How often should we assess pupils? The answer, of course, depends on the purpose of the assessment.

In some lessons, formative assessment will be happening every few minutes or even seconds, as teachers scan the class for instant feedback. If a teacher is explaining something and they adjust their explanation because they see a couple of pupils frowning in confusion, that’s formative assessment. If a teacher adapts their next lesson based on the results of an exit ticket question at the end of the previous lesson, that’s also formative assessment. This kind of assessment is clearly vital, but it is also ephemeral and difficult to capture on paper.

However, the question “how often should we assess?” sometimes refers to the formal assessment points or data drops where schools centrally record pupil performance in the form of a grade or percentage. This is clearly going to take place a lot less frequently than every few minutes. But how often is best for this kind of assessment?

Many schools carry out such data drops three to six times a year. It is worth considering whether such frequency is necessary.

Graded exams such as GCSEs and Sats are normally measures of large domains, which pupils don’t improve on rapidly. Over a short period of time, any improvements or regressions are more likely to be down to statistical noise rather than actual changes.

Such tests also take a lot of time to administer and mark reliably. And they are not designed with formative assessment in mind. That’s not to say you can’t get some formative information from them, just that they are designed with grading in mind, not next steps for the pupil and teacher.

Fewer data drops

For an example of this, consider the difference between two different types of vocabulary test. The first test aims to provide a measure of the number of words a child knows. The second test aims to see if pupils have retained the meaning of five new words they learned in class last week.

On the first test, you would not expect dramatic improvement over a period of a couple of weeks. The average vocabulary size of even young children is thousands of words, and tests have a margin of error. So repeating this kind of test every six weeks probably isn’t going to be that helpful.

The second test, however, has real learning benefits. First, it helps the teacher and pupil to see if the words have been consolidated in long-term memory, which can help them see if they can move on with some new words, or if refresher lessons are required. Second, the act of taking such a test will actually help the pupil learn, as they are attempting to recall something from long-term memory, which helps solidify the memory itself. It would be worth repeating this exact test and others like it more frequently.

If schools had fewer formal assessment points, they could use the time saved to make those assessments as accurate as possible, and to make their separate formative assessments as useful as possible.

Daisy Christodoulou is director of education at No More Marking and the author of Making Good Progress? and Seven Myths About Education. She tweets @daisychristo