Writing a new Maths Scheme of Work – Part 17: Assessments

You can view all the posts in the epic “Writing a Maths Scheme of Work” series on this page. It’s kind of like Game of Thrones, only with slightly less nudity and dragons.

Okay, so assessments. Now, from the outset I am going to say that I am not entirely happy with our assessments, and we are looking to change them for September.

Here is what happens at the moment. We formally assess each year group every half term with a written assessment. This is marked and recorded centrally on the spreadsheet. There are four types of each assessment depending on what class the student is in, and these follow the same structure and names as our homeworks, with Wiles, being for the top two sets, followed by Newton, Gauss and Fermat.

In Years 7 and 8 we use… wait for it… past SATs paper questions. Now, I know this will immediately divide half the readership of this blog (that’s both of you), into the Pro and Anti SATs camp. And to be honest, I am a bit divided myself. Anyway, here are some of the reasons why we chose to do our assessments this way:

• using a programme like TestBase they are pretty quick to create
• there are mark schemes available
• the questions can be challenging and do promote problem solving
• many of the questions are decent

However, they do have some notable cons:

• the style of the questions are not necessarily aligned with the GCSE
• whilst there are a fair few questions, there are a finite amount, and the same questions keep cropping up again and again. I must have seen that seen that flipping question about screen wash about 936 times!
• often not enough marks are allocated (in my opinion anyway) for working out
• unless you use full papers, you are left guessing at the level boundaries. Or, worse still, making them up after the marks are in!

Have a look at a set of our Year 8 assessments, so you can get a flavour of what I am talking about:

Wiles

Newton

Gauss

Fermat

As I say, I am not entirely happy with these.

In Years 9 and 10 we do something similar, but with GCSE questions. This solves a few of the issues with the styles of questions themselves, but by picking and choosing questions you are once again faced with the difficulty of assigning levels or grades to the outcome.

So, what is the alternative?

Before considering that, you need to ask yourself what is the point of an assessment, and what is the most important thing to you?

• Do you want to assess prior knowledge?
• Do you mind if students see questions on topics that they have not met before?
• Do you want the questions to have been created by an external body?
• Do you want to know level/grade boundaries before the paper has been sat?
• Do you want the questions to be in the style of the GCSE exam that students will be sitting and upon which they (and you) will ultimately be judged on?

Myself and my Head of Department, Karen, are currently wrestling with these questions. One option that is proving rather appealing is an idea I got from a recent workshop delivered by the lovely Just_Maths team. In their school they use complete GCSE Foundation papers for each assessment for Year 7 and 8. Students are then given a grade – specifically a “fine grade” with each letter broken up into things like E1, E2 etc – and these grades are tracked not only by the teacher but also by the students on the front of their maths folders as a “flight path”. That way students can, in theory, see their 5 year progression in mathematics.

Now, there are lots of potential problems with this approach (students won’t have met some topics, the GCSE papers are readily available online), but there are also lots of potential benefits. Not least of which is that students will be exposed to a whole range of topics, which very much falls in line with the rationale behind the Revision section on each homework. And, of course, we have predetermined, objective grade boundaries to use. And in terms of topic specific practise, so long as our fortnightly homeworks and weekly diagnostic question quizzes are good enough, then we should be fine.

Of course, we now have the additional problem that there are not enough Sample Assessment Materials for the new GCSE available to allow us to use these for assessments across all year groups. But, as Karen and I discussed this morning, there doesn’t seem to be too much wrong with using existing Foundation papers, as these are likely to be more accessible any way.

Something else worth mentioning is a fascinating approach adopted by William Emeny. In his excellent blog post here, Will talks about a special kind of assessment his school ran for all of year 7  which was designed to identify specific gaps in knowledge that would then hinder students’ progress further on. Just look at the wonderful data he got back from it:

This, I believe, has huge potential. Especially in a world without levels. Moreover (and this will come as no surprise to anyone reading this), I believe you can do something similar using Diagnostic Questions. For example, look at how I can compare the understanding of one student to the rest of the class:

And I get the extra benefit of seeing exactly which questions she went wrong on and the reasons she put:

Or isolate a specific cohort, such as 7 to 11 year olds:

But of course, there is still the need for paper based assessment! Deary me, this is tricky!

So, as you can see, we are not really there yet with our assessments. I would be fascinated to here how other schools approach this issue.