November 2018

To Test or Not to Test? That’s the Question

For many people testing is a symptom of a very sick education system because all our educational energy is taken up with covering the content of the test, developing test taking strategies and preparing the students for their exams. Where, in all that, is the education of the individual?

There are lots of different reasons to test. You can test to see which class a student should join (placement), or where they need help to improve (diagnostic), how much of a syllabus or curriculum they know (achievement), whether they have improved in an area (progress) or more generally – measuring a student’s wider ability (proficiency). All of these kinds of tests share three key aspects:
• Informative: they tell us something about the student that we didn’t know before, or that the student didn’t know about themselves.
• Objective: theoretically at least, everyone is judged by the same criteria.
• External validation: it’s not just us saying it.

Teachers can tell you lots of things about their students; if they are imaginative, creative, intelligent people, if some are better communicators than some others, if some are more introspective, if some have a great time in class and some of them would rather not be there at all, whose English is better etc. So why seek external validation? Many teachers say the results of external validation tells them what they need to focus on to help all the students achieve their goals. It also gives them the evidence they need to adapt the curriculum and informs the decisions they make about lesson objectives and classroom content.

Language Tests

Language tests are a bit like buying clothes: for some students the clothes will be too big, or too small and for some others the style will just be completely wrong.
There has been much concern about testing. It often appears to be used as a tool of government to bludgeon schools, educators and students into submission – after all where students are so focused on passing their tests they have little time left to develop critical thinking skills or to apply critical thinking to the world around them. Nonetheless, we seem to live in a world where it is necessary to measure ourselves against each other – to determine our worth to prospective employers by comparing our abilities in the context of grades and examination results.

Similarly, tests provide evidence for the claims that we make for ourselves. When we seek medical treatment we want to be sure that our doctor knows what s/he is talking about – education and qualifications do not necessarily make a doctor a “good” one, but they do determine a minimum standard that needs to be achieved.
The same applies to language testing. They determine a minimum standard that needs to be achieved.

The techniques of language testing in use at any time tend to reflect the view of language and the way it is used at that time. What is being tested and the kind of task or item type chosen as a means of testing can be expected to show the influence of current thinking on what language ability is and what exactly we are doing when we use language in everyday life.

The respectability of this evidence though, rests upon the reputation of the organisation administering the test and awarding the results. A candidate walking into a job interview with a certificate awarding B2 level competence from “X Language School and Coffee House” will have their ability more seriously questioned than the candidate with a certificate from a well-established and internationally recognized organization.

The main problem with tests is the disproportionate influence they have on everything that goes on around them – they are like the black hole at the heart of the educational galaxy, sucking everything and everyone inexorably towards them. Some of the factors that “washback” and “impact” studies (research that examines either the influence of a test on events that precede the test; or the wider consequences of a test respectively) have identified as being influenced by tests are:
• Government education policy
• Curriculum and syllabus design
• Lesson planning and content
• Teaching practice
• Course book selection
• Course book design & content
• Teacher education & development
• Student goals
• Student expectations
• Parents’ goals & expectations
… and this is not an exhaustive list!

Tests should be an aid to teaching and learning, not a goal in and of themselves. As things stand at the moment, my concern is that the majority of tests simply don’t function in the way they should. They don’t tell us much about the learners’ needs, wants or abilities – they only tell us how good the learner is at reproducing the information required by the test. Maybe the focus of testing is wrong. Instead of telling us whether students have passed or failed, they should tell us in which areas the students have excelled. In other words performance should be subject to a more nuanced analysis and not simply allocated to a general band. •