Thursday, 16 April 2015

Assessment without levels ... a successful trial!

I didn't go out of my way to start an 'assessment without levels' trial. It just seemed to be the only sensible way forward.

My teaching career started in Year 5, but moving to Year 1 after a spell in EYFS, I couldn't face going back to the world of APP. It made no sense in light of the new curriculum and the upcoming demise of levels.

I managed to gain agreement to move straight into assessment without levels for the 3 classes in my year group. It was officially a trial, but I never saw it that way. There wasn't really anything to go back to, so we were more of an advance party moving ahead of the main field.

If there is such a thing as 'being in the right place at the right time', then this must have been it for me. Having recently taken on board the revised early years framework, the parallels with assessment without levels were uncanny ... 

Next Step 'Trees'

I blogged last year about 'Next steps in EYFS not needing to be a cottage industry' and introduced the simple yet effective 'next step trees' that I had created. They paint a clear picture of where children are in their learning journey and are easily shared with parents/carers. Here's an example ...

I'm a big fan of the way core subjects have been revised in the new curriculum (Here's my blog post on maths). The objectives are now mostly clear and succinct, with some even written in child friendly language! It seemed an obvious first step to create some 'next step trees' for Year 1. Our classes are named after sea creatures so the tree became seaweed, but the concept is just the same. I added the Early Learning Goals (ELG's) from EYFS at the bottom, and chose a few 'early start' objectives from Year 2 to go at the top. Here's the one I created for maths ...

We decided to focus on number, splitting some objectives up for the sake of progression, but it would be easy enough to add some key objectives for geometry and statistics etc. I also created sheets for reading and writing, all of which where laminated and placed inside the front cover of the children's work books. The children took to them instantly, having stickers added for the ELG's they had already achieved, and then for the easier Year 1 objectives.

You may be thinking there are too many targets for the children to focus on, and those with experience will know that many Year 1 children aren't capable of reading them at the beginning of the year. However, some children started counting the stickers they had got. Then they counted the ones they hadn't got. Then it caught on ...

"I've got 3 stickers."  Good.

"I've got 4!"  Well done!

"I've got 6 out of 19, and I only need 12 more to be the winner."  Impressive maths skills ... actually you need 13 more so you might not be the winner yet.

The children grasped the size of their learning challenge for the year and were positive about the progress they were already making.

The target trees were central to discussions at our first parent's evening of the year, and feedback was very positive. Some parents occassionally come in and look at the trees when they drop their child off in the morning, checking how they're doing and picking up next steps. It really has led to a higher level of effective home support.

DfE Draft Performance Descriptors

Next came the very timely arrival of the DfE's Draft Performance Descriptors ...

Up until now we'd been using similar terminology to EYFS in terms of how a child was progressing,  e.g. 'working towards, at, or exceeding age expectations'. The DfE now gave us some new words to play with.

Your daughter is "Working at National Standard" sounds very official indeed.

Your son has achieved "Mastery" sounds positively, well, masterful!

The document said we weren't allowed to use the new words yet, and in any case they are only supposed to apply to the end of each key stage (which apparently was also the case with levels!) We're supposed to come up with our own ideas on how to assess in between. So this is a big thing, a turning point. There are two ways of looking at it ...

1/. Really!? The DfE aren't going to tell us what to do, and we're banned from using their idea. Here we go again, thrown into chaos.

2/. Interesting ...  so we're being encouraged to use our own ideas, and they've told Ofsted that they'll have to use whatever we give them. The DfE even say they trust our professional judgement.

Needless to say I chose the second way of thinking. I've long been frustrated by how downtrodden we let ourselves feel as a profession, and the relatively low status we hold in society. The door is now open ...

Can you imagine a GP filling in an 'APP' style sheet to arrive at a patient assessment? Actually, they'd be told to fill just one sheet in for each group of similar patients and give them the same diagnosis. Madness! With proper checks in place we trust their professional judgment.

Admittedly GP's pay may have something to do with our relatively low professional status, but surely we're just as trustworthy? The DfE have made a point of saying we are.

