Assessing performance or proficiency?

In their new book The Art and Science of Language Teaching, Lara Bryfonski and Alison Mackey, in their chapter on assessment, remind us that there is a distinction to be made between assessing performance and proficiency. For example, in an oral test, if a student talks about their family, does this tell you they can reproduce some rehearsed language (performance) or that they have an underlying ability to generate original utterances in a creative way (proficiency)?

Teachers of novices and pre-intermediate school students know that the majority of their learners often have little ability to extemporise when speaking. This is because they have not had enough time, input and interaction to internalise (‘acquire’) enough of the language system (grammar) to speak spontanously. The same is basically true of written production, although in this case students do at least have time to reflect on what they want to write.

So with less experienced students we often end up assessing performance, and with good reason. Students learn and practise sentences and paragraphs to reproduce during a test. In other school subjects we have few qualms about testing what learners have memorised  Indeed, in musical instrument exams, it’s all about performance. So why should we worry too much about this in MFL/WL lessons? In time, those students who elect to continue with an additional language for, say, a fourth year and beyond will gradually be able to be creative with language. At advanced level the assessment regime can reflect this by including elements where students are forced to think on their feet and generate new utterances, revealing that they have underlying proficiency.

So with our younger learners, including many of those who are taking the GCSE in the UK, it is right that we allow students to ‘perform’ in oral and written tests to a large extent. This lets students show off what they have learned, what they can do. If we did not include this element of rehearsed performance, many students would flounder and be discouraged in their pursuit of learning a language. To what extent in your own setting do unit tests, school exams and standardised tests assess performance or proficiency?

At levels up to GCSE (CEFR A2 more or less), we can in fact combine assessment of both performance and proficiency. At GCSE we have always done this in the speaking test by letting candidates do a short presentation or give answers to predictable questions that have been practised before, while adding some unpredictable elements. The new GCSEs (first teaching 2024) allow for pre-learning of photo card conversation questions which are largely predictable. The same is true of the role-plsy tasks. Skilled teacher-examiners will encourage weaker candidates to show what they have rehearsed and give opportunities to more skilled students to make up new language on the spot (e.g. by asking unexpected follow-up questions).

For writing papers there will also be a strong element of pre-learning, which is also fine. Indeed, knowing how much weaker students struggle with written papers, we would be delighted of more students prepared more thoroughly and reproduced their knowledge under exam pressure.

Listening and reading are another kettle of fish. In this instance, we are always more likely to be asessing proficiency rather than performance. This is because the content will always vary a good deal, so learners have to deploy their various acquired skills to achieve success: distinguishing sounds and intonation patterns, recalling words/chunks and parsing the grammar to make meaning. This means that, overall, when you take into account a whole exam including all the four skills of listening, speaking, reading and writing, we can always say that we are assessing proficiency to a degree. GCSE success is, taken as a whole, a reasonable measure of proficiency, I would argue.

My main takeaway here, then, is that, rather than fretting too much over performance and proficiency, we should acknowledge the distinction, be aware of it, but perhaps focus on what students KNOW AND CAN DO, whether it be via pre-learning or spontaneously, on the spot. Our current exams in England do this pretty well overall, even if they can be criticised for other reasons. If a key goal is for us to engender self-efficacy in students it’s only fair that we let them learn, memorise and reuse known language.