Agency Calls for Better Science Indicators
Washington--The National Research Council has recommended that educators and policymakers develop new sets of indicators--including more extensive student and teacher testing--to improve the assessment of precollegiate science and mathematics instruction.
In a report released here last week, the council, a branch of the National Academy of Sciences, also proposes establishing a national research center to facilitate the creation of such testing measurements.
"The indicators we now have are vastly better than we had in 1960," before the creation of the National Assessment of Educational Progress, said Richard J. Murnane, professor of education at Harvard University and chairman of the nrc panel that developed the report.
"But there are an awful lot of things that are not known, and areas that can be improved," he said.
The report, Improving Indicators of the Quality of Science and Mathematics Education in Grades K-12, recommends that new types of tests measuring students' higher-order thinking abilities be developed.
It also urges that policymakers collect data on student attitudes toward science and math, that teachers be tested to determine their subject-matter knowledge, and that curricula be measured against national standards.
While the report does not esti4mate the costs of developing the new indicators, Mr. Murnane said that panel members assumed they would be borne by the federal government.
'More Penetrating Insights'
The report is the second in recent years by the council's committee on indicators of precollege science and mathematics education.
Its first report, Indicators of Precollege Education in Science and Mathematics: A Preliminary Review, issued in 1985, presented data from existing indicators.
But the review showed that more research was needed to develop indicators that would provide "more penetrating insights" into the quality of instruction, the new report states.
Specifically, it recommends new forms of student assessment--particularly for grades K-5--to replace the multiple-choice tests currently available. Such tests, it says, "are not adequate for assessing conceptual knowledge, most process skills, and the higher-order thinking skills that scientists, mathematicians, and educators consider most important."
The proposed national research center would facilitate the development of the new tests by producing, evaluating, and distributing assessment materials to states and districts.
Noting that "the science and mathematical literacy of adults is a major goal of science and mathematics instruction," the panel also recommended the testing of a random sample of adults every four years.
But the report notes that testing for student achievement alone is not a sufficient tool for evaluating an instructional program.
"Unless information about behavior and attitudes is known, in addition to test scores, for a reasonable sample of the student population,'' it states, "the focus for ... policymakers becomes blurry and decisions tentative."
To evaluate such attitudes, the report recommends that data be collected on secondary-level course enrollment and the amount of instructional time devoted to science and math.
In addition, it recommends a greater effort to collect data on teacher quality, including periodic testing of teachers' knowledge of their subject matter.
The purposes of such tests, the panel stressed, would be "to assess the extent to which students are taught science and mathematics by teachers who have mastered the knowledge and skills they teach, not to denigrate the ability or aptitudes of teachers."
The panel also called for the collection of comparative data on beginning teacher salaries and pay scales in other occupations, as a way of determining the quality of teachers' working conditions.
Noting that what is taught is also key to determining how well students learn, the report recommends ways to measure curricular quality.
Specifically, it urges national organizations to develop exemplary curricular frameworks, and that state guidelines, textbooks, and tests be matched and rated against them for content coverage.
Copies of the report are available for $17.50 each from the National Academy Press, 2101 Wisconsin Ave., N.W., Washington, D.C. 20418.