WASHINGTON--Average verbal scores on the Scholastic Aptitude Test dropped this year to an all-time low, and average mathematics scores declined for the first time since 1980, the College Board reported last week.
Announcing the results of the college-admissions test for the high school class of 1991, board officials also noted that the gap between high-performing and low-performing students appears to be widening.
Although the average scores dropped by two points each--to 422, out of a possible 800, on the verbal section, and to 474 out of 800 on the math section the scores of those who also took the board’s Achievement Tests remained 100 points higher on each part than those of all s.A.T. takers. The Achievement Tests are required for admission at many selective colleges and universities.
Such a gap suggests that there is a “disturbing pattern of educational disparity,” Donald M. Stewart, the board’s president, said at a press briefing here.
“If the dichotomy continues,” he said, “we could end up with a small class of educational elite and an underclass of far less prepared students.”
Mr. Stewart noted that the S.A.T. results also included some encouraging signs.
The percentage of test takers who were members of minority groups--28 percent--reached a record level, he noted, although the test scores for minorities showed mixed results.
In addition, officials pointed out, the number of students who said they took academic courses in high school continued to rise. Students in 1991 reported studying a mean of 18.7 course-years in such subjects, up from 18.2 in 1987.
Moreover, said Bill Honig, superintendent of public instruction in California, the overall averages mask genuine improvement in student performance on the test. A closer analysis of the results, he noted, showed that the proportion of the senior class who performed at high levels on the exam has risen steadily since 1983. Such increases have occurred, he pointed out, even while the number of students taking the S.A.T. has risen.
“If that’s our goal, if that’s performance, then schools are delivering,” Mr. Honig said. “The scores are going up.”
“Do I believe kids are scoring at the level they need to score? No,” he continued. “It’s progress, but not strong enough.”
Reversing a Decline
Developed and administered by the Educational Testing Service, the S.A.T. is primarily taken by students in the Northeast and the West Coast, and is used for admissions decisions by many of the most selective colleges and universities.
Results of the other major college admissions test, the American College Testing program, which is taken mostly by students in the Midwest and the South, are expected to be released this month.
The S.A.T. report found that last year’s drop in the number of test takers appears to have reversed.
In 1990, the number of students who took the s.A .T. declined by 6 percent over the previous year, prompting concerns that colleges may begin to see the effects of the decline in the college-age population.
But in 1991, the number of students who took the test--1,032,685-represented a slight increase over 1990. An estimated 42 percent of the class of 1991 took the test, the report notes. In addition, the report notes that about 200,000 students took one or more Achievement Tests in 1991, a 1 percent increase over the 1990 total. As with the S.A.T., the number of students who took such subject-matter tests had declined last year.
‘Free Fall’
In examining the results of the test, Mr. Stewart noted with alarm the decline in the verbal scores, the fifth straight year such performance has dropped.
“A free fall is taking place,” he said. “That’s disturbing.”
Mr. Stewart also pointed out that average scores on the Test of Standard Written English, a 30-minute multiple-choice test administered along with the S.A.T., reached their all-time lowest level in 1991. The average score on the T.S.W.E. dropped from 42.5 to 42.1 on a 20-to-60 scale, a decline over the previous low recorded in 1981.
The two results reflect the fact that students do not read much, Mr. Stewart suggested.
“We have a national problem of too much TV, too many videos, and a decreasing amount of time spent reading,” he said.
But the board president was at a loss to explain the drop in math scores, which have held steady for the past four years.
Lawrence W. Hecht, a senior research scientist for the College Board, cautioned that the one-year drop in math scores could be an aberration.
“A two-point decline over many years of stability could be a statistical blip,” he said. “It doesn’t necessarily signal a downward trend.”
But Mr. Stewart said that the decline comes as a surprise after years of efforts to improve math instruction in schools.
“Wouldn’t we have thought that, given all the emphasis on math, there would have been a blip up rather than a blip down?” he asked. “I was taken aback.”
Minority Participation
In addition to school factors, board officials noted that the changing composition of the test-taking population could also affect the average scores. A panel that investigated the steep drop in S.A.T. scores during the 1960’s and 1970’s found that about half the decline was related to the changing population.
Ethnic minorities constituted 28 percent of all students taking the test in 1991, up from 27 percent in 1990 and 13 percent in 1973, the board reported. Although white students’ average scores have been declining since 1976, the report notes, whites continue to outperform all ethnic groups on the verbal section, and all except Asian-Americans in math.
The report also states that the substantial gap between whites’ and blacks’ scores continued to narrow in 1991. Since 1976, one-fourth, or 64 points, of the 258-point difference between whites’ and blacks’ composite scores has been eliminated, it notes. Moreover, it says, the number of black students taking the test reached an all-time high of 100,209 in 1991.
The report also notes that a record 8 percent of students in 1991 said English was not their first language, and that another 8 percent said they were bilingual.
Such increases could help explain the decline in verbal scores, said Robert G. Cameron, a senior research associate at the College Board. Students who studied English as a second language had an average verbal score 77 points below the national average, he pointed out.
Course-Taking ‘Paradox’ Mr. Cameron also noted a “paradox” in the fact that average scores declined at a time when students were taking more academic courses. In general, students who take 20 or more academic courses in high school tend to score 50 points above the national average on each section of the S.A.T., while those who take 19 or fewer courses score below the national average.
Despite the increase in academic course-taking, however, Frederick H. Dietrich, vice president of the board for guidance, access, and assessment services, noted that the number of students who said they took four or more years of English in high school declined from 88 percent in 1987 to 85 percent last year.
The report also found that, as in past years, the average S.A.T. scores of students who said they planned to major in education in college lagged far behind the national average.
But the performance of such prospective teachers declined less sharply than that of others. Such students’ average verbal score remained stable, at 406, and their math scores declined by one point, to 441.
Moreover, the proportion of students who said they planned to major in education rose to 8 percent, up from 7 percent in 1990.
Copies of the College Board’s report, “College-Bound Seniors: 1991 Profile of S.A.T. and Achievement Test Takers,” are available free of charge from the College Board, Box AF, 45 Columbus Ave., New York, N.Y. 10023-6992.
