State and Testing Officials Dispute Gloomy View of Pupil Achievement
By Robert Rothman
State officials and testing experts last week took issue with Secretary of Education Lauro F. Cavazos' gloomy assessment of the status of student achievement.
In releasing the annual state education-performance chart this month, Mr. Cavazos cited the fact that overall scores on college-admission tests remained stable for the fourth straight year as evidence that achievement has "reached a plateau." He also charged that the school-reform effort thus far has failed to boost performance.
But the state and testing officials argued in interviews that the Scholastic Aptitude Test and the American College Testing Program test are inappropriate gauges of student performance.
Moreover, they said, alternative interpretations of these data, as well as other performance data not included on the "wall chart," suggest that achievement has improved over the past few years.
In particular, the officials noted:
- College-admission test scores have increased for every demographic group. But overall scores remained stable because lower-performing groups took the tests in greater numbers.
- Results from the National Assessment of Educational Progress indicate that performance on lower-level items has increased steadily over the past decade, and that the gap between blacks' and whites' performance in most subjects has narrowed considerably.
- Publishers of commercially available achievement tests report that national norms now are substantially higher than they were earlier in the decade, particularly in mathematics. This phenomenon, they noted, helps explain why most students perform "above average" on the tests.
These trends suggest that, far from failing, school reforms have succeeded in raising the basic skills of most students, especially those who were low-achieving, said Archie E. Lapointe, naep's executive director.
"Schools delivered on what we asked of them," he maintained. "They went back to the basics, and worked on minimum competencies. That mission is largely accomplished."
But Charles E.M. Kolb, deputy undersecretary of education, said that despite these gains, overall achievement remains too low.
"We never said the wall chart is the sole set of indicators," he said. "But if you look consistently, the trend is clear: There is a lot of room for improvement."
Ramsay W. Selden, director of the state education-assessment center of the Council of Chief State School Officers, cautioned that it is still premature to make judgments about the reforms instituted in the mid-1980s. Most of the students whose performance shows up on national indicators, he noted, started school before the reforms went into effect.
"We're still not seeing kids who have been through 12 years of schooling under the reforms," he pointed out. "We may see an upturn as kids come out" of school in five or six years.
Ever since it was launched in 1984 by then-Secretary of Education Terrel H. Bell, the wall chart has sparked a debate among educators over the quality of data on student performance.
State officials charged that comparisons based on college-admission test scores were unfair. Such tests, they pointed out, measure the performance of a self-selected sample of college-bound high-school students, and do not evaluate the full range of the school curriculum.
The wall chart, said Gregory R. Anrig, president of the Educational Testing Service, which administers the SAT, "attempts to draw sweeping conclusions about the quality of education in the U.S. by taking a snapshot at one point in time with instruments that were not designed for that purpose."
As a result, said Daniel Koretz, senior social scientist with the rand Corporation, Secretary Cavazos' conclusion may be correct, but does not follow from the data.
"I think it's probably the case that the reforms of the last decade have not brought about dramatic improvements," he said. But "to infer that from the SAT scores is unjustified."
Mr. Kolb acknowledged that the admission tests do not reflect performance of a national sample of students, but he added that naep reading scores, which are based on a representative sample, show a similar pattern of stagnation.
Bill Honig, California's superintendent of public instruction, said a closer look at college-admission test scores reveals evidence of improvements in student achievement.
Nationally, he pointed out, scores for minority students have risen much faster than those for white students. But the overall scores have remained relatively stable because the minorities, who tend to be lower-performing, made up a growing proportion of the test-taking population.
Between 1984 and 1989, for example, blacks' SAT scores rose by 22 points, to 737 out of a possible 1,600, while overall scores rose by only 6 points, to 903. On the act, blacks' scores rose by 1 point from 1985 to 1989, to 13.6 out of a possible 35, while the national average remained stable at 18.6 over that period.
"If you have more minorities [taking the test], you expect scores to go down," Mr. Honig said. "You reach deeper into the talent pool."
In fact, he said, "all groups went up. Each group is making progress, but the mix is changing. You have to swim faster just to stay even.''
Mr. Anrig added that the increase in minority test-taking, and the substantial rise in their scores, is an unheralded accomplishment that schools should celebrate.
