees. As in the states, we will implement a Census in the Schools program, specifically tailored to Puerto Rico, and, with the partnership of the Puerto Rico Department of Education, school kits will be made available to every teacher, which is an expansion over 1990. I particularly want to mention the contribution of the Secretary of Education, Victor Fajardo, who is also a member of the Census 2000 Race and Ethnic Advisory Committee on Hispanic Populations.

We are also partnering with community groups and professional associations who 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 counted and that the data collected meet the needs of Puerto Rico data users.

We partnered with the Museo de Arte de Puerto Rico, which permitted reproduction of Manuel Hernández Acevedo's painting "Plaza San José" for use in a poster that is being widely distributed for use in Puerto Rico.

And we will use private companies to manage our efforts to promote Census 2000 more visibly and effectively. For the first time, we will use paid advertising to promote participation--both in the states and in Puerto Rico. 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. Y & R Puerto Rico, the Puerto Rico-based firm selected to develop this campaign, has nearly completed production of what we believe will be a very successful campaign.

The Census Bureau is strongly committed to excel in the conduct of Census 2000. Our plan effectively tackles the problems experienced in past censuses, especially concerns about improving the completeness of the count. The plan will allow the Census Bureau to meet the multiple, and sometimes competing, challenges that face the census in Puerto Rico. To ensure a census that will serve Puerto Rico, we will continue efforts to seek and implement suggestions from a wide variety of sources.

Mr. Chairman, that concludes my testimony. I will be pleased to answer any questions.

ATTACHMENT 1

Address List Development--CITY STYLE

RESULTS

Original Master Address File Addresses 90,304,226

Plus Additions 11,094,789

Included in both LUCA and Block Canvass 3,072,069

Included in Block Canvass Only 5,687,293

Included in LUCA Only* 2,335,427

Minus Deletes 5,034,356

Deleted by both LUCA and Block Canvass 184,577

Deleted by Block Canvass Only* 4,606,730

Deleted by LUCA Only 243,049

Minus Conversions from Residential

to Non-Residential 379,208

Both LUCA and Block Canvass 4,712

Block Canvass Only 356,024

LUCA Only 18,472

TOTAL 95,985,451(numbers as of 9-99)

*Final result will change after including the findings from "field" confirmation which will be completed in November, 1999.