A Test Colleges Don't Need

The New York Times reported last month that a national commission appointed by the Bush administration is considering the use of standardized tests to discover how much college students learn. Campus newspapers across the country have commented extensively on the story.

Such interest is understandable. There are no tests similar to SATs to tell us how much undergraduates know. State legislators, who appropriate billions of dollars each year to higher education, are naturally interested in finding out what they are getting for their money. So are students, who invest heavily in their education. Unfortunately, however, mandating standardized tests is not a promising way to find answers, let alone improve learning. There are other, more promising solutions.

As it happens, researchers have discovered a lot about the progress students make in college. Their findings justify the growing concern for more accountability.

Although professors regard improving critical thinking as the most important goal of college, tests reveal that seniors who began their studies with average critical thinking skills have progressed only from the 50th percentile of entering freshmen to about the 69th percentile.

Tests of writing and of literacy in mathematics, statistics and computer technology suggest that many undergraduates improve these skills only slightly, while some actually regress. Many corporations have to offer programs to teach their college-educated employees how to express themselves.

Although most colleges require students to study a foreign language, they rarely require enough study to achieve a reasonable competence. Only 15 percent of undergraduates enroll in courses of the kind needed to acquire real proficiency. Fewer than 10 percent of seniors believe that they substantially improved their language skills during college. The rest, in the words of one critic, "know enough to read a menu but not enough to compliment the chef."

Teaching methods are often inadequate for the goals faculties are trying to achieve. Important courses such as expository writing and foreign languages are frequently taught by untrained graduate students and underpaid adjunct teachers. Efforts to develop critical thinking falter in practice because too many professors still lecture to passive audiences instead of challenging students to apply what they have learned to new questions. Rather than examine students on their problem-solving skills, instructors test mainly for recall and comprehension of material and offer only skimpy, belated feedback in return.

Despite these problems, standardized tests are a poor way to improve the situation. It is extremely difficult to capture what students should be learning in a single set of exams, especially when colleges and their student bodies are so diverse. In practice, such tests tend to include much that is trivial while leaving out much more that is important. They are unlikely to give an accurate picture of how much undergraduates have learned.

It is equally hard to use test scores to bring about needed reforms. If nothing much turns on the results, faculty members will ignore the tests. If lawmakers try to employ financial incentives to pressure faculty members into concentrating on boosting test scores, professors will either resist stubbornly or escape from teaching basic courses into training graduate students and giving advanced seminars.

Useful reforms can come only from within the universities. Academic leaders will have to work with their faculties to develop methods of assessing student learning that are appropriate to their institutions. They will need to provide funds to experiment and evaluate new teaching methods. They can offer more extensive training of graduate teaching assistants and young faculty members. Above all, they should try to emulate other well-run organizations by initiating a sustained process of improvement in which they continuously evaluate their educational programs, identify weaknesses and experiment with new ways to remedy their deficiencies.

Ultimate success or failure, however, will depend on the faculty. Conceivably, professors will refuse to cooperate, invoking academic freedom or accusing critics of meddling in matters they do not understand. But better things may be in store. Reports from the Carnegie Foundation for the Advancement of Teaching, located at Stanford University, reveal that faculties have become much more interested in undergraduate education since 1970. Groups of professors on many campuses have formed to talk about new ways of teaching. More than half of American colleges are assessing student progress toward at least some important educational goals.

Granted, few of them as yet are using these assessments to build a comprehensive process for identifying weaknesses in their programs and experimenting with possible improvements. Still, the elements of such a system are in place on many campuses. In time, more and more colleges and faculties may find that searching for better ways of helping students learn can be as engrossing as seeking new knowledge in the library and laboratory. The next 20 years might just turn out to be one of the most productive periods in the history of higher education.

The writer was president of Harvard University from 1971 to 1991 and will return to Harvard as interim president on July 1. His latest book is "Our Underachieving Colleges."