The setting argument: a classic case of the poverty of genericism

During my summer holidays, I generally try to learn something new in a more structured way. For example, in the summer of 2016, I attended Cambridge University’s excellent medieval summer school (and I’m going again this year!). In 2017, I spent four weeks taking the intermediate German course at a language school in Bamberg. Both were great experiences.

At Cambridge’s medieval summer school, the classes were not set by any form of ‘ability’. The only admissions criterion was that you could speak English. This meant that the courses I attended (on challenges to royal authority and the Black Death) contained people with a wide array of prior experiences. Some had history degrees, or even PhDs, and others had no higher education qualifications. Some had been attending the school for years and knew a great deal of medieval history; others were rather new to the whole period. I would say the lectures were pitched at ‘undergraduate’ level, but they were certainly accessible to everyone there.

At Bamberg’s language school, we were set from the outset. There were, I think, five different sets based on the European languages framework (A1, A2, B1, B2, C1). We did not have an entrance test, but instead we were set based on prior study. If the level was too easy or hard, we were able to move. I found the B1 course was just right for me. I have had the frustration before of being in a language class where the level is too easy or too hard: in my first year of university I attended the lower intermediate French class, and quickly realised that my GCSE-level French was woefully inadequate. Similarly, I have attended German classes where the level was a bit too easy: it was good practice, but possibly not the best use of my time.

I share these two examples because they illustrate a point often lost in the great debate about setting. Subjects simply do not work in the same way. In a subject which is more cumulative in its internal structures – such as history, or English literature – it is perhaps not particularly necessary to set. I have taught history in six different comprehensive schools, and five had no setting. Nationally, I think, mixed ability is the norm in history. In a subject which is more hierarchical in its internal structures – such as mathematics or a language – setting is far more common. I think in all the schools where I have worked, those subjects were set.

In my last blog post, I argued that we should be very cautious about making sweeping statements about generic teaching methods, as our starting point should always be ‘what is right in this subject’. This is of course not the final word on the matter. There are first the institutional constraints we all face, with the classic school one being the timetable. Maybe English teachers prefer not to set and Maths teachers prefer to set, but the way pupils are timetabled means it has to be one or the other? More importantly, we have to hold ourselves to wider ethical standards: if it was shown that electrocuting a child helped her learn history, I still say “no, this is wrong”. Perhaps any form of setting is wrong on ethical grounds (although I have not seen many cases at all of mixed-ability maths teaching which has not involved the pupils being set by table at which they sit.)

We often talk about wanting a more nuanced debate about teaching. Whether to set children or not is I think a good example of where this nuance needs to come from recognising that not all subjects work the same way.