Steady improvement by black students in reading, writing, arithmetic and other subjects has reduced the gap between blacks and whites on standardized school tests, according to research unveiled at the American Psychological Association's convention.

A research team headed by University of North Carolina psychologist Lyle Jones found "very clear evidence" that current black students are markedly closer to their white classmates in basic skills than earlier generations of blacks were. "The consistency of these trends . . . suggests that a further reduction in the black-white average difference for these test scores will be seen in the future," Jones said Monday in Anaheim, Calif., at the APA's annual meeting.

One striking example, Jones said, appeared in the verbal skills section of the National Assessment of Education Progress, a test given each year to a national sample of students aged 9, 13 and 17. When people born in 1953--today's 30-year-olds--took those tests, blacks averaged 20 points lower than their white contemporaries. Among students born in 1970--today's 13-year-olds--blacks' scores averaged 10 points below whites'.

Jones and his assistants were hesitant to list reasons for the blacks' comparative improvement. They suggested, though, that blacks' enhanced financial status and various public education endeavors, such as the federal Head Start program, may have contributed.

The study is based on work by Jones, Ernest Davenport, a graduate assistant at North Carolina, and Dr. Nancy Burton, now of the Educational Testing Service. It supports findings of a more limited report last fall by The College Board, which runs the national Scholastic Aptitude Test (SAT) for college applicants. The board found a narrowing in 1981 and 1982 tests of the traditional black-white gap in scores.

The Jones study is considerably broader. The North Carolina researchers reviewed 1972-78 scores on three national tests: the National Assessment, which is given to primary and secondary students; the SAT, which is taken by high school seniors, and the Graduate Record Exam, which is taken by college seniors and graduates.

Burton said yesterday that black performance vis-a-vis white performance generally improved on all three tests.

During the period studied, Burton said, blacks tended to enhance their scores faster than whites in areas of general improvement, such as basic reading and math. In areas where the total population's scores declined, such as general science, blacks' average scores fell less than whites'.

The researchers noted yesterday that their findings indicate a clear performance gap still exists between the average white student and his black counterpart on standardized tests.

"We should not lose sight of the continuing large achievement differences between white and black children," Jones said.

Nonetheless, Burton observed, the findings of the new study seem to establish more clearly than any previous work that the achievement gap is closing.

"Of course, we'd rather have five more years of data," she said. "But I think these results, which we find to be consistent regardless of student age or subject matter, indicate pretty strongly that something real is going on here."

In their comparative study, the North Carolina researchers used a narrow definition of "whites," which excluded orientals, Hispanics and American Indians.

However, Burton said, the reporting of racial information is inexact, and sometimes test administrators guessed students' race based on appearance.

The current gap between white and black performance varies somewhat with the test, the subject matter and the students' age, Jones said, so it is not possible to state the difference as a single figure. The key finding, Jones said, is that the gap seems to be narrowing for all tests, subjects and age groups.