Once again, the Education Inspectorate paints a worrying picture of education: its quality is under pressure, and the learning climate in secondary education is insufficiently stimulating. What in The State of Education 2026 What remains underexposed is how testing culture contributes to this. Getting students into secondary education an average of 102 figures per year. But the standard test with a grade is a poor indicator of learning performance and is detrimental to good education.
Tests mainly show what can be retrieved in the short term, not what has been learned sustainably. Imagine: two days before the test, your child is crunching facts, lists and definitions. She or he passes the test, the grade is good. But a month later, most of the material has disappeared from memory. Cramming for a test is learning to forget.
This makes sense if you understand how sustainable knowledge is built up in the brain. Knowledge is stored in two steps. First, the knowledge goes to a temporary waiting room, the hippocampus, a small area deep in the brain. After this, a complete sorting and selection process takes place. Only what is really important is recorded in the great library of long-term memory, the neocortex. What remains in the waiting room is thrown away.
Not a reliable indicator
We think that passing a test means that students have built up knowledge. But when students take a test, we cannot distinguish between knowledge that is in short-term memory and will disappear again, and knowledge that is really stored permanently. A pass on the test is therefore no guarantee that the student will still remember the material a month later. You use buttons to check whether water flows from the tap, while you do not know whether the stopper is in the bath.
The problem is not only that tests are not a reliable indicator of sustainable knowledge building. Testing also reflects on teaching and leads to lessons that have little stimulation. What will be assessed at the end determines what will receive attention along the way. But not everything is equally easy to test, and that throws the focus out of balance.
The effect of testing is reflected in two frequently asked questions in the classroom. As soon as a lesson becomes richer and more meaningful than what fits into a test, students ask: is this for the test? But as soon as the lesson narrows down to what is testable, the other question follows: what good is this for me? Two sides of the same coin.
Taking and marking the test should not take too much time. Each test question must have one clear correct answer and answers must be easy to score. The test mainly focuses on what is easy to assess: definitions, lists and sums in a highly simplified context. And what is difficult to capture in a test question – coherence, reasoning, real application – fades into the background.
You use buttons to check whether water flows from the tap, while you do not know whether the stopper is in the bath
This mechanism is known as Goodhart’s law: as soon as a metric becomes a goal, it loses its value. Testing is a textbook example of this. It is a system error that equally affects teachers.
But without tests, do we still know whether learning is taking place? Measuring is knowing, right? No, measuring is only knowing if you measure what matters. Most tests mainly measure whether students can reproduce individual facts shortly after learning. If you really want to know whether students have learned something, you should check this more often during the learning itself.
Understanding the world
Therefore, regularly gauge where students are standing, without a grade. Have students perform various rich tasks: give a presentation, have a discussion, create a portfolio or construction, and evaluate with them whether the learning goal has been achieved. This approach is not more optional, but more demanding: it actually tests more rigorously whether the goal of education is being achieved. Facts and definitions are valuable, but only if they become part of something bigger: understanding the world and the ability to participate in it.
As soon as the test no longer limits how the material in the lesson takes shape, space is created. Don’t give students a formula to mash, but dice with seven, twenty-four or a hundred faces, and let them come up with a formula for the chance of an even number. Such a lesson brings together exactly the elements that make learning sustainable: wonder, making connections and learning by doing. These elements act like a stopper in the bath: they increase the chance that the brain will mark knowledge as important, so that it ends up in the long-term memory.
That is why we must abandon the standard test with grades as a measure of learning performance. Testing policy is a choice. School leaders can start small: investigate in the first year what happens if you stop. Parents and students: do not long for the false certainty of a pass. As long as the test remains the benchmark for learning, students will not progress further sweating, knowing and forgetting.