I see the pay issue as a 'chicken and egg' one, tied up with professional status. We need to get into an upward spiral, raising both. Thinking about it, we'll probably have to include educational standards in the equation as well.

So let's not complain about a lack of clarity or direction. Let's take the opportunity to show how proactive and professional we really are.

Forgive me, I may have got a little sidetracked. Back to the Levels ... sorry, I mean performance descriptors. Can we still use the word level, as in 'Mastery' level? Surely we can.

Assessing against the new curriculum

As we moved towards the end of the first term, we got properly organised to make assessments. We used a two-prong approach ...

1/. Testing. This word brings all sorts of baggage and pre-conceptions with it, but please hear me out.

2/. The target trees.


As a school we invested in the Abacus maths system for the new curriculum, which comes with short half termly tests to check progress. We also bought similar Rising Stars tests for reading.

The most attractive feature of the above tests is that they are cleverly designed to check how children are progressing in each half term. They get harder, meaning that a score of 70% (or similar) in each half term is maintaining 'National Standard'. It took us a while to trust the results, but we do now and they correlate very well with the target trees.

We think of them as tests, but the children don't.

The tests are carried out on a 1-2-1 basis in a quiet area just outside the classroom. Initially this was driven by their inability to read, but we wouldn't change a thing now that they mostly can.

There is a test paper involved, but it's more like a coaching session. We don't ask every child to try every question. We coach them on the questions they haven't got right (without giving a mark) and look to move them on. For them it's 5-10 minutes of positive attention, showing off what they can do, and learning. They sometimes queue desperately by the door, hoping to be next!

You end up with so much more than a simple mark at the end of each session. You know exactly who needs intervention and what that intervention needs to be. Given two children who score 95%, you can tell which one is working 'Above' and which one has truly 'Mastered' what you have taught.

It is very time consuming though. It takes three quarters of every day for a week in each half term. Whilst the teachers are out, the children choose independent learning activities under the TA's supervision. It's not perfect, but we're all getting better at using the time effectively. It's a week we all look forward to!

If I found myself back in Year 5 tomorrow, this is the first thing I'd put in place. I know it would make a difference.

The children also do some independent writing during the week which we use as a basis for assessment. I have to admit we're still levelling the writing as it enables us to moderate across year groups. We do turn the levels into new assessments though, e.g a Level 1c is 'National Standard' in Autumn, but 'Working Towards' in Summer.

Whilst I'm on the subject of moderation, it's interesting to see that the DfE will be publishing exemplification materials for the end of KS1 and KS2. Another parallel with EYFS who have been using exemplification for a while now.

Target trees

These work very well to give an overview of where a child is. You can see how many and which stickers they have got, enabling you to make a 'best fit' judgement. I'm scared by the number of commercial assessment systems I see that are based on counting up the number of objectives a child has met. Not all objectives are equal! This is not professional judgement.

The systems that tried to do this in EYFS have all but died out now. EYFS know all about avoiding 'tick boxing' and how to use their knowledge of the child to arrive at professional judgements. Teachers with recent EYFS experience will take to assessment without levels like ducks to water!

The target trees have correlated well with the test results in general, and really help us to spot those children who under or over perform in tests. They also work well for moderation because they are transparent.

You think carefully before giving each sticker, as anyone could sit down with a child and check to see if they really can do it ... a parent, their next teacher, the deputy, even an Ofsted inspector. There was never an easy way to sit down next to a child and check if they were a 2b in maths or not! This has to be a big step forward.

Recording & analysing new assessments

Recording systems were few and far between at the time. We considered using a tablet based EYFS system that had been extended to cover the new curriculum for year 1, but it meant taking lots of observations and creating an online evidence base. This is fine in an EYFS environment, but when children's book work has to be marked on a daily basis the additional workload would be impossible. I refused to run a dual system and no-one was going to be persuaded to drop the books. It was time to dust off the Microsoft Excel skills!

It's a big mind shift moving from 'average point score' to 'percentage of children at National Standard', but it's no different to measuring children reaching a 'good level of development' at the end of Reception.

