'HIDDEN HOMELESS' MAY ELUDE CENSUS

SPECIAL COUNT MARCH 20-21 WILL BE ONE OF BUREAU'S TOUGHEST

The Census Bureau, the Roman Legion of population measurement, is about to take on one of the toughest tasks in American number-counting: the question of how many homeless people live in the United States.

A bureau count of the homeless is scheduled for March 20-21, 1990 -- shortly before the census of the whole population on April 1.

State and local governments urged the special count, reflecting their desire to know more about the distribution and characteristics of the homeless: Are they mainly disturbed people? Do they include families who are employed but homeless because little low-rent housing has been built since President Ronald Reagan drastically slashed federal funds for new construction?

The count may help determine policies the nation could adopt to deal with homelessness.

Despite the massive effort, the homeless census will not be perfect. Cynthia Taeuber, the bureau's specialist on the homeless, said there are many "hidden homeless" who cannot be located because they are not at the shelters and abandoned buildings known to be used by the homeless. But she said the bureau will be making the best effort possible.

Advocates for the homeless, while generally welcoming the count, have their concerns about how it might affect the national debate. "The danger is that when the Census attempts its count, it will be misconstrued as a complete count," said Maria Foscarinis, Washington counsel for the National Coalition for the Homeless.

"It's virtually impossible to get any accurate count of homeless people in large part because the most significant group now consists of people who don't fit the stereotype," she said. "They are people who don't look homeless, who stay in abandoned buildings and are asleep at 4 to 6:30 and won't come out for the Census Bureau to count, and people who are doubled up, tripled up with friends and relatives and who can't stay in any one place for too long. So there are a large number of people who won't be picked up by any count."

The debate and confusion over the size of the homeless population heated up in 1984 when the Department of Housing and Urban Development estimated the homeless at 250,000 to 350,000. Advocacy groups charged that HUD underestimated the number to reduce pressure on Reagan to spend money on the homeless.

Two years later, a Harvard economist, Prof. Richard Freeman, estimated 350,000.

Last year, the Urban Institute put the number at about 600,000. At that time, Foscarinis called the institute's figure too low and said, "Our estimate is 3 million."

As Mitch Snyder of the Community for Creative Non-Violence here put it, "Just about everybody in America who is involved thinks the figure is 2 million to 3 million."

In the bureau's nationwide count of all the homeless who can be found, about 9,000 workers will be employed. Taeuber said that March 20-21 was chosen because "many of the state and local shelters close their doors on April 1" as cold weather ends.

The project begins in September, Taeuber said, when the bureau will send letters "to 39,000 local government units asking them to identify all the shelters for the homeless -- permanent and temporary, public and private -- that they know about, plus any nighttime sites, such as subway entrances and abandoned and boarded-up buildings where the homeless are known to congregate."

On March 20 from 6 p.m. to midnight, bureau enumerators will fan out to an estimated 5,400 shelters and count the homeless there.

From 2 a.m. to 4 a.m. March 21, enumerators will visit street locations such as subway entrances and attempt to count the people there. And from 4 a.m. to 6:30 a.m., they will station themselves outside abandoned buildings and count the people who emerge after having slept the night.

Because of possible physical danger, the enumerators will not be asked to enter abandoned buildings but to remain in cars. Police will not be used to escort the enumerators into the buildings because people inside might fear arrest and because the information gathered is considered confidential.

Taeuber said some groups, not at the locations to be visited March 20-21, will not be counted. For example, "We are not counting people riding on buses and in subway cars because there is no objective way for the enumerator to know which people on the bus are homeless.

"And then there are the hidden homeless -- nobody knows where they are. Some hide in tunnels, under bridges, in holes in the ground, abandoned cars, haystacks. We just don't know where they go."

On April 1, during the full decennial census, enumerators will also visit shelters for abused women, commercial campgrounds, drug and alcohol treatment centers and maternity homes for unmarried mothers and identify those who report no other usual residence.

Because of the "hidden homeless" problem, the Census Bureau will not report a total figure. It will list separately the homeless counted March 20-21 and the tallies for each of the separate places visited April 1. Later, the bureau will report its findings on income, race, sex, age, and social and economic characteristics.

Articles appear as they were originally printed in The Washington Post and may not include subsequent corrections.