Exploring language assessment literacy with Professor Anthony Green

You are now well a well-established figure in testing. How did it all begin?

I suppose it goes back to when I was a kid taking tests, sometime wondering who makes these things and why exactly do they choose these questions. Early on in my teaching career, I got roped in to do a placement test and found it really interesting. It was a great learning experience, partly just working on a project with more experienced colleagues, and partly because I got hold of Arthur Hughes’ book Testing for Language Teachers (CUP, 2013), a very accessible book that really opened up the whole area for me. I then got more involved in assessment work, went on to do my MA at Reading, where I got to study with Arthur Hughes as well as Cyril Weir and Don Porter, then I went on to do my PhD, followed by a job at Cambridge Assessment. By that time, there was no escape!

So if you had to choose one key element, what still fascinates you now?

Like most language testers, I have a teaching background so the most interesting questions for me are about how testing and assessment [CB1] works for teachers, how they can help teachers to understand what they are doing better, and to get insights into learning processes. My interest now is not just in large-scale external tests, but also in how teachers can make better use of assessments in the classroom.

How much work do you think there is to do to improve assessment literacy in the classroom so that teachers are more at ease and comfortable with what testing and assessment really is?

It touches on teachers’ professional development and how far current teacher training really deals with assessment. We have an approach to teacher training that marginalizes assessment. In some programmes it’s just not there at all. When it is there, it tends to be left to the end of the course. If you look at the CELTA syllabus, for example, there are five different topics and assessment does not come in until the very end of the last topic. You learn how to teach, and then assessment is added on as an afterthought. For me, that is completely wrong. Assessment has to be part of everything that teachers do. Each module should include how you can get effective information from assessing your students that will help the learning process. As a teacher, you need to build a picture of where you are going with students. That means you also need to be able to decide whether the students have got there – or how close they are – which requires assessment. Without including this element, you are never going to have an effective teacher training system. When teachers understand how assessment works and how tasks can help to reveal what learners know or don’t know, they are better teachers. There is a need for assessment to be really embedded in teacher training.

What then should be the next steps for tangible improvements in assessment literacy?

From my point of view, one very important thing that teachers can learn from testers concerns quality. For my book, Exploring Language Assessment and Testing, I came up with the acronym PRICE (Planning, Reflection, Improvement, Cooperation, and Evidence) to convey to policymakers and managers the idea that you have to invest in assessment as well as pay attention to those five elements.

In planning for lessons, teachers often think about what’s going to be fun and what is going to engage the students. But they don’t think enough about what they, as teachers, will learn from what the students are able to do, building into their plans what the task will reveal about students’ strengths and weaknesses, and how the task can help them to see where to go next with the class. Considering the value of the task in the classroom and whether it actually reveals to you what the students were able do requires a process of reflection. Does it show how you can move their language forward? If not, how could you make it work better – the improvement element of PRICE. This can work at the level of casual question techniques to check for understanding, but also at more elaborate forms of assessment, such as portfolios – in fact anything we do to find out where students are in their learning. It works for feedback mechanisms as well: what did you say to the student? How did that help them to learn and were they able to make use of the feedback you gave them to improve their performance?

Successful reflection also depends on the two other elements: cooperation and evidence. Typically, we don’t collect enough evidence from the classroom – recording what we are doing or noting what went well or not so well – but we need this kind of evidence to support reflection and improvement. It is also unfortunate that teachers too often work in isolation: not sharing and discussing enough with colleagues on what we do in class and what brings success. This kind of cooperation can really help us to improve our practice.

How and where do we bring this into teacher training so it becomes an integral part of what teachers do?

