Mr. Chairman and Members of the Committee:

I am pleased to have the opportunity to discuss with you this morning the Commerce Department's plans for the 2000 decennial census.

The importance our nation places on this obligation was evident in 1787 when representatives at the Constitutional Convention for the first time in the history of any nation wrote a requirement for a census into the Constitution.

Since that time, the U.S. census has served not only to ensure the equitable balancing of political representation, as mandated by the Founding Fathers, but as the primary source of data on the Nation's economic and social life. The decennial census is the only comprehensive survey of the entire American population. It tells us where to put schools, hospitals, and commercial facilities. The information we provide to the American people and economy through the Census are as much a part of our infrastructure as are our roads, ports and bridges.

We have a strict Constitutional responsibility, therefore, to provide a census to the American people that is accurate and, therefore, fair, that enjoys broad support within the scientific and technical community; and that uses the taxpayer's money with restraint. This is a responsibility entrusted to me by the Congress through the Census Act and that I regard as second in importance to none.

Before I address how we intend to fulfill this obligation for the 2000 census, let me take a moment to review the 1990 census -- and the lessons learned.

The 1990 decennial census managed to account for 98.4 percent of the population. But it did not meet some basic goals. It was not accurate: its errors were skewed across the population, largely affecting minority groups. It was not timely: it took too long to get results and the delay contributed to the untenable situation in which two separate estimates of the population were in circulation. It took six years for that litigation to be decided by the Supreme Court. And, finally, it cost much too much.

In 1992, Congress mandated that the National Academy of Sciences-National Research Council undertake a decennial census study on how the government could achieve the most accurate, cost-effective, population count possible. After an exhaustive review, these organizations reached four central conclusions:

1) It is fruitless to continue trying to count every last person with traditional census methods of physical enumeration;

2) It is possible to improve the accuracy of the census count by supplementing a reduced intensity of traditional enumeration with statistical estimates of the number and characteristics of those not directly enumerated;

3) Once a decision is made to use statistical estimation for completing the count, a thorough review and reengineering of census procedures and operations could achieve substantial cost savings in the next census, even as accuracy is improved; And,

4) Proposals to drop the long form in the next decennial census and substitute a continuous month survey need substantial further research. Therefore, the 2000 census should include the long form.

We have used the recommendations embodied in this report as our guiding principles in designing Census 2000. Today, I would like to talk about two issues: sampling and management. Let me begin by saying that Census 2000 will make an unprecedented effort to count everyone living in the United States.

Like the eminent scientific and non-partisan government groups that have reviewed the question of sampling, we believe that the judicious use of statistical sampling is a prudent, cost-saving and legally permissible means for helping our nation achieve the best census ever.

Proponents include the National Academy of Sciences, the American Statistical Association, the Population Association of America, and the National Conference of Mayors, as well as such oversight bodies as our own Inspector General and the General Accounting Office.

Sampling provides us with a scientifically sound and widely accepted method for accounting for the people whom we cannot count. It is a widely understood and known process used in many research and business efforts in our economy.

Despite our plans to improve our address lists in conjunction with local governments, to have extensive outreach and promotion, and to use more convenient and user-friendly forms, there always will be some people we do not reach or find. Sampling allows us to impute characteristics of those people. Moreover, there always are some systematic errors in the census that can be corrected by using statistical sampling as a quality check.

It is important for me to note here that even with the use of sampling we will fulfill our commitment to have personal contact with at least 90 percent of the U.S. population. Let me repeat that, we are committed to contact personally at least 90 percent of the U.S. population.

That said, let me affirm that we believe that sampling is a known, mature and objective process that, if introduced into the 2000 census, will improve census accuracy while saving taxpayers hundreds of millions of dollars.

The second issue I would like to discuss is management. The census is a large and complex undertaking, the largest peacetime mobilization in our nation. To meet this responsibility, we are implementing a variety of innovative efforts. Among these are computers that scan forms to read handwritten responses. We also will be outsourcing parts of the effort that can be more effectively conducted by private partners. These changes should lead us to a census that is easier to implement and that costs less.

Bringing these efforts to fruition is a management challenge, and I am aware that concerns have been raised. We are working hard to address these concerns. And we are finalizing a census design that we will rehearse in 1998 and implement in 2000. I am confident, therefore, that the Bureau is on track to implement its plan and to provide the American people with a census as good as the one that the Constitution envisioned for us. I am confident, but not complacent.

I intend to take a strong and on-going interest in the management of this activity, and I look forward to working with this Committee and with the Congress to realize the goal we share of a census that is accurate, fair, cost-effective and that is worthy of its Constitutional origin.

Thank you for the opportunity to testify before your Committee. I will be happy now to answer any questions you may have.