Census mirror on Americans takes too close a look for some

Next Tuesday, "census day," America holds up a mirror and looks at the number , wealth, education, housing, jobs, and a dozen other vital statistics of its citizens.

The vaunted professional pollsters will step aside for the Bureau of the census's closer-to-real view of the country. The standard name-age-sex-race questions will be asked, along with others that are much nosier:

* "Do you enter your living quarters directly from the outside or through a common public hall; or through someone else's living quarters?"

* "Do you have complete plumbing facilities in your living quarters, that is, hot and cold piped water, a flush toilet, and a bathtub or shower?"

* "What is the value of this property?" "What is the monthly rent?"

One out of five Americans will be asked to bare still more details about his or her life. These results will then be used to draw inferences about the total population. Some of these questions:

* "How well does this person speak English?"

* "Does this person have a physical, mental, or other health condition which has lasted for six or more months . . . ?"

* "If this person is female, how many babies has she had, not counting stillbirths?"

Population and geographical distribution data are required under the US Constitution to apportion congressmen and other elected officials. The accompanying questions, however, are for social- planning reasons.

But are those questions about the quality of your bathroom, access to your house, your language capability, your mental or physical health, or number of babies, really necessary? Yes, says the Census Bureau, if this is to be an accurate inventory of America.

The Census Bureau points out that the bathroom question, which appeared on the 1970 version, is an attempt to measure housing quality. The access-to-your-house question is a way of making certain the respondent occupies a bona fidem dwelling unit, not just a few rooms in someone else's house.

Language ability is a way of finding how many Americans do not speak English and, therefore, where bilingual education or minority-aid should be emphasized.

The health question, the Census Bureau admits, "tested poorly" in trial runs, "but so many public agencies need disability data that the question will appear anyway." And the number of babies question, which is the same as one asked in 1970, is used to measure the fertility of American women and thus to predict the population growth rate.

All of the questions were derived during 70 public hearings over the past decade. Several congressmen looked warily at the intrusive questions during a 1977 examination, and enough round-about pressure was exerted on the Census Bureau to temper some of the questions and shorten the form. But, says one congressional staffer: "Congress took a basically benign attitude back at that time. We couldn't interest this institution because 1980 was so far away."

Michael J. Ferrell, staff director of the House subcommittee on census and population, says that while most of the 1980 questions resemble those of the past, "the census hasn't actually changed with the times. Public attitudes about privacy have changed considerably in the past decade. You can call it the post-Watergate, Proposition 13 mentality if you like, but people don't want their lives pried into."

Rep. William Lehman (D) of Florida, former chairman of the census subcommittee, says he is convinced the Census Bureau maintains enough safeguards to protect confidentiality. "But," he says, "there is no question that both census forms are self-defeating" due to the nature of the questions. "The perception of invasion of privacy is going to turn people off."

The two questionnaires, plus the follow-up activities of Census Bureau workers, are so broad in scope and so sweeping in coverage -- and so much is riding on the statistical results -- that any number of interest groups have logged complaints before the fact.

The fear of invasion of privacy concerns most of the citizens who have qualms about answering. Anti-draft protesters this week warned that, directly or indirectly, census data might be used to locate and prosecute young men of draft age who refuse to register.

But census officials say it will be virtually impossible to connect statistical data with personal names. In the 190-year history of the census, not only has the Census Bureau's constitutional directive to count the "whole number of persons" resident in the United States been upheld, but no court case has been won against the Census Bureau for wrongly divulging private information.

Census officials say this time a sophisticated optical scanner will read each census form. The device is prevented from reading the name at the top of the sheet. Then the form will be stored for 72 years. Only the Census Bureau, the person who filled out the form, or an heir or relative of the person who has shown a valid, legal reason why he should have access to it, are allowed to see it. The forms are not destroyed because of their future historical use.

"Our best defense," says Census Bureau director Vincent Barabba, "is to demonstrate what our record is. There has never been any evidence of unauthorized release of information."

His assurance aside, at least one group plans to protest the census by refusing to answer -- an act punishable by fines ranging from $100 to $500. Says Lawrence Samuels, director of the California-based Voluntary Census Committee: "We feel the census is too personal, too much like '1984,' and we object to its coercive nature."

Mr. Samuels's group is planning public protests, which will include the shredding of census forms, on April 1 and 5 in Los Angeles and elsewhere. About 3,000 persons took part in a similar protest in 1970, he says. This year, he predicts, "half a million will protest."

Not only are the questions too prying, says Jay Hilgartner of the Libertarian Party in Washington, but "the major objection is that the accumulation of data is used to maintain a very large government apparatus and to prop up this planning mentality." Libertarians support a voluntary census.

Others, too, have problems with the census. Minority groups say census workers are sure to count too few of them. Some governors and big-city mayors fret that their areas will be undercounted also.

Besides political representation, no fewer than 36 government programs -- from Headstart to federal aid for highways -- depend on census information to guide the allocation of funds to state and local communities.

In 1970, the Census Bureau admits, 2.5 percent of the population was not counted. This year, say some minority groups and politicians, the Census Bureau should adjust the figures upward, automatically, to take the error into account.

The opposite approach is sought by a group of senators, congressmen, and "immigration reformers." They say the large number of illegal aliens in the country will distort the census and give undue political representation to Southwestern states and other alien- intensive areas in the 1980s if the Census Bureau does not automatically deduct them.

The two viewpoints, says Mr. Barabba, can best be reduced to this argument: "Should we count only the people who are able to elect governmental representatives? Or should we measure the burden to be borne by an area which is to be represented?" Until now, Congress has chosen the second course.

Sen. Daniel P. Moynihan (D) of New York has introduced a bill to adjust the census statistics to compensate for undercounted minorities. And Sen. Walter Huddleston (D) of Kentucky has introduced a bill to adjust the statistics to eliminate illegal aliens from the count.

The Census Bureau is asking that it be allowed to conduct its count and determine the possible errors before deciding whether to make an adjustment.

This determination will be made, says Mr. Barabba, by taking a post-census census of certain regions to find out how close the count actually was. Census data also will be cross- referencing with already available vital statistics, such as social security, income tax, and birth/death data.

Then the political battle can begin.

By Jan. 1, 1981, after round-the-clock data processing by Census Bureau computers, the bureau must present the President with a count of the nation's population by state, for congressional apportionment. Three months later, more detailed population counts for counties, cities, and other political subdivisions must be given the governors of the 50 states.

All other data compiled are then stacked, shuffled, and massaged in many ways to provide planners, business persons, educators, political scientists, and anyone else with the closest thing to a mirror-image of the American people at the "census moment" -- midnight April 1, 1980.

Demographers can make their statistical adjustments for people overstating their education, understating their age, and for dissidents who have shredded their census forms. A credible statistic on almost any aspect of the country can be derived from the census.

The Census Bureau even has statistics that shows how inclined Americans are to become statistics on April 1.

As of March 21, says Mr. Barabba, more than 90 percent of the population had heard about the coming census. Of these, he says, more than 90 percent have indicated they will cooperate.

We want to hear, did we miss an angle we should have covered? Should we come back to this topic? Or just give us a rating for this story. We want to hear from you.