The U.S. Census Bureau is coming to town a few years early in 21 communities in southeast Los Angeles County and the San Gabriel Valley--communities that, because of their high minority populations and housing density, have been tough to poll in the last two censuses.
Although census takers will not set up shop until October and are to begin mailing questionnaires in March, 1986, bureau officials already have met with local officials to coordinate the census.
Dubbed the Central Los Angeles County Pretest, the experimental census will try to determine the best ways of counting not only Los Angeles County residents but urban dwellers and members of minority groups in general.
Making an accurate count is imperative, because census data determines representation in Congress, and federal, state and county agencies also use the data as one basis for distributing aid money.
It is more difficult for the Census Bureau to count minorities than it is to count the general population, bureau officials said, and the problem is compounded in urban areas.
But even if the census found these missing people, experts interviewed differ on what impact a more accurate census would have on how government aid is distributed, or how it might affect lawmakers who represent the test area in Washington or Sacramento.
Bureau officials chose Los Angeles County as a test area for 1986 because it has a high minority population and is urban and densely populated, said Peter Bounpane, assistant director for demographic censuses at the bureau's headquarters in Maryland.
Despite difficulties in finding some people, under-counting is not widespread. The Census Bureau estimates it misses only one-half of 1% of the total American population every 10 years.
"The problem is that even though it's a small total of the population, it represents the minority population," said Mike Flanagan, census coordinator at the bureau's Los Angeles office.
In fact, the bureau estimates that in the 1980 Census, it missed 2% to 4% of the minority population alone. For Latinos, the estimated "miss rate" in 1980--the first time the bureau counted Latinos as a subgroup--was 4.5%; for blacks, it was 5.5%, lower than the black miss rate in the 1970 Census.
"The guess is that there are people in the minority population who for one reason or another fear putting their name on the Census form," Bounpane said.
Fear of Discovery
He said some members of minority groups may fear losing welfare benefits, or, especially among Latinos, may fear that the information will be turned over to the U.S. Immigration and Naturalization Service in its search for illegal aliens.
"All these are unfounded, but people do fear that," he said.
Even legal or naturalized residents sometimes refuse to return the census form, said John Huerta, associate counsel for the Mexican-American Legal Defense and Education Fund.
"They have so much at stake in staying in this country, they just don't want to rock the boat," he said, adding legal residents may be fearful because, although they are in the country legally, they still can be deported.
Housing conditions in densely populated urban areas also make polling difficult. Multiple families packed into a dwelling often try to conceal themselves because they fear they are violating zoning or health codes.
But these people, as well as illegal aliens, have nothing to fear, bureau officials said, because the bureau does not release information on individuals, only on averages for each census tract.
Tries to Overcome Concern
The fears exist, though, and the bureau, one official said, is "constantly trying to overcome them." One of the purposes of the pretest is to convince people to trust the census.
The actual nationwide Census will not be conducted until 1990. A census is held every 10 years, as mandated by the Constitution.
Censuses are conducted by first determining an area's addresses, which are plotted into census tracts. Questionnaires are mailed to the occupant of each address. Residents send the questionnaires back to the bureau office. The bureau then sends pollsters to the addresses that did not respond.
The experimental census will test polling procedures. To establish itself in the test communities, the bureau will open three offices in the area. Instead of trying to get the word out themselves, and to build more trust between the bureau and the community, census takers will ask community groups and leaders to encourage residents to respond to the census.
"Almost the entire work force will be local people," Flanagan said.
To cover part of the cost of the pretest, the bureau will spend about $1 million on salaries, services and supplies for the local offices, and expects to hire more than 1,000 workers, census officials said. No census official could give a complete cost for the pretest.