There are good tests and bad tests. The latter should be eliminated. The former can tell us quite a lot about how much and how well kids are learning and how well their schools are doing.
No, they can’t tell us everything we - parents, teachers, school leaders, policymakers, taxpayers - should want to know about the educational performance and progress of our children and their schools. They can’t tell us much about creativity or motivation, about character or leadership, about honesty or kindness. And even the most sophisticated tests aren’t great for gauging whether a student has learned to do research, conduct experiments, compose a poem, or write a coherent essay.
But when it comes to basic knowledge and essential skills in such core subjects as English, math, science, and history, good tests can be enormously informative. And by good tests I mean, just for starters, tests that are carefully aligned with a school’s or teacher’s educational objectives, curriculum, and standards.
Tests come in many forms, from the 15-minute Friday afternoon review of some of that week’s lessons - a new math skill, for example, or spelling words, or whatever - to the three-hour Advanced Placement exams that seek to probe an entire year’s worth (and more) of relatively sophisticated learning.
At the “macro” level, tests such as the National Assessment of Educational Progress don’t tell us anything about individual students or schools but are enormously informative about how entire states and populations within them are performing. Similar “sample-based” tests reveal how the United States is doing vis-à-vis other countries.
Almost nobody loves testing but it’s an efficient and reasonably objective way to determine how the educational process is doing. And it’s become more important - and surely more controversial - as a result of two recent developments.
The federal “No Child Left Behind” act has focused huge attention on student and school performance in English and math and has triggered much argument about possibly narrowing the curriculum, squeezing out other subjects, and turning too much of the school year into test-prep. (These concerns are not unfounded, though NCLB test results have also revealed more than we’ve ever known about how our schools—and subgroups of pupils within them—are doing from grade to grade.)
The second recent development is the widening use of student test results as part of teacher evaluations. The logic here is irrefutable: just as a chef’s main product is nourishing meals that people want to eat, a teacher’s main product is the nurturing of skills and knowledge in her pupils’ heads. If one wants to evaluate the teacher’s effectiveness, shouldn’t the starting point be whether the girls and boys in her classroom are learning what they should? And how will we determine that other than by testing them?
But that isn’t how teacher evaluations have typically been done in the past. Unless done with great care, it can be unfair to teachers, in effect holding them responsible for learning (or ignorance) that isn’t their doing. Test results - even the sophisticated “value-added” kind - don’t get at every aspect of a teacher’s performance that is valuable to her students, her school, or the larger society. And plenty of teachers work in areas (e.g. art, gym, even history) where there are seldom any usable test results or teach kids in grades that aren’t part of the state testing regime.
So we ought not get carried away with the use of student test results as the basis for teacher evaluations. But it would be crazy not to use them whenever possible as part of those evaluations.Chester E. Finn, Jr. is a senior fellow at the Hoover Institution of Stanford University and president of the Thomas B. Fordham Institute.
All statements and opinions expressed on this blog are those of the individual contributors, and not of the Bill & Melinda Gates Foundation or NBC News.