The Maryland Department of Education has scrapped the results of its functional reading competency test, given last fall to about 61,000 ninth graders across the state, because officials found at least seven of the questions were too ambiguous to be fair.

The test, which cost about $80,000 to administer, must be passed by all students in the state before they can graduate from public high schools. It includes questions based on street signs, labels and other items students can expect to encounter in practical situations.

"We blew it," said William D. Grant, chief of program assessment for the state department who was in charge of preparing the new test. "But we're willing to admit that, and we're going to fix it up."

"Good, well-developed competency tests can do a lot for kids and teachers," Grant added. "It's a little more difficult (to write them) than we thought it would be."

In a memo released yesterday, David W. Hornbeck, the state school superintendent, said the fall test results would not be counted because at least seven of its 72 questions were "technically defective."

"To do otherwise would possibly penalize students through no fault of their own," Hornbeck said.

The affected students will take a different version of the 90-minute test next fall when they are 10th graders. Those who fail will be given special remidial work, Hornbeck said, and will have three other chances to try to pass before graduation.

The problems with this year's version of the reading test are not the first that the state minimum competency program has encountered in Maryland.

After the test was given in the fall of 1978, officials discovered that the Baltimore school system had distributed "study guides" beforehand that included two of the actual questions and several others that were very close.

Grant said the test results were invalidated in Baltimore, but were counted elsewhere in the state. He said the department then speeded up development of the alternate version of the test, which included the ambiguous questions.

The new test was given first in Baltimore last spring, Grant said, but the problems with it were not detected then.

"It was only after some of the results were in from all over the state that we realized we could have confused the kids," Grant said. "The kids could be right."

One sign that "something might be funny" with the test, Grant said, was that large groups of students picked the same wrong answer on several of the multiple choice questions. "That showed there might be another plausible way of answering the question beside the one we were looking for," he said.

Grant refused to disclose the wording of the ambiguous questions because he said revised versions of them might be used again. But he gave the following sample of the sort of confusion involved.

Question: If a ship is wrecked in very deep water, how far will it sink?

A. Just under the surface.

B. To the bottom.

C. Until the pressure is equal to its weight.

D. To a depth which depends in part on the amount of air it contains.

Grant said the correct answer is B, but he said a bright student might pick D because the ship might remain only partially submerged.

Including the special spring test in Baltimore, Grant said about 78 percent of the state's ninth graders passed the reading test last year. The 22 percent who failed were allowed to take it again this fall. But Grant said the state received no information on the number who repeated the exam, which has now been discarded.

He said these students, currently 10 th graders, will be allowed to take the test again in 11th and 12th grades. The reading test goes into effect as a statewide requirement, starting with this class, which is scheduled to graduate in June 1982.

Even though Maryland has reported no statewide results by race, officials in Montgomery County disclosed that among their students last year 28 percent of the black failed compared to just 7 percent of the whites. Montgomery School Superintendent Edward Andrew said he is forming a committee to check the test for possible bias.

Grant said the group of officials who developed and reviewed the test including many blacks. "We are satisfied that at least on the surface, the test does not have any race bias," Grant said.