Comparing S.A.T. Scores Won't Reveal The State of American Education
On two separate occasions in recent weeks, the Reagan Administration has used Scholastic Aptitude Test data to draw faulty and misleading conclusions about American education.
One of these occasions has already been the object of a critical commentary in this newspaper (see "Papering Over the Reality of Education," Education Week, Jan. 25, 1984.) The College Board, sponsor of the SAT, was also dismayed to observe these misuses of sat data, because they reflect and perpetuate the erroneous perceptions about the meaning and use of SAT scores.
The first occasion (and the subject of the previous commentary) was a press conference called by Secretary of Education Terrel H. Bell in January. At this conference, Mr. Bell displayed state-by-state scores on the SAT and the American College Testing exams. He used data from the test most widely used in each state as one of the comparative measures of a state's educational performance.
Although the Secretary issued caveats about the use and interpretation of SAT scores, these were largely ignored by the press and public--a predictable reaction considering the human tendency to reduce complex issues to simple concepts and numbers.
The second occasion was President Reagan's State of the Union address. Mr. Reagan said that "a 600-percent increase in federal spending on education between 1960 and 1980 was accompanied by a steady decline in SAT scores." He implied that increased expenditures for education do not produce improvements in education, but are associated with, and possibly cause, a decline in sat scores.
The fact is that during the past two decades most federal assistance to schools has been focused almost exclusively on students who were least likely to take the SAT or the ACT--elementary-school students with reading and mathematics problems, handicapped students, bilingual students, economically disadvantaged students, and students in vocational-education programs. To focus federal assistance on these students, cite the test scores of the college-bound, and then conclude that federal assistance "caused" the achievement of the tested groups to decline is shoddy reasoning at best.
The decline in sat scores has been well-researched. In 1975, the College Board and the Educational Testing Service appointed a distinguished panel headed by former Secretary of Labor Willard Wirtz to investigate the reasons for the decline in national SAT scores that began during the 1960's. The panel attributed most of the decline to the substantial increase in the number of students taking the SAT. Thus, an equally valid interpretation of the score decline as it relates to federal educational outlays might well be that these outlays, particularly student financial aid, made it possible for large numbers of minority and poor students to have access to higher education.
Unfortunately, attention to the score decline produced the notion that the national average SAT score is an accurate measure of the quality of American education and is a fair way to compare the educational performance of schools, districts, and states. These assumptions, evidently shared by the Administration, are not true for the following reasons:
First, college-bound students constitute about half of all high-school graduates, and only one-third of college-bound students take the SAT.
Second, the SAT measures individual verbal and mathematical skills that develop as a result of experiences both in and out of school. The test is thus sensitive to many social and economic factors and is not well-suited to identify the effects of formal education programs alone.
Third, colleges receiving SAT scores are not uniformly distributed across the nation, and so the proportion of students taking the SAT in each school, district, and state varies widely.
Fourth, the student populations in schools, districts, and states vary widely, as do the educational programs developed to meet their disparate needs and abilities.
Thus, SAT score averages are not an appropriate or scientifically valid basis for comparing the overall performance of teachers, schools, districts, states, or other groupings, or for judging the overall quality of American education. They must be used with extreme care and for a limited if valuable purpose--most notably to assess the academic performance and needs of individual students.
The College Board annually publishes a national summary of SAT scores and provides summary statistics to appropriate education authorities in those states, districts, and schools in which a substantial number of students take the SAT. These reports help school officials put individual scores into context, and, over time, suggest changes in the academic performance of college-bound students.
Similar uses can be made of state, district, and school averages, as long as officials bear in mind that SAT scores are not definitive measures of academic performance. Score summaries, with other information voluntarily provided by test-takers, can also help colleges plan their curriculum, faculty staffing, student recruitment, facilities, financial aid, extra-curricular activities, and student services.
These uses--not the assessment of American education--are appropriate uses of SAT scores.
Vol. 03, Issue 22, Page 18