Moderation is the means of ensuring accuracy and consistency in marking or assessing students' work or performance.
Moderation is an important part of the assessment process - to ensure that standards are maintained and that elements of subjectivity do not bias the judgements of assessors. Faced with the task of marking an assignment (or assessing a student's performance), it is a temptation for the informed assessor to subconsciously take account of extenuating factors that may have affected that student's performance. Whereas there ought to be a place for consideration of such circumstances, it is important that these are openly and transparently acknowledged.
Moderation is also an important part of the learning process for those involved in the delivery and assessment of learning, in that it allows them to engage in dialogue about standards with fellow professionals.
Unfortunately, moderation is time-consuming (and involves extra effort) and can be perceived as an additional and burdensome layer of bureaucracy. The NVQ Framework, for example, is moderated by a system of internal and external "verifiers" whose primary task is that of ensuring consistency of assessment standards.
Moderation in Practice.
The way in which moderation is carried out depends on both:
The type of assessment may relate to:
Unfortunately, there are no clear-cut boundaries.
Achievement may be reported as:
Some moderation processes require "second marking" of all assignments. This may lead to a considerable amount of discussion in order to negotiate an agreed final mark. Whereas this is time-consuming, it does ensure a high level of accuracy and consistency.
Moderation of method (a) (pass/fail) is perhaps the easiest to implement - although there can still be variations in the process. A suggested method is for the Assessor to set out all work in "order of merit" and then to indicate "where to draw the line" (Obviously, this can only be done where there is a physical product or assignment etc.) This means that the Moderator can focus attention specifically on those cases around the "borderline" with only cursory attention to the remainder.
Moderation of method (b) (grade/band) increases the numbers of "borders" - which can lead to the process becoming more unwieldy.
Moderation of method (c) can be the easiest (where it is applied to paper-based tests that have clear-cut correct answers) or it can be the most difficult (where it is applied to assignments whose assessment may involve an element of professional judgement). Effective assessment of assignments in this way requires very specific assessment criteria and the process of moderation can become sidetracked on interpretation of these criterion descriptors.
Consideration also needs to be given to the size of the sample submitted for moderation and the way in which this sample is selected. This will, of course, depend on the total number of assessments being moderated, the number of assessors and the requirements laid down by awarding bodies. A sample size of less than 10% might be considered too small (unless the total number of assessments makes a larger sample size impractical - as for some national examinations).
Methods of choosing the sample can also vary. For a sample size of 20%, for instance, one could:
This process can be further complicated where students submit work for different modules (where it makes sense to moderate each module separately) or where students have been offered choices regarding the form of assessment.
Of course, the whole process is additionally complicated by the fact that there may be multiple assessors - possibly operating from different centres - each assessing different modules - who may also be acting as moderators - possibly of their peer assessors. It's a challenge!
Useful website - Standards & Levels of Achievement