I really think my performance as a teacher would have been more effective if assessment issues – such as constant quality control and the connection between what happens in the classroom and how tests work – had been introduced at the very beginning of my initial training. When I completed my course, almost nothing had been said about tests. It was a completely absent element. In my first teaching job, the first class I taught was a First Certificate in English (FCE) preparation course and I knew next to nothing about how tests work. I fell back, as many teachers do, on looking at example tasks and focused on practising those, giving my students endless rather boring practise on the test. I think I did that because the thinking behind the test was not clear to me: how the test was supposed to represent real-life communication. That again comes down to a certain lack of planning and reflection in the teacher- training process. Training needs to include assessment and how it can help improve students’ learning from the very earliest point. I believe motivation as a teacher is enhanced by seeing that you are successful in your work. If you simply feel you are just delivering what is in a test or coursebook, that is not very motivating.

But if you can trace a process of the students developing new abilities, then that is much more inspiring?

There is a tendency to try to separate classroom assessment and testing into different worlds. Some testers will say that assessment is just ‘good teaching’ and teachers will say formative assessment is different from large-scale testing. I think that testing is an educational experience just as teaching is, and both sides can learn from the principles that govern each domain. I believe educational testing would be improved if we could make taking tests more of a learning experience. That is, if test takers could leave a test feeling that they had really learned something from it. Equally, teachers would benefit if at the end of a lesson they can say they had really understood something new about their students, taking more of an assessment perspective on what they have learned about their students.

So if we teach only to test and we see that learners only want to pass the test, they end up with a piece of paper but no tangible learning, which makes your point about teaching an exam class being demotivating very important…

Yes, of course, learning something about a test – being aware of how the test-taking process works and what you will be required to do – is important, but it is only tangential. There is a much bigger picture behind any good test and test developers don’t always communicate as well as they might about why the test is the way it is. When a learner takes, for example, FCE Speaking, there is a paired activity where they speak to the other candidate and arrive at a conclusion… so why is that activity on the test? What is it supposed to represent? The handbook for teachers will tell you quite a lot about what the task is, but almost nothing about why it is there or what kind of real life language it is supposed to represent. I think it does represent aspects of real life language use – an informal conversation discussing an issue of current interest. Test developers could do a lot more to explain that to test takers and teachers, and to explain why including a task like this on the test counts towards revealing how learners might perform in the real world. That would help teachers to articulate to the learners that there is life beyond the test.

Are formative assessments for children up to 16 years of age a valid and reliable form of assessment?

We cannot assume that the use of any kind of assessment is valid and reliable, but must collect evidence and use that to form a judgement. Of course, you could look at the kinds of assessment being used and you could apply some of the PRICE principles and establish that it shows whether a student can do things with the language, whether you are able to intervene, where necessary, to help them learn. If you could show that it did, it would be valid for its formative purpose. If, for example, you find out that your questions are revealing nothing about the students’ language abilities, then it would not be valid. The causes would need to be investigated. There can be no assumption that any kind of assessment is valid. You have to look at whether it is working to achieve the purpose for which it is set.

Should assessment and testing be more fun?

More fun for who? (laughs) I think this idea that it is isolationist and impenetrable is a little harsh. The foundational book, Robert Lado’s Language Testing[CB2]  (1973), is very clearly a book for language teachers and intended for language teachers to communicate to learners and help them to develop language tests. It was not isolationist in that sense of being for specialists: it was for all language educators.

Maybe language-testing specialists have not always been very good at communicating with non-specialists about the subject, but I don’t think it’s impenetrable. After all, most language testers are teachers in their original training and we have managed to penetrate it – so it can’t be that hard! But I’d agree it’s not usually great entertainment in the way some teacher training can be, so, yes, I guess it could be made more accessible. There should be more exchange between teachers and testers… both have things to say to each other. Whether it is fun or not, better communication would be helpful.

Tests themselves should be as painless and as engaging as possible, but for many people they can be traumatic. I remember in one of the first placement tests I developed for a language school, even though the results counted for very little, one of the students taking the test burst into tears.  If we can make tests less anxiety-inducing, that would be good – and the more we can do that, the better.

It should be more approachable, and maybe we should avoid using terms such as isolationist and impenetrable to prevent it from becoming a self-fulfilling prophecy.

Indeed, we need some communicating to do and some clear ideas to work on. Thanks for taking the time to share your thoughts with us.

Neil Bullock