How we could reform assessment and certification

[Originally published on on 5th November 2015]

In my recent post ‘Why we need to reform assessment‘, I outlined a number of issues which give me concern over the assessment of SQA National 3-5, Higher and Advanced Higher courses, introduced as part of the delivery of Curriculum for Excellence.

Whilst there may be many teachers who would wish for a return to the simpler assessment arrangements of the Intermediate 1 & 2, Higher and Advanced Higher qualifications of the ‘Higher Still’ era, which have been replaced by the CfE courses, I feel that one of the major shortcomings of CfE is its failure in doing anything to fundamentally change the nature or the purpose of assessment.

The new system, as with all of its predecessors, places almost the entire value of the certified qualifications on the terminal summative assessment of the course – the exam. Though some courses have significant elements of coursework, and many include an extended project or research task, still the majority of what our students, schools and staff are judged upon happens in a narrow time frame of a few hours at the end of almost an entire year of study. The results of these high stakes assessments supersede those for the individual unit assessments completed during the courses studied, rendering them effectively worthless.

In addition to this, there is the issue of the complexity involved in marking and recording the results of the unit assessments (which I outlined in ‘Why we need to reform assessment‘) which makes it difficult for students to understand whether they have passed units or not and increases the administrative burden on teachers.

In order to overcome these issues I propose the following changes to the methods of assessment and certification.

Unit Assessment via online e-Assessment

The SQA currently makes use of its own system for online e-Assessment, SQA Solar, for a range of courses across Nationals, Higher, HNC etc. Centres and candidates have unique, secure logins ensuring security of the assessment, and the system allows assessments to be scheduled at a time when the student is ready to be assessed.

This system could be expanded to incorporate all unit assessments in all subjects at all levels, and could be set up so that students performance could be recorded against the many individual criteria necessary to achieve a pass in a given unit. Any reassessment required could automatically be tailored to the specific areas not achieved at the first attempt. Given a sufficiently large bank of assessment items, or a sufficiently adaptable format, allowing numerical data to change for calculation-based questions (as it does on Heriot-Watt University’s Scholar VLE), it might be possible for students to make multiple attempts at assessments until the required standard is reached.

As the system is fully automated, this would free up teachers’ time for teaching and supporting their students learning, rather than using it for the bureaucratic administration of data. It would also reduce the ‘data chase’ required to ensure that SQA data is kept up to date on school MIS systems for transfer to SQA systems.

‘Points’ allocation and certification for internally assessed components –

Most courses have individual unit assessments which must be passed by students in order for them to achieve a grade in the final examination. Although these unit passes are included on students’ certificates, there is no explicit value placed upon them in comparison to the exam grade achieved. By allocating all components of all courses a number of points at the relevant SCQF level, students could potentially build up points across a number of courses whilst being able to choose whether or not to sit the final examination. This would reduce the ‘high stakes’ nature of the final examination, and allow for students, departments and schools to be judged and compared over the full range of their performance.

Points allocation for units could be based on the ‘size’ of the units, whilst exam grades could be allocated points determined by the band of pass. In my own subject, Physics, for example –

N5 – points awarded at SCQF level 5

3 x units, each with 10 points = 30 points,

Exam grade bands – A1 = 30 points, A2 = 25 points, B3 = 20 points, etc

Higher – points awarded at SCQF level 6

2 x full unit, each with 10 points + 2 x half unit, each with 5 points = 30 points,

Exam grade bands as for N5

Revision and separate certification of assessed course ‘Added Value’ units and ‘Assignments’ – 

Many courses have an internally assessed ‘Added Value’ unit, which at N4 has to meet every one of a significant number of individual criteria. Teachers are allowed to provide feedback to students in order to modify their submissions so that these criteria can be met.

The equivalent component of most N5 courses is an externally assessed ‘Assignment’, a formal report which is completed ‘under close supervision’ after a period of research which may include practical experimental work. Though guidance is given to students from their teachers, no feedback may be given on the report produced which is sent to the SQA to be assessed. The final mark for the assignment, given out of 20, forms a small proportion of the final score and hence the final grade.

