Home » Main » The Next Generation

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 8 other subscribers.

The Next Generation

Borg
Next Generation CompSci student

The new GCSE specifications for Computer Science have now been submitted to Ofqual by the four main exam boards, each with their own take on the 80 / 20 coursework (aka non-exam assessment, aka controlled assessment) split.

At first glance they seem to be a step-up from the existing qualifications and that’s understandable based on the demands of the new A-levels and KS1-4 Computing curriculum. Subject content has been shifted down a Key Stage. Where basic programming skills have been taught at KS3 this now finds itself down to KS2. Obviously the new GCSEs have to reflect this and dove-tail into the more challenging A-levels.

The assessment of these qualifications seems to be the main deciding factor as to which board becomes most popular and there are quite a variety of different models to choose from. Comparing these qualifications side-by-side helps understand the differences:

Assessment comparison

Exam board Exam Non-exam assessment
AQA Paper 1: Computational thinking and problem solving
Written (‘programming’) exam 1h 30 (40%)

Paper 2: Written assessment
Written (other theory) exam 1h 30 (40%)

 

Solve a practical programming problem
Report, 20h (20%)
C#, Java, Pascal/Delphi, Python, VB.Net
Half the marks from code listing and explanation
Edexcel
Component 1: Principles of Computer Science
Written (other theory) exam 1h 40 (40%)
Component 2:Application of Computational Thinking
Written (‘programming’) exam 2h (40%)

 

Component 3: Project
Report, 20h (20%)
Python, Java, Pascal/Object Pascal, VB.NET, C derived
Approx. 1/3 of marks for code
Eduqas Component 1: Understanding Computer Science
Written (other theory) exam 1h 45 (50%)
Component 2: Computational Thinking and Programming
On-screen (‘programming’ in Greenfoot) exam (30%)

Component 3: Software Development
Report, 20h (20%)
Basic derived, C derived, Java, PHP, Python, Pascal/Delphi
Approx. 1/3 of marks for code
OCR Component 1: Computer Systems
Written (other theory) exam 1h 30 (40%)
Component 2: Computational Thinking, algorithms and programming
Written (‘programming’) exam 1h 30 (40%)
Component 3: Programming project
Report, 20h (20%)
“A suitable high-level language”
Half the marks from code listing and explanation

A similar structure for most but the Eduqas on-screen exam stands out, even more so as it is fixed in Greenfoot. I like the practical exam at KS5 for A-level but I’m not entirely convinced that it will work at KS4 especially with larger class sizes. It also begs the question: is this the best way to test understanding of problem solving skills if programming ability is already examined in the NEA?

I like that the AQA and OCR NEA awards so many marks to the production of successful code. From past experience both students and I have had enough of report writing as a means to prove that someone can program (not to mention the added marking workload). Report writing often seems artificial and is a burden outside of the natural way these solutions are developed. It seems that the AQA requirements for documentation seem the least onerous option. The OCR version is rooted in the previous specification and I can’t help feeling there is a lot of paperwork involved.

Verdict: Eduqas offer a bold alternative but AQA reward the ability to produce working programs with minimal paperwork.

 

Subject content

All qualifications are based on a core specification provided by the DfE. The level of detail varies from board to board. In much the same way as their A-level, AQA have painstaking amounts of detail whereas OCR seem to skimp on the details. In my mind as a teacher this is a bad thing. I would want to know what exactly pupils are going to be tested on and OCR seem vague in this regard.

In terms of coverage all of the exam boards have most of the same content. The things that stand out for me are:

  • Eduqas is the only board to specify that pupils need to learn HTML
  • AQA use the term Structured Programming above the expectation that sub-routines are used and understood
  • Edexcel don’t mention Low-level programming / Assembly language or XOR
  • Eduqas are the only board to mention RISC and CISC

Verdict: AQA and Eduqas provide teachers with plenty of detail about what to cover but you may have to wait for past papers to see what OCR want your pupils to learn.

 

What everyone else is doing!

Lemmings
To boldly go where everyone has gone before…

Teachers are like Lemmings! Where one goes the rest seem to follow. The undisputed champion of GCSE Computing qualifications is OCR. They started the ball rolling and they are by far the most popular exam board for the subject. There are numerous resources and CPD courses supporting the running of this qualification and this will only benefit those teaching the new specification.

There will be gaps however, and with all four boards offering the same qualification with the same structure this bias may change over time. In the short-term though you can call on the support of your fellow teachers to see how the OCR specification works.

Verdict: Most teachers currently use OCR so there’s plenty of support out there to help you get started.

 

Summary

I favour the AQA specification due to its level of detail and the fact that their NEA rewards the most marks for producing a solution without having to document every last step. That said, it’s great to that there is so much choice which can only benefit teachers and pupils alike. With so many viable alternatives available it certainly helps to keep the exam boards honest!


4 Comments

  1. Thanks for producing this summary. It’s perhaps a bias of mine, having done AQA A Level and OCR GCSE that the AQA SAM looks very ‘A-level’ like to my eyes. Some of the tasks in all the SAM looks really challenging and some of my C/D borderliners are really going to struggle.

    The WJEC/Equqas Paper 2 intrigues me. I love Greenfoot as a platform for Java, but fear that the level of challenge – again, thinking of the C/D borderliners – is quite tricky. But then starting with Python and moving onto Java might be a good way to more explicitly force students to consider subroutines and data types than is the case with just using Python alone. The issue of introducing OOP to 14 year olds, though – that’s a challenge right there.

    I think I’d prefer the WJEC offering as a student, but then I know I’d be towards the top of the curve (given where I’ve ended up). For those not so far round I think AQA/OCR look to be the most viable alternatives.

    • Thanks Mark – glad it’s of use!

      I agree with you that the AQA SAM seems tough. I have found weaker pupils can produce a solution that works in a fashion but struggled with the paperwork through lack of quality or perhaps motivation through the onerous requirements. I don’t necessarily think the skills required are different but perhaps the presentation of the task is daunting.

      Ultimately it boils down to which course suits the pupils in your class better.

  2. I’ve plumped for OCR. I’m not a complete fan of the specification, but it’s what I want because I can develop from the GCSE and it plans for that progression. Couple of things wrong in the report:
    OCR does specify you have to learn HTML, CSS and JavaScript
    The exams in OCT are 2 hrs 30 each

    I do quite like the flexibility of OCR, meaning I can cover the content and it allows me to throw in “other stuff” like PHP, AGILE and scrum development, which is think is important for the kids to learn. But having never taught AQA or EQUAS I can’t comment on their specs.

    • Thanks for the feedback Ben. I think you are looking at the A-level specification whereas the post compares the GCSEs. I agree though that flexibility is a strength of the OCR specifications. I would just worry about missing something as seen with the recent OCR paper that asked about opcodes. It wasn’t specifically mentioned and a teacher might use the term operator instead.

Leave a comment

Your email address will not be published. Required fields are marked *