Estimates from census clear hurdle

The Register-Guard
November 23, 2002

The U.S. Justice Department won't appeal a ruling ordering the U.S. Census Bureau to release figures estimating how many people were missed in the 2000 population count, a decision that could affect how billions of federal dollars are distributed.

Justice lawyers had until Friday to appeal last month's decision by a three-judge panel of the 9th U.S. Circuit Court of Appeals in San Francisco. The judges said the public is entitled to see Census Bureau figures adjusted by statistical sampling.

``We are not filing anything,'' Charles Miller, spokesman for the Justice Department's civil division, said hours before the midnight deadline. The Census Bureau had no immediate comment.

Oregon state Sens. Susan Castillo of Eugene and Margaret Carter of Portland filed the federal lawsuit that was ruled on last month.

Castillo was elected state school superintendent in the May 21 primary election and will take office in January. She couldn't be reached for comment Friday night but, informed of the appeals court ruling last month, said the census numbers dispute was a "credibility issue. When we have a high level of confidence in census information, the more people are willing to take part."

Democrats, big-city politicians and civil rights groups have charged that the 2000 census missed roughly 3.2 million people - most of them minorities and the poor - and that many communities are being shortchanged in receiving government funding distributed by population. The funding helps pay for Medicaid, foster care and other social service programs.

``People would like to see this data,'' said Tom Sussman, Castillo's and Carter's attorney. ``We don't know if it's helpful or harmful or not, but you've got to see it first.''

The population count is taken every 10 years. The bureau sends census takers and questionnaires to every household, though not everyone responds. The bureau tabulates the results and sends them to federal social service agencies, which use them to determine how much money goes to each state.

The census also is used to reallocate House seats based on population changes. A 1999 U.S. Supreme Court ruling bars the use of adjusted numbers for reapportioning congressional seats.

After the 2000 count, the Census Bureau used mathematical formulas to estimate how many people were missed, a population termed the ``undercount.''

Opponents of releasing the adjusted data, mainly Republicans, have said the complicated statistical methods used to determine the undercount would add more error into a census that the bureau deemed to have one of the lowest national undercount rates ever.

Critics also have said that while adjustments count missed people, they may not allocate them to the proper neighborhoods because the formula is less accurate on the local level.

The Bush administration in October 2001 backed the bureau's decision not to release sampling data. Castillo and Carter, both Democrats, filed a Freedom of Information Act request to see the adjusted figures and the bureau asked for an exception to the law.

A month later, U.S. District Judge James Redden of Portland ordered the government to release the undercount.

Last month, the appellate court sided with Redden, rejecting the bureau's arguments that releasing the data would expose sensitive internal debates and have a ``chilling effect'' on future policy discussions at the agency.

Regardless of what the bureau does, the amount of money distributed to each state won't change. However, state and local governments have the option, if their laws allow, to use the adjusted numbers to redraw state and municipal political districts and reconfigure how federal dollars are disbursed to the local social service programs.

``Now that these data will be made available, the public and scientists can finally judge for themselves which data was more accurate,'' said Rep. Carolyn Maloney, D-N.Y.