"There ought to be the equivalent of A Nation at Risk proclaiming improvement in minority scores," he said. "These are youngsters who, in the past, were lower performers. That's lost in the wall chart."
But Chester E. Finn Jr., a professor of education and public policy at Vanderbilt University, said educators have for years cited changes in the test-taking population as an explanation for the decline in admission-test scores.
"It was partly true then, and it's partly true now," he said. "I don't think it explains away the absence of progress. Other data don't give me grounds to say things are conspicuously better."
Good News Is Hard To Find
But Mr. Honig and others argued that data from other testing programs, including naep and state tests, also suggest that student achievement has improved.
Although reading scores remain stable, the California superintendent noted, an unpublicized NAEP report shows that 17-year-olds' math performance increased substantially between 1982 and 1988. Nearly all high-school students, he said, demonstrated basic skills and beginning problem-solving abilities, and 60 percent could perform "moderately complex'' procedures and reasoning.
"Can the nation survive if only 60 percent can solve moderately complex problems?" he asked. "We probably need 75 percent at that level. But we're making good progress."
Mr. Lapointe of naep said that the 1988 test results are "less reliable" than other naep data, since theywere not based on a representative sample of students. But he confirmed that, between 1982 and 1986, 9- and 17-year-olds showed progress on the math tests, and that black youngsters at all ages showed "steady, significant growth" in the subject over the period.
Mark D. Musick, a member of the National Assessment Governing Board, pointed out, however, that blacks' scores still lag behind those of whites, and that overall performance remains unacceptably low.
"When we were preparing press releases [to accompany naep reports],'' said Mr. Musick, president of the Southern Regional Education Board, "we looked for positive points, but they were hard to find."
Robert L. Linn, a professor of education at the University of Colorado, also noted that commercial achievement-test publishers have also reported gains in student performance.
In testing samples of students to develop national norms, he said, publishers have found that students now perform better than those in the past.
"Higher achievement is needed to be 'average' in 1988 than in 1981, when the test was normed the last time," he said. "That suggests something of an increase."
Paul L. Williams, director of re and measurement for ctb/McGraw-Hill, which publishes the California Achievement Test and the Comprehensive Test of Basic Skills, confirmed that norms are now higher for math, but added that reading performance has remained stable.
Mr. Honig explained that reading performance has been slower to increase because of the pervasive damaging effects on performance of television, and because the growth in immigration has increased the number of students with limited English skills.
The steeper increase in math performance, he added, might also reflect the national attention to that subject in the wake of concern over economic competitiveness.
Critics, such as the physician John J. Cannell, who drew national attention with his revelation that most elementary students score above average on achievement tests, have suggested that much of the increase in performance is due to "teaching to the test" or outright cheating.
But Mr. Selden of the CCSSO disputed that view.
"Some increase is due to practice and teaching skills measured on the test," he said. "But it's also due to the general increase in student performance."
Mr. Linn said the tests measure a relatively narrow range of knowledge and skills. Better methods of assessment, he said, are needed to determine precisely how much students know and can do.
"These are gains in achievement measured by tests, not gains in more broadly defined knowledge and skills," he said.
In addition to citing gains in achievement, critics last week also challenged Mr. Cavazos' assessment of the reform movement.
'Throwing Cold Water'
At the press conference at which he released the wall chart, the Secretary pointed to a chart that showed that performance remained flat at a time when spending on education increased substantially.
Edward D. Roeber, supervisor of the Michigan Education Assessment Program, said there is no direct link between the reforms that have been enacted and student performance.
"To expect student-achievement improvement [from the reforms] is unrealistic," he said. "Such reforms set the stage for improvement, they don't cause it to occur. Paying teachers 10 percent more is not going to get 10 percent more in achievement."
Frank Newman, president of the Education Commission of the States, said the gains among low-achieving students suggest that the "reform movement to date hasn't been a failure," but that "it hasn't been successful enough."
More fundamental reforms are needed, Mr. Newman said, because "the kinds of things we want students to know are changing."
"We make a mistake to talk about schools failing," he said. "The world has moved away from the schools."
But Mr. Honig insisted that such changes can only occur if national leaders acknowledge the progress that has been made.
"I'm not saying everything is fine, and that we don't need to restructure," he said. "But if you throw cold water on people if they've tried, that's a terrible way to lead. People do more if they see results. It snowballs."
"The body is starting to get up," the California superintendent said.