We need to talk about IELTS, and the case for testing English ‘in-subject’

‘English language’ testing is a critically important area, and also a highly sensitive one for English-medium higher education around the world.

An International English Language Test System (IELTS) score is tantamount to a passport for international students, and tantamount to a barrier to higher educations institutions wishing to increase their international student numbers. IELTS is now the only secure English language test verified by the UK government.

Tensions exist on the one hand between wanting to increase international student numbers, but at the same time vouching for the ability of those students to follow courses of study in English. Students themselves are often desperate to come to the UK, and in the majority of cases will no doubt assume if they have met a certain IELTS score then their English is good enough for them to study here. Lecturers too might want every student to have a top score (9.0 in IELTS), but have to operate on the assumption made by their institution that a score of 6.5 will be sufficient for their students.

However, is it correct to assume that IELTS is actually testing the ‘English language’ that we want students to have to succeed on their degrees? To ask the question slightly differently, is it possible that IELTS tests a different kind of English, one which may in many ways appear to be the ‘English language’ that students may need to succeed, but is in fact so only in appearance? In other words, what if the underlying subject context were so important to the use of English, that the ‘English language’ was unique according to whether students were studying computing, nursing or history and so on? If that were the case, then the English that students need to succeed on their degrees would be very different to the English that IELTS tests.

Our research shows that the whole approach to how we test ‘English language’, and our whole approach to how we research the validity of ‘English language’ may in fact be based on a false assumption. We test English based on the assumption that it can be taken away from its context, analysed elsewhere and tested separately to the subject. However, what we have found is that when lecturers in subjects such as Nursing or Design use particular words, they do so with their own unique subject based understandings.

Not only this, but each subject has its own key aspects that underpin and operate with the ‘English language’ used. For example, in Nursing, the language is underpinned by emotion and empathy, in Business it is the idea of profit generation that underpins the language, and in Design the visual and the philosophical are key. What is more, we have found many non-verbal and non-text based elements to be of fundamental importance to student success, and, significantly, are inextricably linked to the ‘English language’ used.

The traditional approach to recruitment would suggest that English + Subject in native language = preparedness for university study. Yet, this approach is highly questionable if the English cannot reliably be assessed away from the subject area.

What exactly is it then, that IELTS is testing? IELTS is testing IELTS, and so if students want to do a degree in IELTS, then IELTS is a good indicator of their ability. If however, they want to do a degree in sound design or engineering, the English they need will be very very different. Analogously, one IELTS topic we found was ‘home composting’. To show the English in IELTS students would need to talk about home composting with fluency and accuracy, to show the English needed to talk about home composting in design they would talking about the process of the design of the composter, or the colours used. In nursing they may be talking about elements of hygiene, or of how it could be used with patients with dementia who have memories of gardening. Would the English be different? Of course it would.

What should we do? We we need to rethink the language testing policy we currently have. If we want to test whether students’ English is sufficient for for their degrees, we should be testing this English in their subjects. One way to do this could perhaps be to use English-medium A-level tests or their equivalents. Another way would be to interview students in the context of their subjects, with the use of subject lecturers. We could also require them to demonstrate their subject knowledge in English.

Institutions which recruit students by evaluating English in the subject context will select students better prepared for their studies. They are more likely to succeed, and to be satisfied with the experience. We need to talk about IELTS; avoiding the issues fails both students and their institutions.

To find out more, you can find our article in the journal ‘Power and Education’ here.