Do you remember back in 2013 when Michael Gove announced the end of levels? At the time it seemed that there were more questions than answers - and that was probably true inside the Department for Education as much as it was on the outside. What might have seemed like a simple and productive move actually opened a can of worms.

First, there was the teacher assessment debacle. I’m not persuaded we’ve solved that one yet, but it seems that getting rid of key stage 1 Sats and removing teacher assessment for reading and maths in Year 6 may be part of the sledgehammer solution. It speaks volumes about the chaos of the system that the loss of teacher assessment judgements has been mourned by virtually no one.

Then there was the challenge of sharing test results with parents. After 20 years of getting used to the simple levels system, the new system of comparing to the scaled score of 100 was never going to be straightforward. Again, I’m not sure we’ve really solved that one yet, but schools are doing their best to help parents to understand the system.

But here we are, five years later, and some problems still have no resolution.

Making progress?

At first, we stuck with the old levels system long enough for the department to try to prepare a system for measuring progress across key stage 2. The resulting system actually makes some sense: individual children are compared only with others who had similar starting points, and schools get credit for the improvements made with all children, not just those who cross the key thresholds. But I think it’s fair to say that it’s a limited audience who really understand how the prior attainment groups are worked out.

But if you think that was tricky, imagine the frustrations of those at the DfE responsible for coming up with its post-levels replacement. Up until now - and, indeed, for one more year - we’ve had pupils coming through KS2 with their KS1 judgements in old levels. The old, familiar APS system is still used to give a points score to mark their starting point, but what happens to that when we reach the first cohort without levels in 2020? Those pupils are already in Year 5; it’s not that far away.

The obvious solution, you might think, is to take their test scores as a starting point. After all, every child took the same tests in Year 2, with every score matched to a scaled score using the common thresholds. It makes for a nice easy comparison of starting points. Except nobody thought to collect the test score data. Maybe a few local authorities here and there have a spreadsheet in a folder somewhere, but for the most part all we have is the teacher assessment judgements.

Even if we overlook the inherent challenges of teacher assessment (i.e., the fact that it’s wildly unreliable, particularly for junior and middle schools), there’s the challenge of the simplified categories. Up until 2015, children could leave Year 2 with one of six outcomes for most subjects (level 1, 2c, 2b, etc.). From 2016, that was reduced to just three: “working towards”, “expected” or “greater depth”. The nuances that used to exist have been lost.

In the first year of the new tests, more than half of children achieved the “expected standard” in each subject - but that represents an enormous range of ability that we can no longer differentiate between.

It may seem a way off at the moment, but I can’t help but think that progress measures in 2020 are going to be a bit of a mess.

Michael Tidd is headteacher at Medmerry Primary School in West Sussex. He tweets @MichaelT1979