Let's begin with a multiple-choice question. Which of the following best describes your definition of the term "test" when applied to education?
a) An out-of-date token of an exam-factory school model that is damaging our students' learning;
b) A blunt tool that is used to clobber teachers at every available opportunity;
c) A vital tool for education;
d) You are too surprised by option c) to give an intelligent response.
Get teachers (or students, or parents for that matter) to answer this question and the answers are likely to be variations on options a) and b). This is because it is seemingly impossible to distinguish the idea of testing from the deep-seated prejudices we have about our high-stakes exam regime.
I would argue, however, that this is exactly what we have to do. Testing doesn't just measure our students; more importantly, it is an effective tool for improving learning.
The benefits of learning through testing are many, even if they seem counter-intuitive at first. For example, rather than leave a test until the end of a topic or scheme of learning, research suggests that we should start with one. It also suggests that we should test more, using regular low-stakes assessments such as quizzes or short questions, to harness memory.
A key text in this area is Henry Roediger and Jeffrey Karpicke's 2006 research study Test-enhanced Learning: taking memory tests improves long-term retention (bit.lyTestEnhanced).
This describes the positive impact that taking a test has on students' ability to retain information. But research has been coming to similar conclusions for many years. Back in 1917, psychologist Arthur I Gates explained the beneficial impact of testing, or "recitation" as it was termed, on learning (bit.lyRecitationStudy).
More recently, Mark McDaniel et al's 2007 paper Generalizing Test-enhanced Learning from the Laboratory to the Classroom (bit.lyTestLabToClass) showed that taking a test on a given topic proved more memorable than simply revising the material.
This evidence should force us all to reconsider the worth of regular low-stakes testing when we are designing a curriculum.
Time for reassessment
First, we need to start with a little rebranding. There is a stark difference between high-stakes exams and low-stakes testing for learning. Testing in this sense could involve students simply being quizzed on taught topics (a "quiz" doesn't sound as daunting as a test). Alternatively, it may involve students having to map out what they know in a visual way without the benefit of their notes. In essence, whichever method you use should draw upon the difficulty of retrieving information from memory but without the strain of a planned exam.
The timing of tests is important, too. As mentioned earlier, taking a test before beginning a topic can be very effective. In their study The Pretesting Effect: do unsuccessful retrieval attempts enhance learning? (bit.lyPretestingEffect), Lindsey E Richland et al found that even when students bombed on a test before studying the topic, they still learned more effectively in the long term.
The evidence suggests that tests can prime students for learning in a positive way. Getting it wrong can tease out initial misconceptions, or guide students towards the ideas they need to remember when they come to a given topic.
There are, of course, hurdles to overcome in this approach. The mere mention of testing will trigger fear and loathing in many teachers. Under the all-seeing eye of an inspection body that seeks rapid progress, the idea of students struggling on a pre-test seems unacceptable. But there are benefits to "learning ugly" and we must share and communicate these better.
Another consideration is that such an approach could potentially lead students to feel they're failures. To counter this, teachers must gain the trust of their classes and explain the benefits of what they're doing. We can, and must, recharacterise failure and struggle as essential elements of progress, because they are fundamental prerequisites for effective and memorable learning.
Here are three strategies to get you started on using testing as a tool for learning:
1 Take the pre-test
As already stated, a pre-test can be extremely effective. However, you have to ensure that it's used positively. Follow up the test by identifying the gaps in students' understanding, giving them precise feedback on their misconceptions and using this assessment to plan meaningful future study.
You can also fix any potential damage to pupils' self-confidence by highlighting how far they've moved on from the pre-test as you progress through a topic.
2 Questions, questions
Assessing your students' understanding needn't look or feel like a test. It can easily be transformed into an active process, in which the pupils ask the questions.
Once more the "testing effect" is harnessed at the beginning of the learning process. Introduce the topic to be studied and then get students to come up with questions about it.
The crucial next step involves pupils organising the questions coherently. You could have a question wall to help them find patterns in what they've asked and to categorise their questions according to different criteria - which one was asked most often, for example.
The final step is to get students to answer the questions and share their understanding. This process highlights prior knowledge and enables you to modify planning and teach more effective lessons.
3 My favourite mistake
Back in 1998, Sheryl Crow sang about her "favourite mistake". Which of her famous ex-boyfriends is the subject of the tune is unclear, but you can harness inspiration from this pop gem when giving purposeful feedback on tests. Whether it is a past exam question or a weekly quiz, draw attention to your students' misconceptions by making errors visible.
There are a variety of tools for doing this. Circulating the room and selecting your favourite mistake is a lo-fi option. Getting students to write their own mistakes on cue cards or whiteboards works just as well.
For the technologically minded, displaying a mistake on a visualiser for the whole class to see - and then exploring the misconception - does the trick brilliantly. You can unpick the error and right the wrong.
This can have the doubly positive impact of normalising struggle and failure as necessary steps in the learning process, while clarifying the most common misconceptions being made by our students.
Alex Quigley is director of learning and research at Huntington School in York. You can find him on Twitter @huntingenglish