April 6, 2001

Hispanic Leaders Fight to Correct Census Undercount

By Yvette tenBerge

Hispanic organizations and Latino leaders joined this week in condemning United States Secretary of Commerce Donald Evans for his decision to allow state governments to use unadjusted Census 2000 data in their redistricting processes, a move that could cost Californians billions of dollars in federal funds.

A report released by presidential members of the U.S. Census Monitoring Board indicated that last year's census may have left a total of 3.3 million people uncounted nation-wide and may have counted millions of other people, like college students living in dormitories, twice. The report also found California to be the most under-counted state, with 529,782 people - 1.5% of the population - not included. This data, along with their conclusion that most of those undercounted belong to minority, racial and ethnic groups, has Hispanic organizations demanding that the Census Board be allotted more time to continue their analysis efforts.

"There is much at stake for the Latino community politically and economically. The repercussions of not releasing data adjusted to correct the under-count will extend far beyond our political system," said Arturo Vargas, Executive Director of the National Association of Latino Elected and Appointed Officials Educational Fund (NALEO), last week before a U.S. Senate Committee that oversees the Census. In this lies the crux of the dispute for Hispanics. The information gathered by the census does far more than simply provide an accurate count of the number of people in the U.S. or an accurate description of where these people live and their ethnicity. What state and national governments then do with the census data is crucial, and it is this next step that has direct impact on each and every citizen.

"The Latino community should not have to live with a 10-year error because the Census Board had a 3-month deadline in which to decide whether to correct the Census 2000 data," said Mr. Vargas. The "3-month deadline" to which Mr. Vargas refers was April 1st, the date by which census numbers, by law, must be released for redistricting. The "10-year error" to which he refers comes in two parts. First, a national census is conducted only every ten years. Therefore, the data collected in 2000 will be used to set many federal standards until the year 2010.

Second, under public law 94-171, the government is allowed to use the data from each census to determine district boundaries, as well as the allotment of federal funds to those districts for the next ten years. Redistricting, the process of redrawing federal, state and local legislative districts according to population, also allows politicians to realign congressional and state legislative districts to assure "equal representation" for their constituents in compliance with the "one-person, one-vote" principle of the 1965 Voting Rights Act. Since the Census 2000 count is estimated to have missed over half a million Californians, as well as over 1 million Hispanics on a national level, the impact of using unadjusted Census 2000 data could be enormous. It is for this reason that politicians like Lt. Governor Cruz M. Bustamante are lining up to fight for the release of the adjusted numbers.

Bustamante who was appointed to the U.S. Census Monitoring Board by former President Bill Clinton, has called for the Bush Administration to release "corrected" data from the 2000 Census. "Despite a model Census outreach campaign, more than half a million Californians went uncounted. We have the science to correct the data, and not to do so, because of partisan politics, is an injustice to all Californians."

In the 1990s, in order to account for the large number of people that typically go uncounted in each census survey and to deal with those who are erroneously included, Former President George Bush commissioned the National Academy of Sciences to devise a method that would enable the Census Bureau to accurately adjust these numbers. To this end, scientists at the Census Bureau developed the Accuracy and Coverage Evaluation (A.C.E.) survey, a post-census survey of about 314,000 housing units designed to estimate the net coverage of the 2000 Census. The Census Bureau uses data from the A.C.E., which is based on housing unit matching, interviewing and person matching, combined with the actual census results and independent research to come up with "corrected data."

According to Howard Hogan, the Division Chief for the Decennial Statistical Studies Department at the Census Bureau, though, there was such a high degree of uncertainty surrounding the accuracy of the adjusted data, itself, that the Bureau's recommendation to Secretary of Commerce Evans was against incorporating these numbers unless further time was granted to analyze the data.

"The final decision was made by Secretary Evans, but we were asked to make a recommendation. The results of the census, together with the results of the A.C.E. and the results of independent benchmarks from other demographic analysis, essentially provided us with enough uncertainty about whether the adjustment was going to improve the census that we did not feel that it was prudent [to recommend the use of the adjusted numbers]," said Mr. Hogan. "For example, the A.C.E. is saying that the undercount is 1.5% while some of the independent research [based on birth, death and immigration data, also allowing for undocumented numbers] indicated that this number was much lower."

Although Mr. Hogan states that the Secretary of Com-merce's decision to base redistricting on the unadjusted numbers is as "final as anything gets in this world," he does admit that the Census Bureau will be allowed to continue to analyze their data. Using the unadjusted numbers for redistricting may be a done deal, but there are a number of important federal programs, like the community improvement program Community Development Block Grant, the educational program Chapter 1 Funding and Medicaid, that base their funding on the annual population estimates released by the Census Bureau.

Ed Schafer, the Senior Demographer at SANDAG, believes that the effect that the census undercount will have on the issue of federal funding is far more significant to San Diegans than the effect that it will have on redistricting. "Generally, the undercount is known to occur more heavily in democratic, urban areas. As far as redistricting is concerned, the districts could be drawn in such a way that one party might benefit more than another party, but this can not be guaranteed," says Mr. Schafer, who confirms that the distribution of federal funds are at least partly based on population.

"It is true that there is an annual estimate made and that some federal programs use population estimates to determine their funding, but what has not been decided [by the Secretary of Commerce] is whether the Census Bureau will be allowed to use the adjusted numbers as a base from which to begin making these annual estimates," says Mr. Schafer.

Although Hispanic organizations and Latino leaders may have lost the battle in extending the deadline for redistricting, they still have time to push Secretary of Commerce Evans to allow the Census Bureau enough time to accurately analyze their data. Billions of dollars hinge on whether the Census Bureau is allowed to use their adjusted numbers as a foundation from which they can build their annual population estimates.

Return to the Frontpage