These arrangements make it much more demanding for an N4 student, who may find the task much more challenging than most N5 students. A poorly completed N4 AVU would not meet all of the critera, resulting in the student not meeting the requirements of the unit, and subsequently not receiving an overall award for the course. A poorly completed N5 Assignment carries no such penalty, and would simply give the student a lower final score – without denying the student an overall pass.

Revisions should be made to the assessment of AVU tasks to make them fairer on the students. Perhaps an AVU could be consider to have been passed if a significant proportion of the criteria for the task, say 10 out of 15, were met by the student.

In addition to the significant differences in the assessments of these equivalent tasks across SCQF levels, AVU and assignment tasks are often very similar in related subject areas. This results in significant duplication of effort and repeated assessment of skills across a number of a student’s subjects.

By assessing these tasks on a skills basis, rather than within subjects, a single AVU or assignment could be completed by a student studying more than one science, or social subject. Students could choose which subject or subjects their assignments could cover, potentially allowing more meaningful, challenging, inter-disciplinary work to be undertaken. Though this might make the assessment of students’ reports more complicated, it might offer an opportunity to make the assessment criteria more flexible, as they are for the Baccalaureate qualifications undertaken by some students in S6. If nothing else, a reduction in the number of these tasks would significantly reduce the workload on students and reduce the SQA who have found it increasingly difficult to recruit sufficient markers for these tasks since their introduction.

I recognise that these proposals would require significant change to our current systems of assessment and certification, and that the Scottish teaching profession has experienced unprecedented change throughout the development and delivery of Curriculum fro Excellence. I further accept that one of the main reasons for avoiding radical change in the exam system has been concern that parents, employers, colleges and universities, might not fully understand the significance of new qualifications. In reality, it could be argued that these groups don’t fully understand the significance of the current qualifications system, and haven’t done so for a long time, if they ever have at all.

On a superficial level, it is easily understood that a student with an ‘A’ grade in a qualification is in some way ‘better qualified’ than another with a ‘C’ grade in the same subject, and that a student with five Higher passes is ‘better qualified’ than another with three Highers and two National 5s. But unless one has recently studied a course, or taught it, there is little chance of understanding what knowledge and skills are really involved gaining such a qualification, let alone how that qualification compares with other subjects or other levels.

It is often argued that we need these qualifications to allow universities to choose between applicants for places on their undergraduate courses. Without wishing to belittle this assertion, it does bear comparison to the ‘Sorting Hat’ in the Harry Potter novels – e.g. ‘AAAAB’ at Higher being the minimum requirement for a Law degree (Slytherin?). Increasingly, however, universities apply their own assessment requirements (BMAT, UKCAT exams), conduct entrance interviews, or consider applicants on the broader indicators of their personal statements, reducing their reliance on the crude measurement of ‘ability’ given by exam results alone.

In many ways the awarding of badges by organisations such as the Boy’s Brigade or Scouts to indicate the achievements of their members is a much more understandable form of accreditation. Indeed many professional and vocational qualifications are already ‘badgified’ in this way using industry standards, against which ‘badges’ are referenced and accredited. Mozilla, the organisation behind the Firefox we browser, support such a system for teachers to award ‘Open Badges‘ to their students using ‘open standards’ – where the criteria for which the badge is awarded are embedded as meta-data and awarded digitally. These badges can be electronically attached to a student’s digital profile via their blog, Google or other online account, and shared with prospective employers, colleges and universities.

Some work has already been undertaken by the SQA to develop this approach to accreditation, outlined in this press release from 2013, with small scale projects being adopted by some FE colleges, including Borders College, for accrediting both the work of students and staff CPD.

Open badges may not solve all of the short comings of our current system, indeed other, better systems may be in use elsewhere, or currently under development. Such a system, if combined with students’ unique Glow account could potentially stay with them throughout their schooling and beyond, perhaps even following them beyond further and higher education and into employment. The development of such a ‘Scottish Learner’s Account’, integrating assessment, certification and the accreditation of skills could form the foundation of a truly radical approach to these issues upon which students at all stages could build throughout the ‘Lifelong Learning’ that lies at the heart of the Scottish Government’s ambitions for the future of education.

