Adjusted census figures suggest a 2% population undercount in El Paso

Staff, Erica Molina
El Paso Times
December 6, 2002

An estimated 12,705 El Pasoans went uncounted in the 2000 U.S. census, according to controversial data released Thursday by the U.S. Census Bureau.

The nearly 2 percent difference between the official and adjusted counts for El Paso could mean that federal money that should come to the region is instead being distributed elsewhere.

"Unfortunately, for El Paso it means that there will be approximately 2 percent of the formula funding (from federal programs) that could potentially come here under normal circumstances that would be lost," said Tom Fullerton, an economics professor at the University of Texas at El Paso.

The Census Bureau's official count for El Paso County in the 2000 census was 679,622. But using controversial statistical adjusting techniques, Census Bureau data released Thursday estimated the county's 2000 population at 692,327 -- 12,705 more than the official figures.

Adjusted figures for Do?a Ana County estimated the 2000 population at 178,877, or 4,195 more than the official count. Nationally, the adjusted figures suggest that the 2000 population was almost 3.3 million -- or 1.2 percent -- more than the official count of 281.4 million.

Jesse Acosta, a county planner and former city demographer and Census Bureau specialist, said the difference in numbers is not very significant.

"If that's the case and that is the undercount, I believe the census did a fantastic job," Acosta said.

The Bush administration opposed releasing the adjusted numbers, on the grounds that the official head count was more accurate. Other officials, including big-city mayors, disagreed and won a court order forcing the Census Bureau to release the adjusted numbers.

Census figures are used for a wide range of purposes, including as distributing federal funds and determining the number of members each state gets in the U.S. House of Representatives. The adjusted figures released Thursday are unlikely to be used for any such official purposes, but they will probably renew the debate over the best way to count the nation's residents.

That debate has generally broken down along party lines -- Republicans favoring a strict head count and Democrats arguing for that count to be supplemented by statistical sampling methods, which could account for people missed or counted twice.

Rep. Dave Weldon, R-Fla., chairman of the House Civil Service Census Subcommittee, has opposed releasing data that he calls "flawed." He told USA Today on Thursday that their release will "serve no useful purpose whatsoever."

Demographers say the head count conducted every 10 years since 1790 is bound to miss large numbers of people, especially given today's highly mobile society.

"Undercounts happen because people are transient, people don't wish to be counted, and people are living in situations that are not exactly understood by the census in terms of the way they form the questions," said Cheryl Howard, an associate professor of sociology at UTEP.

The data collection methods also contribute to census inaccuracies, Howard said, including the use of different techniques to collect the data and the reliance on information from the U.S. Postal Service to provide current information.

Minorities, young males and children are traditionally those who are not counted, she said. El Paso has large numbers of each.

Rapidly growing areas are also subject to inaccuracies.

Acosta said the release of block-by-block figures of the census data differences will provide a better idea about how significantly El Paso would be impacted.

Bill Kaigh, a statistics professor at UTEP, said the American Statistical Association, of which he is a member, favors using adjusted census figures.

"From our perspective we realize errors are going to be made, and we think it's foolish not to adjust to them," he said.

Adjusted numbers

How adjusted census numbers released Thursday differ from the 2000 census counts:

Area 2000 census Adjusted count Difference % change from census

El Paso County 679,622 692,327 12,705 1.87 percent

Do?a Ana County 174,682 178,877 4,195 2.40 percent

Texas 20,851,820 21,215,852 364,032 1.75 percent

New Mexico 1,819,046 1,854,322 35,276 1.94 percent

United States 281,421,906 284,683,782 3,261,876 1.16 percent

Check out adjusted data for every state