An article yesterday on the importance of geometry for minority students incorrectly reported the odds of completing college without it. The odds are 1 in 40 for a black student and less than 1 in 60 for a Hispanic student. (Published 9/25/90)

Black and Hispanic students who take at least one year of high school geometry vastly improve their chances of getting into college and receiving a bachelor's degree, according to a study done for The College Board.

The central finding of the one-year study by Pelavin and Associates, a Washington research firm, suggests that requiring all students to take geometry is the most effective way to increase the number of minorities in college. Compared to white high school graduates, smaller percentages of black and Hispanic graduates entered college during the 1980s, even though minorities were a growing proportion of the traditional college-age population.

The study of almost 16,000 students found that the gaps between college-going rates of whites and minorities virtually disappeared among those who had taken a year or more of geometry, a mathematics course usually offered in the ninth or 10th grade. Typically, schools require geometry students to first take a year of algebra.

"We therefore recommend that schools consider the strategy of requiring mastery of algebra and geometry of all students and that schools develop a plan to encourage college aspirations in all students," authors Sol H. Pelavin and Michael Kane concluded.

"We don't think kids should just take geometry. We don't think they should just aspire to go to college. It's the combination of the two that gives you the big-bang effect," Pelavin said.

Donald M. Stewart, The College Board's president, called math "the gatekeeper for success in college" and recommended "serious consideration of a national policy to ensure that all students take algebra and geometry."

Education Department statistics on the high school class of 1982 used in the study indicate that geometry was taken by 40 percent of whites, 19 percent of blacks and 17 percent of Hispanics.

During the 1980s, some states increased the number of math credits required for high school graduation, but did not specify which courses should be taken. One response has been for students to take more courses in general or consumer math, courses that do not advance their knowledge of the subject.

The new study appears to corroborate a connection between studying math and college success found by other educators have. The Quality Education for Minorities Project has proposed that all students take algebra in middle school. Florida's chancellor of higher education has warned students not to expect admission to the state's colleges if they have not taken algebra.

Math courses were found to have a stronger relationship to college enrollment and completion than courses in laboratory sciences or foreign languages, which the study also examined.

"I think we're looking at something that is more basic than those other courses," Pelavin said. "The logical-thinking skills taught in algebra and geometry are some of the basic skills needed in college."

Pelavin and Kane found that, four years after receiving a high school diploma, 58 percent of whites, 47 percent of blacks and 45 percent of Hispanics were enrolled in college. But among students who had taken at least a year of geometry, the college attendance rates were nearly equal: 83 percent for whites, 82 percent for Hispanics and 80 percent for blacks.

There was a similar finding on college completion. Of high school graduates who enrolled directly in college, 33 percent of whites, 17 percent of blacks and 11 percent of Hispanics had either earned a bachelor's degree or accumulated enough credits to be seniors four years later. Among those who had studied geometry, 40 percent of whites, 34 percent of Hispanics and 27 percent of blacks had reached those milestones in that period.

The researchers put the odds of a black student finishing college without having taken algebra at 1 in 40, and at less than 1 in 60 for a Hispanic.