The ELG's are measured in three bands (emerging, expected and exceeding), but I think the DfE are right in suggesting four bands for KS1 and five for KS2. Using 'Above' and 'Mastery' will help keep focus on the most able as they rise through school.

There is an issue with 'Below' though, as it's an awfully big bucket! The DfE have promised a code system to bridge between 'Below' and the P-scales, which will hopefully give us some guidance for next year.

Once you have shifted your mind, the data analysis all heads towards the new floor targets for the end of Year 6. 85% of children need to be working at 'National Standard' in maths, reading and writing.

From a statistical point of view, working with bands rather than Levels is a dream! It's easy to work out group, class and year group averages, and you can use standard deviation to measure gap closure.

Our fledgling ideas in Excel have been taken forward into a full application. If interested, you can find out more by clicking here.

Assessment without levels commission

Just as we were expecting the draft performance descriptors to be confirmed ... along came the 'Assessment without levels' commission. The way it has been set up sends another signal that the DfE want the profession to be proactive.

The DfE announcement called it a 'teacher-led commission' and to quote the School Reform Minister Nick Gibb ...

"This commission will continue the evidence-based approach to assessment that the government has already put in place, and will support primary and secondary schools with the transition to assessment without levels, identifying and sharing good practice in assessment."

I admit that I got quite excited about this, and seeing that Nick Gibb is my local MP I decided to write and tell him about our trial. He liked the sound of it and is coming to see it in action in a few weeks time.

Update : Nick Gibb's comments ...

On the target trees ... "This is exactly what it's all about, knowing exactly what the children can and can't do." Nick sat down with some children and was impressed that they could talk about and demonstrate the objectives that they had met.

On Evoke (the excel based recording and analysis application developed during the trial) ... "You've gone further than other trials I've seen and are successfully using an electronic system. It makes it very easy to see what's going on."

The stress of workload and change

I like to get involved in change. Experience in previous careers has taught me that change in a complex system, when managed well, results in win-win situations. The trick is to make sure that all parameters and concerns are taken into consideration. I trawled though the recent DfE publications, press releases and speeches, to pull together the parameters that they have put on the table ...

(This slide show demonstrates how I arrived at the above criteria using the various clues that the DfE have left us ... click here)

I can't think of anything to add to this, and it's up to us as a profession to get the balance right. In my view, workload reduction is the biggest win we need as classroom practitioners. I refuse to allow any more duplication, double recording, tick boxing and the collation of evidence folders!

To draw a final parallel with EYFS, county moderators are just as interested in what a teacher can tell them about a child as the physical evidence they can provide.

The processes that our trial has led us into are simple and efficient. There's no duplication or excessive paperwork. There is plenty of professional judgement though. Senior management have commented on how well we know the children, and progress is clear in the data we provide them. They trust us. The trial is rolling out across the school for the summer term.

And finally ...

Can I sum up what 'assessment without levels' means to me in a single sentence?...

In the past, being perfectly honest, my use of 'teacher assessment' was a fudge to help me through the near impossible task of using APP. I never really developed a feel for where a child was in every subject, or managed to remember what a level 3c in maths meant they could actually do. I knew how many children had to make 2 sub levels progress though!

'Teacher assessment' means something very different to me now. Working with such a succinct and measurable curriculum (I know it has its issues!) and being trusted to use my professional judgement has ended up with me ... well, being a professional who can exercise judgement effectively.

I can sit down with a child and their parents and explain exactly what the child can already do, and what they need to do in relation to the 'National Standard'. I can show them the key objectives for a subject on a single sheet and demonstrate how their child is progressing. This can only lead to a meaningful and honest discussion. In fact, a formative one. And isn't that the whole point! The DfE themselves describe what we should be doing between the end of key stages as 'formative assessment'. 

I have no doubt that some teachers have always managed to achieve this, but not me, and I'm sure that 'your child is a level 3c' was never a great place to start from.

So no, I can't sum it up in a single sentence. I think that reflects the size of change that this is. It also means that running your own trial in advance of September would probably be a smart move.

The Edgazette
Related Posts Plugin for WordPress, Blogger...