Maryland Transportation Secretary Richard H. Trainor objects to dropping census questions about how people get to work and how long it takes them because he says the change would rob the state of information "used extensively in support of highway and transit planning activities in the Baltimore and Washington metropolitan areas."

Executive Director Robert E. (Jack) David of the South Carolina Employment Security Commission said the questions provide information on "important employment characteristics." And Alderman Norris Larson of the 8th Ward of Evanston, Ill., said the questions provide "block by block data" deemed "critical" to determining housing needs and fair-housing issues.

Economic research chief H.N. Goodson of the Texas Employment Commission said the census questions are "vital to the ongoing Labor Market Information programs" of the commission. Without them, he wrote, the capacity to analyze and monitor labor conditions, "especially at the substate level," would be "seriously compromised."

The questions in question are traditionally asked by the U.S. Census Bureau. And since the Office of Management and Budget suggested cutting back on the number asked in the 1990 census, the protest letters have been rolling in.

Under the Census Bureau's tentative plan for the April 1, 1990, poll, five-sixths of the 106 million dwelling units in the United States will get a "short form" with 17 basic questions, and the other one-sixth will receive a longer form -- the basic 17 questions plus about 50 others.

But the forms, which will be used in a March 20, 1988, tryout in three locales and then in the 1990 head count, must be approved by OMB under the Paper Work Reduction Act.

OMB, looking at the forms from the point of view of simplification and reduction of the paper work burden on the public, has questioned the necessity of up to 30 of the questions -- two-thirds on housing -- and has asked the Census Bureau whether the same information could be obtained by other, less extensive sampling.

OMB's Wendy Lee Gramm has emphasized that her office has not formally proposed dropping any of these questions or made any decisions on whether to do so, only asked for the justification for including them. OMB will notify the Census Bureau by next Tuesday of its decision on which questions should be dropped or changed.

Until then, OMB is receiving additional material from the Census Bureau to justify the questions, and it has asked for public comment from state and local government groups, businesses and academic centers that use census data.

More than 600 letters have come in so far, and judging from about 200 viewed at random, virtually all of them are strongly opposed to deletion of any of the 30 questions, particularly those on transportation, the respondent's residence five years earlier, fertility of women and job experience. A great many also urge retention of all housing questions.

The letters typically make the same point: that while general regional and national information can be gleaned from less extensive sample surveys, detailed information about specific neighborhoods, cities, towns -- needed for policy and economic decisions at the local level -- is simply not available anywhere else but through the decennial census.

The "public" response, while expected, has led some officials involved in the matter to speculate that OMB will eventually approve the questions, though it might force a few of the housing questions to be dropped and others to be shortened or combined. So far, the OMB is keeping mum.

The letters, meanwhile, cover the American landscape. The Las Vegas Regional Transportation Commission needs the questions for a "transportation planning program in southern Nevada." In Kentucky, the questions help provide "benchmarks . . . for nine service delivery areas plus our affirmative action and estimated labor supply reports."

The Kaw Tribe of Kaw City, Okla., called housing and labor questions "vital to the funding process of the Kaw Tribe." And Wilma P. Mankiller, principal chief of the Cherokee Nation of Tahlequah, Okla., said deletion of various labor and housing questions would "seriously limit our ability to evaluate the progress and status of over 100,000 Indians in our area."

And the University of California at Davis wrote that the census information provides the "only subnational level data we have been able to locate which comes close to measuring" an issue under study by one department -- "the displaced farmer."