Accentuating the Positive: Lessons for Future Assessment from the 2021 Grading Process
What can we take forward from schools’ experiences in summer 2021?
by Mr Neil Smith, CHS Head
The following blog was written for the Rethinking Assessment group and is also available to read here.
This summer, grades for pupils in England were calculated in a different way for the third consecutive year.
Unlike the ‘traditional’ model in 2019 based on performance of students in terminal exams/NEAs, or the Covid ‘1’ model in 2020 in which grades were based on teacher projections of what a student would most likely have achieved if they had sat end of course exams, in summer 2021 grades were awarded by teachers on the basis of assessed work which the students had completed or would sit in school before mid-May; an approach deemed to be necessary in response to the disruption to learning which had taken place as a result of Covid-19.
Aspects of the grading process and the impact which it had on grades have been strongly criticised, with particular attention drawn to grade ‘inflation’, the lack of a uniform approach across schools in how pupils were assessed and how grade boundaries were set, and concerns that pupils at independent schools appeared to disproportionately benefit from the increase in top grades awarded.
Whilst there is, to varying degrees, weight to each of these criticisms, the experience of this summer also resulted in schools behaving in a way which could have a positive impact on future assessment, if elements of the 2021 model were maintained in the future, creating an opportunity to reflect on the desirability of continued reliance on terminal exams as the primary method of assessment for students, especially at GCSE.
Here then, are a few thoughts on some of the lessons from schools’ assessment experience during the last year which are worth learning from:
1. Perhaps the most obvious lesson from this year is that using a process involving multiple assessments removes the tyranny of the one shot nature of traditional high stakes assessment. As Alison Peacock has written in a recent blog, exams are principally used to provide feedback, to consolidate learning and to enable future choices to be made, and a strong case can be made that a model including multiple assessments over time is more compatible with the purpose of exams than that traditionally used at the end of Years 11 and 13. Furthermore, such a model is less likely to replicate how students are assessed in other areas of study and assessment post-18 at university and the workplace.
2. The way that a number of schools approached assessments this summer also mitigated against the impact of exam board marking errors. Issues with examiner agreement on what marks should be awarded to a script have been well known for many years(1), and even acknowledged by Ofqual(2). In order to arrive at fair judgements, many schools built in extensive standardisation and moderation procedures to agree on marks awarded to individual scripts and to agree how far individual candidates’ had exhibited the qualities required by the JCQ grade descriptors. As a result, one of the significant effects of the TAG process in 2021 was that teachers appeared to gain greater professional confidence (and proficiency) in discussing what and how to assess, and what competencies they expected to see demonstrated at each grade level. Looking ahead, the proposals articulated by John Dunford provide a constructive approach to further improve the quality of assessment in a system using teacher judgment, particularly with regard to tackling potential accusations of teacher bias.
3. In turn, use of these grade descriptors also facilitated a mastery rather than quota approach to grading, with candidates graded on what they could demonstrate and not where they fell within a fixed quota. The use of descriptors to inform grading certainly proved easier to apply in subjects where criterion referencing was more easily applied, but their use across the curriculum should not prove an insurmountable problem.
4. Future use of grade descriptors by teachers could result in teachers providing better quality and more directed feedback to students, with teachers able to state specifically which skills, or indeed content, students needed to improve in order to progress to the next grade.
5. More responsibility for grading to school could also result in a more fundamental shift in what is being assessed, with schools able to gain greater autonomy on courses which they can teach and indeed the attributes which they wish to assess.
6. This summer highlighted what could be thought of as ‘secondary’ benefits to schools which are worth factoring into possible reform of the assessment system, such as maintaining pupil focus throughout the course of study, making better use of the whole duration of the course (rather than rushing to complete a course by the Easter of Year 11/ 13, then revising, then study leave).
A concern expressed by some advocates of exam reform is that the way in which a system based on teacher judgement was implemented could have done immense harm to the argument for reforming the examination system. Whilst the methodology applied was far from ideal, for the sake of future generations of students, the positive lessons from how it operated in practice must not be overlooked.
References
(1) How accurate are examiners’ judgments of script quality? An investigation of absolute and relative judgments in two units, one with a wide and one with a narrow ‘zone of uncertainty’ Tim Gill and Tom Bramley Paper presented at the British Educational Research Association annual conference, Heriot-Watt University (2008)
(2) Marking consistency metrics An update, Stephen Rhead, Beth Black and Anne Pinot de Moira, Ofqual (2017)