In a previous essay I suggested that the vast majority of teachers should be freed from the tyranny of tests. Instead of making teachers and schools prove themselves every year, let’s assume that they are doing a good job unless there is some reason to believe otherwise. If less than 10% of our teachers deserve to be fired, why burden the other 90% (and their students) under a crushing regimen of testing?
A problem with this argument is that once a teacher is flagged as suspect, one obvious response is to test that teacher’s students to see if they are learning as much as they should. But to do that, it is important to know what is “normal” for that subject/grade level. Don’t we need to test everyone to get that information?
No!!
Tests do need a standard, but not every student in every classroom needs to take a test to generate such a standard. If, every year, say, 1% (or some other small number) of fourth grade math students are randomly chosen to take a test, that is still a lot of data. If the teachers are not told ahead of time who will be selected, they will not teach to the test, they will teach what they think is important the way they think is best. There will be a wide range of results, because not all teachers will have taught the same material the same way, but still, fourth grade math is fourth grade math: by the end, students should be able to do most of it. The results will go into a data bank; it will not be construed as a measure of the quality of the teacher or the school; there will be no pressure, simply a measure of what students can do. The exception will be those teachers who have been flagged as suspect—their results will be scrutinized. Also, teachers/schools/districts may be flagged if their students are grossly below average on these random tests.
Another relevant question is, who constructs these diagnostic tests, and what should be on them? After all, it took years of argument and thousands of man-hours from highly educated people to come up with the Common Core standards. Won’t this be just as hard?
It shouldn’t be.
What is fourth grade math? I have no clue. I took it 55 years ago. But there are textbooks out there for fourth grade math. I’ll bet that there are about 5 textbooks that are used by 90% of the students in the country. (Based on what I hear about the influence of the Texas School Board, 5 may be an overestimate.) Further, I’m guessing that they are all pretty much the same. Sure, they do things in a different order, explain things somewhat differently, etc., but they all cover the same 10 topics/concepts/skills in the course of a year (and the books used by the other 10% of students nationwide do too). Students graduating from fourth grade in one district or state should be able to do more or less the same as students elsewhere.
So here’s how to construct a test. Choose 6 semi-random problems from each of these 5 books. That’s 30 problems. Give kids 1 hour to do the test. That’s 2 minutes per problem.
What does semi-random mean? Simply that in the final selection of 30 problems, each of the 10 major concepts should be represented by at least 2 problems, and further, no 2 problems about any of those 10 concepts should come from the same textbook.
Problem solved!