Supreme Court to hear Census case

North Carolina, not Utah, got House seat by margin of 857

ELLYN FERGUSON GANNETT NEWS SERVICE

Originally published March 23, 2002

"We believe, under the Census count, we have been given the seat fair and square." -- Roy Cooper, North Carolina attorney general

WASHINGTON -- Brad Miller is one of nine candidates running for a new congressional seat that might disappear.

Utah officials are trying to convince the U.S. Supreme Court that North Carolina got the seat illegally. Utah will argue before the court Wednesday that the Census 2000 used a possibly unconstitutional statistical method for its count, which is used to redistribute House seats every 10 years.

The method is called hot-deck imputation. The Census Bureau uses it when census takers can't find anyone home after making up to six visits or phone calls and can't find anyone who knows about the hard-to-reach household. The bureau then uses data collected from nearby households to assume information about the hard-to-reach household -- such as the number of people who live there.

The method was used to determine less than 1 percent of all households in both states. But since Utah missed getting a fourth congressional seat by just 857 people, that fraction counts.

Utah's argument

Utah argues this guesswork is forbidden by the congressional Census Act, which bans statistical sampling in apportionment of House seats. The federal government and North Carolina argue the method, which has been used since 1960, is just a tool to help provide an accurate count.

Miller, a lawyer and North Carolina state senator, said he isn't worrying about the case. Instead, he's focused on getting endorsements and traveling the new Democratic-leaning 13th Congressional District based in the greater Raleigh area.

"What we're doing is going forward and not paying any attention to that," Miller said.

But other people are paying attention, including voting rights activists and civil rights groups.

Only the people of North Carolina and Utah would be immediately affected by a decision in the case.

But the case could also affect the balance of power in the House if the court ordered North Carolina, which has a delegation of five Democrats and seven Republicans, to immediately give the seat to Utah, whose delegation has two Republicans and one Democrat. Republicans control the House by 11 seats.

Other states affected

In the long run, other states could be affected in the 2010 census if the justices bar any form of imputation in arriving at state population figures used to determine how many House seats each state gets.

But even that effect might be limited because a congressional seat seldom turns on as slim a margin as in the Utah-North Carolina case.

Altogether, the Census Bureau added 1.2 million people, or four-tenths of a percent of the total U.S. population of 281.4 million used for reapportionment, from imputation.

Anita Hodgkiss, voting rights director for the Lawyers Committee for Civil Rights Under Law, said it is not clear how minority voters or minority candidates might be affected if the Supreme Court rules against the Census Bureau.

The Lawyers Committee, a Washington, D.C.-based nonprofit group that follows redistricting cases, and civil rights groups argued during the planning of Census 2000 that sampling would help offset the traditional undercount of minorities, the poor, and urban and rural residents.

"I think there's a question of whether civil rights groups will weigh in on this, whether it would help or hurt (the case) given the makeup of the court. The experience has been in the last few years that the court has been more hostile to minority voting rights cases," Hodgkiss said.

Congress has leeway

U.S. Solicitor General Theodore Olson and Walter Dellinger, a former acting solicitor general in the Clinton administration, are expected to argue that Congress has given the Census Bureau great leeway and that imputation is a way to cope with missing, incomplete or inconsistent data.

In papers filed with the court, North Carolina cites a prior Supreme Court ruling in a case stemming from the 1980 census in which Indiana claimed it had lost a seat to Florida because of imputation. Indiana and Florida later agreed that imputation was not the same as sampling, and the court agreed.

North Carolina Attorney General Roy Cooper was surprised the Supreme Court decided to take the case after Utah lost a different argument in a lower federal court. But he thinks it is unlikely Utah will win.

"We believe, under the Census count, we have been given the seat fair and square," Cooper said.

Seeking consistency

But Thomas Lee, the Brigham Young University professor and former Supreme Court clerk who will argue Utah's case, said the state is asking the justices to be consistent.

In 1999, the Supreme Court said the congressional Census Act forbids the use of statistical sampling to adjust population numbers used for apportioning the 435 seats in the House of Representatives among the states.

The court left the door open for statistical sampling on population figures used in funding formulas for federal programs or for redrawing boundaries for political districts.

Utah officials argue that using imputation to include people in apportionment is the same as using statistical sampling.

"This is the logical follow-up to that case," Lee said.

Lee also will cite the U.S. Constitution's directive to conduct "actual enumeration" for apportionment, meaning some form of direct counting.

Lee said the Census Bureau's use of mail-back questionnaires, which provide about 70 percent of all population information, qualifies as actual enumeration because only households that exist can return them. Imputation, he said, does not qualify because statistical assumptions may add households and people that do not exist.

The Olympian Copyright 2002

back to main Census 2000 index