Census Count Low but Reliable, Officials Say : Population: State’s estimated 1.1-million uncounted residents won’t be added until the figure can be proved more reliable than original tally, spokesman says.

Census Bureau officials acknowledged Thursday that the 1990 census undercounted the nation’s population by at least 5 million people, and undercounted California’s by more than 1.1 million, but they refused to embrace the new estimates as a more accurate count without additional study.

The new Post Enumeration Survey (PES) figures, based on a sampling survey and adjusted to correct for an undercount of people believed to have been missed in 1990, would increase California’s population to 30,888,000, or 3.7% more than the 29,760,021 figure tallied last year.

Nationally, the bureau estimates that there are 253,978,000 residents, up 2.1% from the figures compiled last year from the once-a-decade head count. The 1990 census counted the nation’s population at 248,709,873.

Census data is used for reapportioning the 435 seats in the U.S. House of Representatives and the seats in state legislatures or in city and county government bodies. In addition, population-based formulas are used widely to allocate federal revenues to state and local governments.

If the new figures were used to calculate congressional seats, California and Arizona would gain another representative in the House, while Pennsylvania and Wisconsin would each lose one, according to a study by Election Data Services, a Washington-based political consulting firm.

“It’s fair to say we missed people,” said Peter Bounpane, assistant director of the bureau. “We have no argument with that.”

However, he cautioned that the bureau would suggest using the new figures only if officials are convinced that the statistical count has fewer errors than the 1990 census did.

“As of today, we are still in the process of review,” Bounpane said. “Of course, some people will benefit more than others if we use the new figures. But that’s not our purpose in making our decision. The only issue we’re concerned with is whether doing this is more accurate than using the census numbers alone.”

Adjustment advocates were elated, arguing that the new numbers proved their case.

Los Angeles Mayor Tom Bradley said “preserving the myth that the uncorrected census accurately portrays our population serves no valid purpose.” He called for a recount, saying “failure to do so threatens the integrity of our system of government, which depends upon one person, one vote. . . .”

City Atty. James K. Hahn concurred, saying that the new estimates “show a significant undercount of minorities, which impacts Los Angeles and California even more than other areas of the country because we live in the most racially diverse state in the nation.”

Settlement of a suit brought by officials in Los Angeles and other big cities requires Commerce Secretary Robert A. Mosbacher to decide by July 15 whether the census figures should be adjusted for a possible undercount. His decision will be based, in part, on the recommendation from bureau officials who have been ordered to conduct a statistical survey of the national population and compare the results to the traditional count.

Rep. Thomas C. Sawyer (D-Ohio), chairman of the House census and population subcommittee of the Post Office and Civil Service Committee, said the new figures support the lawsuit.

“It’s becoming clearer and clearer that the census was flawed,” he said. “The PES may not be the most accurate count, but it’s clear the raw numbers are dead wrong. The secretary of commerce has a very uphill battle to prove why on scientific grounds it doesn’t make sense to adjust.”

But states that stand to lose representation say that they will not take the new count without a fight.

“An adjusted census could very well be less accurate and less objective than the original census,” said Sen. Herbert Kohl (D-Wis.), chairman of the Senate subcommittee that oversees census operations.

The PES estimates indicate that people were missed in virtually every state and that most of the nation’s largest cities were undercounted in 1990, including a 5.1% undercount in Los Angeles. The estimates showed that the city has about 3.7 million residents instead of the 3.5 million counted in the census.

In New York, about 3% of the population was overlooked and 2.6% was missed in Chicago.

But the most striking finding was that Inglewood was the nation’s most undercounted city, with 10.9% of its population missed. The largely black and Latino suburb of Los Angeles was the only place in the nation with a double-digit undercount.

“We expected there would be an undercount,” said Norman Cravens, Inglewood’s assistant city manager. “But I’m shocked that it’s that high.”