[CTPP] C E N S U S 2 0 0 0 B U L L E T I N

********** C E N S U S 2 0 0 0 B U L L E T I N ********** Vol. 3 - No. 4 Feb. 25, 1999 The following statement on Census 2000 was read by Census Bureau Director Kenneth Prewitt at a news conference Wednesday (Feb. 24, 1999) at the National Press Club in Washington, D.C.: "I will make a few general observations and then take your questions. "I start with a word on recent history. The results of the 1990 census did not please the Census Bureau, or the Bush Administration, or the Congress, or governors, mayors, and other state and local officials, or a large number of private and public sector data users, or the American public. It was a costly census; it was less accurate than what the country has a right to expect. The Census Bureau was charged to design a more modern census, one that would reduce the number of Americans who are missed -- either because we cannot find them or because they won't cooperate. It did so. That design, however, quickly became mired in political disputes, was litigated, and a month ago was set aside by the Supreme Court. "The Census Bureau had, of course, planned for that possibility. It had presented an alternative design to the Administration and the Congress in mid-January, before the Court ruling. Based on our recently completed evaluation of our Dress Rehearsal experience, we have further refined that plan. Its principal features are the subject of this press conference. "The Dress Rehearsal tells us two things. "First, however hard we try and whatever the level of resources available, Census 2000 will not count everyone. Moreover, this 'undercount' will not be equally distributed across demographic groups. There is what we refer to as a differential undercount. For instance, in 1990 we counted nearly all white Americans, but only approximately 95 percent of African-Americans and Hispanics, and an even lower rate of Native American Indians. Insofar as these less well counted groups are concentrated in some states, not others; in some cities, not others; in some neighborhoods, not others -- these states, cities, neighborhoods do not get their fair share of either the political or economic benefits allocated on the basis of census numbers. "Second, the Census Bureau's design should include a procedure -- described in the updated summary as the Accuracy and Coverage Evaluation (ACE) -- that will identify the magnitude and distribution of the differential account, and correct for it. The Dress Rehearsal confirms the statistical soundness of this procedure. Consequently, I have today informed the Secretary of Commerce and the Congress that it is feasible for Census 2000 to include this procedure, and that by doing so we will produce a more accurate and complete census than would otherwise be the case. "Because the Supreme Court ruled that this more accurate number is not to be used for apportionment purposes, our design also includes a major, labor intensive (and expensive) effort to find and enumerate as many Americans as is humanly possible in the time-frame available. In pursuit of this goal, our first and most important effort is to put a census form in the hands of every single household in America. Census 2000 features many improvements and technical innovations not available in 1990 -- for example, a completely re-engineered Master Address File, the most comprehensive ever constructed in U.S. history; first-ever use of paid advertising; intensified partnerships with tens of thousands of local governments, tribal organizations, private groups and non-profit organizations nationally and locally, a census-in-the-schools initiative. "This plan is operationally robust, and will be conducted with complete dedication by the Census Bureau professionals. This said, the apportionment counts are not likely to be an improvement on the 1990 accuracy levels. How can this be? How can you spend more money, mount improved operations, and yet not increase accuracy? Because all the factors that made it difficult to count Americans in prior censuses are today even more present. In more American families, both parents work, making it difficult to find anyone at home. Transient lifestyles are on the rise. People are busy. More people live in irregular housing. Greater numbers of people are linguistically isolated. Large immigrant populations avoid government officials. Census forms must compete with huge flows of junk mail. More persons are cynical about -- or actively hostile to -- any of the works of government. Census 2000 must overcome decreased levels of civic engagement by the American people. In short, the Census Bureau has to work harder to stay in place. We will produce the best apportionment counts that we can; they will not include everyone. "Allow me to summarize the points just covered, so as to leave no ambiguity. Between the 1st of April and 31st of December, the Census Bureau will count (and assign to an address) everyone it possibly can. The results of this effort will meet our obligation to present apportionment counts without the use of modern statistical methods. But the work will not then be finished. Census 2000 will continue its work with an Accuracy and Coverage Evaluation in order to produce more complete and accurate numbers, which will be ready prior to April 1, 2001. It is the task of the Census Bureau top produce the best numbers possible, not to decide how they will be used. The more complete census counts will be made available in a form that allows them to be used, if it is so decided, for redistricting purposes, for determining the allocation of federal funds, and for ongoing statistical and program purposes. Some may describe this as a 'two-number census,' but it in fact is a census that is progressively more complete, more accurate. "I conclude by reminding us all that the census clock ticks -- relentlessly, ceaselessly. In just 372 days the first Census 2000 forms get delivered. Given the lateness of the hour, we must acknowledge the hard reality that we no longer have the luxury of debates about alternative designs, or substitute procedures. No matter how well intentioned, we cannot now take a chance on untested operations or late additions. The largest peacetime mobilization in U.S. history must go forward based on the considered professional judgment of the career scientific and operational experts at the Census Bureau, who stand with me here today. We are up to the task, but only if we are allowed to do the task." For further information about the Census 2000 Bulletins, contact J. Paul Wyatt in the Public Information Office on 301-457-3052 (fax: 301-457-3670; e-mail: pwyatt(a)census.gov).