Cities Press Campaign to Fix Census Undercount : Population: Ten plaintiffs join lawsuit challenging tally of minorities and poor. Federal aid and political balance in Congress are at stake.

With new indications that the 1990 Census is undercounting the population by historic proportions, Los Angeles Mayor Tom Bradley on Monday kicked off a last-ditch effort by state and local officials around the country to compel the Bush Administration to change the final head count.

At a press conference, Bradley announced that nine more cities and one more county are joining a lawsuit filed by several other states and cities, including Los Angeles. The suit would force the Administration to adjust the census by using a controversial sampling technique that many experts believe is the best way to count urban poor people and minorities, the groups most often missed by the census.

In light of mounting evidence of a substantial undercount, Bradley accused the Department of Commerce, which presides over the census, of failing to live up to “its mandate to the citizens of this country” to count everybody. He said that the census undercount, which missed an estimated 4.6% of the Los Angeles population in 1980, “is sure to be at least as high” for 1990 if there is no adjustment.

“We can say quite confidently that this is going to be the worst census in history,” said Lance Simmens, assistant executive director of the U.S. Conference of Mayors, who attended the press conference. Because state and federal aid to cities is apportioned on the basis of population, a big undercount of urban residents “will have a devastating impact on cities at a time when they are already reeling” from cuts in federal aid, Simmens said.

Simmens described the press conference, one of three to be held this week in cities across the country, as the “last resort” in a two-year campaign to persuade the Administration to adjust the census using the sampling technique known as the post-enumeration survey.

The census is often a fractious affair. Dozens of lawsuits challenged the accuracy and fairness of the 1980 Census. None prevailed. The current dispute over the undercount has turned the 1990 Census into a partisan political fight with a Republican Administration under fire from urban Democrats.

For the cities and counties that are suing, the prize is not just money. Larger urban populations also beget more congressional districts. For a Republican White House, the possibility of more big city Democrats in Congress is not a welcome prospect.

Under a preliminary settlement of the lawsuit pressed by Los Angeles and about 20 other cities, counties, states and civil rights and civic organizations, the secretary of commerce has until July 15 to decide whether to use the post-enumeration survey, which has been completed, to adjust the census.

Commerce Department officials told Congress recently that they doubted that an internal review of the survey will be completed by July 15. If the work is not done by the deadline, the officials said, a decision probably would be made not to adjust the census.

The 1990 Census missed 4.7 million people, compared to 2.9 million left out in 1980, according to preliminary estimates by the Census Bureau. The census traditionally has missed a greater percentage of black people than white, and that gap is widening.

Current estimates say that the census will miss 1.2% of the total population while omitting 6.6% of the black population.

Demographers refer to the difference between the two rates as the differential undercount. That 5.4% differential is up from 3.9% in 1980. If the census is not adjusted and the current numbers do not change, it will be the first time in 50 years that the differential undercount has increased from one census to the next.

City officials said Monday that the suit on behalf of adjustment is being joined by San Francisco, Oakland, Inglewood and Pasadena, and by Cleveland, New Orleans, Philadelphia, San Antonio, Denver and Broward County, Fla. Besides Los Angeles, the original plaintiffs include New York City, Chicago, Houston, Dade County, Fla., and the states of California and New York. Also parties to the suit are the National Assn. for the Advancement of Colored People, League of United Latin American Citizens, U.S. Conference of Mayors and National League of Cities.

In pushing for adjustment of the census, the plaintiffs are telling the Administration to rely on the post-enumeration survey--a technique that relies on sampling and estimating the sizes of various population groups rather than counting heads.

The survey, conducted while the traditional head count was under way, involved re-interviewing the occupants of 165,000 households nationwide and comparing the results of those interviews to the initial head count.

If there is a difference in the findings--for example, if the census form reported five people living in the house and the post-enumeration survey reported seven--a determination is made as to which of the two is more likely to be accurate. If the survey is judged to be correct, the number of people in that household is “adjusted” upward.

The survey allows a similar adjustment to be made for all households belonging to the same stratum of society. One stratum includes single Latina women, ages 21 to 40, living in rental houses. The post-enumeration survey divides the population into more than 1,000 such strata.

The survey is controversial because it has the capacity of compounding a single error by a thousandfold. Once a mistake is made with one household, critics say, it can be factored into the formula for determining the number of people in all similar households.