Students
Teacher survey
HE STEM: MM&PS

Think Pieces

This think-piece is one of a series intended to raise issues resulting from the Transmaths series of research projects. Each think-piece is focused on a major theme that emerged from the research and is designed to be brief and provocative. More details of the research underpinning the ideas presented here can be found in the publications database. We invite you to join the debate if you wish (details of how to do this are given below).

Assessment

Why are the outcomes of public examinations so uninformative for learners, teachers and policy?

It is striking that the huge amounts of information obtained by Awarding Bodies in assessments of 14-19 year olds yield so little information of value to the education system and in particular teachers and learners themselves.

Given the wealth of item level data on whole populations of GCSE and GCE cohorts, for instance, together with the accumulation of individual data longitudinally through the years in national databases (eg NPD), one could expect huge amounts of information to be available every year. As a minimum we might expect:

  1. Formative assessment information on each individualís profile of performance and potential competences (e.g. identifying which criteria the individual learner showed some success with) rather than just a summative grade, so that future educational decisions by both teachers and learners might be better informed;
  2. Information for policy, for stakeholders, and for teachers and learners about which criteria are typically met by students at a given level, and descriptions of the common errors students at various levels typically make, in which areas of the syllabus etc (eg the work on algebra of projects like the Evaluating Mathematics Pathways project);
  3. Improved assessment year-on-year as assessment designers are informed at a very detailed level how their assessment items are tackled by learners.

University and Sixth Form College staff in interviews told us that the grade (at GCSE or GCE A level) a student attains, for instance, tells them little about what the students know and can do at the next stage (A level, or university respectively). Surely the least that could be expected after all the expensive and time consuming examination processes which any individual student has negotiated is an estimate of what competences might be expected of them as an individual, and how this contrasts with a typical student obtaining the same grade?

Compared with PISA and TIMMS for instance, very little information is made public by the awarding bodies. These international studies have a significant impact on curriculum policy based on much less comprehensive data. The assessment leading to GCSE and GCE A Level captures significantly more potentially useful data on an annual basis being extensive in its coverage of the population, content, background contextual data, and so on. The use of the technology that the Awarding Bodies now employ makes data collection and dissemination increasingly straight-forward.

Why are the outcomes of assessment not more informative: what can be done to ensure they better inform teaching and learning?

First, the exams are not designed, or not well designed to yield this kind of information: the priority is to make reliable judgements of summative grades (e.g. for selectors to make inferences) only, not inform users of what this means in terms of competence. This could be changed, technically, if there were an interest at stake other than reliability of grades. Second, Awarding Bodies are increasingly shy of litigation. If individual reports are contestable, and the more information they yield the more contestable are the resulting reports, this places the Awarding Bodies at risk. This is especially true in relation to high stakes awards/examinations.

Third, the information needed is, in many cases, already collected but there needs to be a priority for its organisation and dissemination to be focussed on outcomes other than grades. It is time to make more serious demands of the exam system: we need to consider the outcomes we value and need and ensure that it is made available.

If you wish to add a response to this think-piece please email your thoughts to lawrence.wo@manchester.ac.uk