While some schools have simply replaced the old levels with unhelpful practices, a number of companies are developing solutions to fill the gap, says Joshua Perry
When the DfE announced that levels would be scrapped in June 2013, many people cheered. Finally, schools could focus on formative assessment practices that actually improved outcomes! Professionals suddenly had freedom to devise something better! So what if there wasn’t a clear consensus as to what the ideal system looked like, right?
But the problem is, when you let a thousand flowers bloom, you’re also likely to grow some weeds and – as any gardener will tell you – it’s not always easy to tell them apart.
In recent years, there has been a tendency amongst schools to gather ever-increasing quantities of data. In many ways this is a good thing, but there have also been some dodgier practices, the most dangerous of which tend to involve magically turning teacher assessments into summative scores to compare cohorts or schools.
To understand why this academic alchemy is a bad idea, it’s helpful to recap on what teacher assessment involves. Essentially, a curriculum is divvied up into a bunch of performance descriptors, such as: “Can compare two fractions to identify which is larger”. Then, a teacher will assess each child, using evidence from the classroom, and mark them using a simple scale (e.g. “Emerging”, “Expected” and “Exceeding”). The data gets entered into an online tracking tool, like the ever-popular Classroom Monitor and Target Tracker.
There is another way, and it’s rapidly gaining traction
I should be clear: I’m a big fan of such systems for formative purposes. They can be invaluable to teachers in tracking curriculum coverage and identifying gaps in knowledge. After all, tests don’t test everything, so teacher judgment is a vital part of good classroom practice. However, I do get anxious whenever teacher assessments are aggregated and compared as if they are high-quality summative data, because teacher assessments are subjective. An individual teacher may moderate their own judgments, but as soon as you try to compare judgments from multiple teachers for accountability or performance management purposes, you’re in trouble. The data is simply not up to the job.
The good news is that there is another way, and it’s rapidly gaining traction. More and more schools are using standardised assessments alongside tracking systems to benchmark themselves against a reliable, norm-referenced dataset. Such tests don’t just give a grade for each child; they also offer a standardised score (such as a percentile) which expresses the child’s result in a national context.
Companies such as GL Assessment, Rising Stars, Renaissance Learning and CEM not only offer standardised assessments in English, maths and science; they also provide valuable measurements of well-being and pupil attitudes.
What’s more, the sector is adapting rapidly to cover new areas. A new technique called comparative judgment is being used by No More Marking (also known as Sharing Standards) to standardise writing assessments. This innovative approach does away with traditional marking entirely, and instead gets teachers evaluating which of two essays is better. Once this has been done enough times for each essay, the system puts them all in a rank order to standardise grades. Amazingly, it’s less work for teachers than conventional marking – and considerably more reliable.
It’s time to move away from voodoo analysis
Standardised assessments don’t replace pupil tracking systems. Instead, by using both schools can have the best of both worlds. A growing number of MATs (including Ark and Reach 4) are putting standardised assessments at the heart of their internal processes. It is increasingly common to use standardised assessments annually or even termly across all year groups. Primary schools can now base their entire summative assessment model around standardised assessments. And as adoption and frequency increases, costs (which are admittedly not trivial) should decrease.
Partnerships are also vital. Assembly has been working with standardised assessment providers and Management Information System suppliers to develop analytics dashboards that prioritise standardised measures. We’d also love to see vendors of tracking systems incorporating data from their standardised peers to help validate teacher assessments.
So, it’s time to move away from voodoo analysis, in which formative data is subjected to a bunch of calculations and expected to become somehow summative. School leaders should embrace standardised data for meaningful cross-school comparison. After all, it’s not the size of your data that matters; it’s the quality that counts.
Joshua Perry is director of Assembly, a schools data project founded by Ark and the NEON foundation
Teaching is one of the professions where it is easy to measure effectiveness of the school or teacher, provided the assessment method used is valid. Those schools and teachers that do their job well will welcome testing. Those who are not will oppose testing and they keep going on about ‘parrot fashion’ and ‘teaching for the exam’, as if these are bad things, when in fact they are desirable and improve the child’s levels.
KS2 maths tests include open ended questions and marks for workings. This involves huge expense and time when it comes to marking. It is possible to take a KS2 maths test and rewrite it without any open-ended questions, but keeping the topic of each question the same. The significant point to make here is that THE CHILDREN ARE RANKED IN THE SAME ORDER after taking the simplified tests. You do not suddenly get the children changing their positions. Therefore it was a huge mistake for the government to make the KS2 maths difficult to mark, when they could have made them easy to mark and just as effective. I do not criticise the content, which, as I say, can be preserved after a rewrite.