1 thought on “How we could reform assessment and certification

  1. Paul Cochrane

    Does Drew Burrett realise what he has done? How dare he take the obfuscatory and supplant it with clarity and transparency? I would go so far to venture that, unlike many in CFE, he actually knows what he is writing about!

    Why we need to reform assessment‘ is the first clear attempt at the necessary de-spaghettification of the current assessment arrangements for the Sciences. Drew Burrett’s critical analysis of what is inequitable about N4 AVUs in comparison to N5 assignments is straightforward and correct. Once an N5 student submits their final assignment, it is out of my hands as to the mark allocation achieved. Once an N4 student submits their AVU, I step onto the draft, submit, re-draft, re-submit treadmill until, like a nervous assassin, the pupil accidentally hits their target. This must stop. Indeed, Drew hit upon the thoroughly logical idea I also suggested during the consultation period – why not make it an IDL project? According to our recent lead inspector, that is exactly what HMIE proposed but the SQA rejected. One assumes that the entry fee income had nothing to do with this approach being adopted but it certainly does not suggest a familiarity with the aims of CFE and transferrable skills!

    Drew also hints at the opportunity for the child (personalization and choice) maybe doing more than one assignment if they wish to build up credit. Perhaps the hide-bound regulations in the Sciences that pupil work must be at N5 chemistry level could be removed as well as, to be quite frank, the quality of the course content does not allow an exciting exploration of Chemistry. This could mean doing a generic assignment graded according to the level of response. It could also be argued that the AVU/Assignment could be treated as a whole Unit and allow this to be partial experimental findings and paper research with a final graded mark. If it was a genuine IDL collaboration could this not attract more credit? Pragmatically though, this would require a cull of the content and volume as pedagogy is struggling for variety in the Senior Phase rush to examination.

    Being a good teacher, Drew also suggests some solutions (remember them?). I heartily endorse his idea of e-Assessment for Units. Trying to mark, record and track pupil achievement under the current scheme is an absolute nightmare for staff and pupils alike. Who would have thought that the day would come when the teacher pupil interface would have resulted in,

    “Have you marked our NABS?’


    “Have I passed?’

    “I’m not sure yet.”


    “Well, you’re borderline KU so that will need to be remarked by a colleague, but you missed out on two skills.”

    “So I have failed?”

    “No – I’ll have to go back and check if you passed those skills before.”

    “So I had those skills 3 months ago but now I’ve lost them?’

    “Looks like it.”

    “So I have failed?”

    “No – you may have passed.”

    “Erm…do you know what you’re doing?”


    I would depart from Drew on one matter. He seems to suggest a never-ending series of attempts at passing but I’d restrict that for practical and credibility purposes. I would also go further and emulate Universities. Why can’t the Unit assessments be more challenging and graded A-D (or maybe use Drew’s points idea)? If a student demonstrates a consistent performance at B for example, offer them a B course award and exemption from the final examination or give them the possibility of an attempt to upgrade at the normal diet. Using Drew’s suggested automation, one may find verification, moderation and other needless catalysts of bureaucracy leaving palliative care and seeking their reward in a better place.

    By having these Units pre-checked for skills, a bank of achievement could build up towards other awards such as the Duke of Edinburgh, Young Scot and the dreaded ‘Bacc!’ Drew expounds beautifully on University entrance examinations and other non-standard achievements. I feel unqualified but if the SQA could verify a standard for them and each pupil had an e-portfolio….just imagine, our assessment and certification could become integrated, flexible and supportive.

    In conversations at work over the last few months, many of us have noticed a bizarre unintended consequence of CFE. For decades, HMIE bemoaned the ‘marking time’ of S1-2. Their second biggest gripe was the two-term dash in S5-6. The CFE solution? BGE is now S1-3 and the two-term dash has infected S4.

    Drew has captured a set of suggestions that make sense and would allow professionals to concentrate on pedagogy and effectiveness. This seems to go with the principles of CFE but against the practice of it. Can we get Drew and a few enlightened practitioners taken out and shot….into a position of influence to revamp our current shambolic circumstances?

    Yours sincerely,

    Paul Cochrane
    Science and Chemistry Teacher.”

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.