To a busy teacher, producing good assessments can be a time consuming and daunting prospect.
When it is done well, an assessment can provide you with information that it isn’t possible to find out in any other way.
We have all heard of teachers saying “I’ve set this test to be really hard, no-one will get this right”, but assessment isn’t about you as a teacher pitting your wits against your students. Assessment plays an important part in helping you find out what your students know and, importantly, do not know, in order to better meet their learning needs.
When it comes to assessment, quality really matters, and the difference between good and bad assessment is huge. How can teachers understand what makes one assessment good and another bad? How can this understanding be used to make good choices about assessment?
Having a positive impact on learning
So where to start? It’s easy to become overwhelmed by the amount of research, evidence and pseudo-evidence surrounding how teachers can assess better.
In his blog post ‘Would you let this assessment into your classroom?’ Director of CEM Professor Rob Coe offered a 47-question checklist to point educators in the direction of appreciating quality assessment and making good choices.
Using this checklist as a starting point, here is a quick review of some of the fundamental questions to consider when constructing an assessment for use in your school without it being exhaustive (or exhausting).
1. What does the assessment claim to measure?
Strangely enough, this can be one of the most difficult questions to answer. It is important to take a step back and really think about what it is that the assessment is attempting to measure.
Clarity of purpose is crucial. For example, an assessment that is great for working out a predicted grade might not be good at helping identify pupils’ strengths and weaknesses.
2. Does your assessment measure everything within the subject, or are there gaps?
Some assessments are designed to assess everything within a particular topic area.
Spelling tests, for instance, can check a student’s ability to spell a set of new words learnt that week, but it is unusual for this to be the case. More commonly the assessment is a proxy for the complete domain encompassed by the subject.
In this case we have to select questions that test various areas of the domain. The larger the domain, the more difficult it will be to cover everything.
3. How well do the assessment scores correlate with other measures of the same thing?
Assessment professionals call this “concurrent validity”.
This refers to the extent to which the results of a particular test, or measurement, correspond to those of a previously established measurement for the same concept.
As the name suggests, concurrent validity relies upon tests that took place at the same time. This might mean assessing students around the same time. An example of this would be assessing a group of students with an old GCSE paper, and then assessing them again a couple of days later, using a new GCSE paper.
If your new test compares to a well-established test, you have good evidence that your assessment is working.
4. If you carry out the assessment more than once, do you get similar results?
Ideally, if you re-sit an assessment, you’d want the results to be pretty much the same as when you last did it. This is known as “test-retest reliability”.
If a student sits their maths mock exam paper twice – would they get the same results?
If the results are not consistent, this could suggest that the assessment is unreliable.
5. Could anything about the test takers affect the results?
Ideally your assessment will be as open possible to everyone who takes it, with no particular group put at a disadvantage.
Creating an assessment like this is difficult to achieve, but it is important to bear in mind when constructing the assessment whether it unfairly favours some groups over others.
It is important to think about whether the assessment is free from biases such as gender, social class, ethnicity, reading ability, cultural background.
6. For what ranges is the assessment appropriate?
Academic assessments are normally aimed at a particular age and ability range, both in terms of the language used and the content presented.
Amidst the recent examination reform, schools are having to create assessments without the helpful accumulation of past papers. Are GCSE students therefore being presented with questions that were actually intended for A-level students? Or vice versa?
7. Does taking the assessment or the results it generates have direct value to teachers and learners?
This is the big one. There are four key questions you should consider when looking at the value of the assessment for teachers and learners:
- Why are you asking your students to take this assessment?
- What will it tell you that you don’t know already?
- How will you use the information the results give you?
- Can you use the results of the assessment to directly impact the teaching and learning of your students?
The importance of assessment literacy
‘Assessment’, Professor Rob Coe emphasises, ‘should tell you what your pupils know, understand and can do; it has to support teaching and learning and it should not be used to measure teacher or school performance for accountability.’
CEM has been working with education professionals for over 30 years and there is clearly an appetite for teachers to have a good understanding of assessment, and to be ‘assessment literate’.
Helping teachers give every child the support they need
By Dr Christine Merrell It can be hard for even the best teachers to notice differences in the...
How much difference does going to school make?
By Professor Christine Merrell, CEM Director of Research Development. The contribution of schooling...
Inattentive behaviour in the classroom
By Professor Christine Merrell