An increase in marks does not mean a student is making progress

Standard

I don’t have a mark book. There’s little point.

Having a set of marks for a student doesn’t really tell you much. For them to be meaningful they need a lot of extra data that isn’t efficient to record in a mark book.

Example

Let’s imagine a test that is out of 20 marks; a 1/2/4/5/8 mark questions.

A student does 3 different tests in a term and my mark book looks like this:

10/20. 12/20. 15/20

It seems this student is making good progress; their marks are going up. But of course that isn’t true. There are far too many variables for this to have any sort of meaning.

  • What if the first test they only attempted the 2 mark and the 8 mark?
  • What in the second test they only attempted the 4 mark and the 8 mark?
  • What if in the third test they attempted them all but got 3 in the 8 mark?

None of those tests are anything like each other, other than mark structure. The questions aren’t even the same, nor are their answers. It is absolute nonsense to say that this student is making progress.

Yet so many people are fooled by it. Leaders are placated by the lovely ‘arrow up’.

We need to stop pretending that increasing marks mean progress.

Possible solutions

  1. Give them exactly the same test each time. Even then it is not resolved as the ‘make up’ of their score may be different each time.
  2. Ditch marks/grades/levels. They don’t mean anything and are a poor proxy for progress. Research also is very clear on how using them isn’t a great idea
  3. Use a method that looks at the work itself (see my blog here on how that could be done)
  4. Have an incredibly detailed mark book that records:individual question scores,what the questions were and the things they didn’t do for each question. Not exactly time saving.
  5. Get students to make a record of their answers and what they did/didn’t do. I do this for end of year exams via a google form. It relies on them being honest and understanding their gaps. Even then, it’s just a snapshot of one test.
  6. Redefine progress. Don’t use spreadsheets or data unless it really tells a story. Stop getting teachers to use grades and then using it to analyse progress. Just ask a teacher ‘do you think they’re making progress? How? Why?’. Give subjects time to establish how they think they can determine this.

3 thoughts on “An increase in marks does not mean a student is making progress

  1. Good to challenge simplistic ideas about numerical methods of assessment that don’t make clear what is being measured! I personally think a combination of hard data – the number mark – and soft data – the commentary, judgments and insights of the teacher – provide the richest source of information. Then the challenge is how teachers and leadership record, monitor and intervene effectively based on this richness of data. I don’t know the answer to that yet!

  2. Another possible solution?

    Combine all the questions that pupils got wrong on the first test and retest those questions on the next test. Keep following this cycle.

    That way you’re not re-testing questions they got right first time and you’re honing in on whether they get the questions right that they got wrong before.

    If they were to get these questions right after getting them wrong would that be ‘progress’?

    R

  3. Pingback: The possible fallout from cancelling exams for teachers | missdcoxblog

Leave a comment