90% of teacher training should be subject-specific

Initial teacher training (or education) is currently under review, and one of the issues being raised is the extent to which teacher training does enough to develop teacher knowledge of what they are teaching. Although I see this as a vital component of teacher training, I do think it rather misses the point. For me, subject knowledge is not a ‘component’ of teacher training: rather, I think that subject-specificity should run through around 90% of everything that a trainee does during their training. I want to use this post to explain why.

I shall start with the 10% that I think does not need to be subject-specific. There are clearly some generic things where it makes sense to bring together trainees from a range of subjects. Behaviour management is perhaps the most obvious (though there are still important differences between the behavioural challenges one faces in history, drama and physical education), but there are also things that all teachers need to do that are generic: being a form tutor, talking to parents, handling child protection issues, and so on. For all of these, I see no problem in having generic training.

But what about everything else? What is it that a teacher needs to be able to do at the end of his or her training? This list probably covers around 90% of what a new teacher needs to learn to do:

  • interpret a curriculum
  • plan sequences of lessons
  • plan individual lessons
  • plan activities for lessons
  • identify and adapt appropriate resources
  • give instruction in lessons
  • manage discussion between pupils
  • ask questions of pupils
  • respond to pupil questions
  • give immediate feedback on pupil work
  • mark the work pupils produce
  • advise pupils on how to improve
  • interpret and apply mark schemes

These are all the ‘skills’ of a teacher, and I think these are generic in the sense that teachers of all subjects have to do them. Does it not make sense, therefore, to base teacher training around these skills? Can we not run training sessions on ‘questioning’ or ‘planning’ or ‘feedback’?

My answer would be no. All of these things, I think, should be taught as subject-specific. The reason for this is that all of these skills are underpinned by knowledge.

Let’s take ‘questioning’ as an example. It is my historical knowledge that makes it possible for me to question pupils effectively in the history classroom. I know the kinds of questions that we seek to ask and answer in history (Why did something happen? What were its consequences? How quickly did things change? What kind of change was this? Did everyone experience those changes in the same way?). If I am to ask pupils a causal question (why did something happen) then I have a number of possible questioning routes open to me: I can ask sub-questions about short-term and long-term causes, I can ask about triggers for events, I can pose counter-factual ‘what if’ questions, I can ask about different kinds of causes (religious, political, economic) or I can ask about necessary and sufficient causes. All of these things have a specific meaning in the discipline of history and my questioning in the classroom is underpinned by my knowledge of the kinds of questions that are asked in the subject.

I am sure that all teachers have sat through training on giving formative feedback to pupils, but is this training really that meaningful if it’s divorced from the subject discipline? In order to give formative feedback I need to have in my head a model of what it means to get better at history. This means that I need to know what ‘good’ history looks like, and the different routes by which a pupil might move towards producing that kind of work. Usually generic sessions on formative assessment fall back on things like Bloom’s Taxonomy, which is simply not appropriate for history (and, I reckon, any other subject). If we teach how to give formative assessment in subject-specific ways, then we have a whole range of more helpful and more meaningful feedback available than ‘try to include more evaluation’. In training history teachers, for example, we might get trainees to consider how pupils might get better at using one event to illustrate a trend or addressing the economic dimensions of a question. It is quite simply impossible to get down to this level of detail in a generic session on feedback.

The same I think is also true of some of the insights from cognitive psychology. In my previous three posts I wrote about how the ideas from the book Make It Stick (spaced retrieval, interleaving, etc.) might apply in the history classroom, and it is quite clear to me that, although children are wired up in such as way as to make some things possible or efficient and others not, it is always necessary to ask so what does this mean in my subject? There might well be something to be gained from studying a bit of educational psychology in teacher training (as I wrote about here) but it is always necessary (and I think more time consuming) to then ask how these principles apply in one’s own subject.

My belief that 90% of training needs to be subject-specific does have some implications for the kinds of structures that might make that training possible. This model requires that trainees spend most of their training sessions in subject groups, and this requires some scale in order to be efficient (probably a network of over ten schools). It requires that mentoring (and the associated mentor training) is also subject-specific, which again requires some kind of network. It also requires that the person responsible for running the course (running sessions, training mentors, choosing readings etc.) is also an expert on the teaching of a particular subject. This, for me, is a role where universities are particularly well-placed to serve as hubs, particularly as an expert in (say) history education could interact easily substantive experts in a history faculty and experts such as educational psychologists. Other structures might well make this work, but, as I have written, the role of universities has always been to provide a home for experts in a range of fields, and I think we are missing a trick if we do not capitalise on this.

We have been playing down the importance of subject-specificity in education for some time. It was very fashionable for a while to say things like “I am a teacher first and a subject-specialist second” or “A good teacher can teach anything”. In reviewing teacher training, I think we would do better to start asking ourselves “what is it about my subject expertise that makes it possible for me to teach”. That, I think, should be the starting point for discussions about teacher training.

4 Comments on 90% of teacher training should be subject-specific

  1. I’m guessing you mean for secondary teachers – given that primary teachers must teach lots of subjects. But, what about teachers teaching more than one subject? Or do you think that if there is a solid basis in one subject then it will enable easier transition to teaching other subjects?

    (It would be a nice world where people can teach one subject, but it’s rare due to the peculiarities of timetabling).

    Also, in terms of the 90% – are you including time teaching in that? Only. most of any teacher training route is taught in a classroom where you are implementing the generic knowledge in a ‘subject-based’ way, so the % is already quite high.

    Interesting thoughts, though. Do think subject knowledge is more important than we have previously assumed.

    • Michael Fordham // 28 August 2014 at 11:12 // Reply

      Yes, the post is obviously written from a secondary perspective, but I’ve had to think a bit recently about the logical implications of this for the two scenarios you give (secondary multi-subject and primary) and it’s brought me to some quite stark conclusions.

      Regarding primary, I am now convinced that a one-year PGCE is simply not sufficient. It might be sufficient for preparing to teach numeracy and literacy, but not other subjects. This leads me to think that (a) training to be a primary teacher is harder, (b) it needs more time – perhaps along the four-year BEd model and (c) that there’s an even stronger need in primary than secondary for subject-specific training for NQTs and RQTs. Based on history at least, the evidence we have (e.g. Ofsted, HA survey) tells us that primary teacher training and CPD for history are woefully inadequate.

      For secondary, if we train to teach two subjects, then perhaps it’s just about manageable in the one year, though clearly if one is training to teach two subjects then that requires more work than training to teach one. I think scientists have a big problem with this and I can see strong arguments for moving to two-year science PGCEs, though that certainly wouldn’t help the recruitment problem. One way around this would be to allow scientists (and indeed in other subjects) to shift into a ‘teaching stream’ after the first two years of undergrad which would allow extra time for training, a kind of half-way house between our system and the German system (as I understand it).

      All of this of course requires some structural changes, and my sense is that many don’t really recognise the challenges involved in this.

  2. Interesting ideas there and I agree about the importance of subject specific training although you haven’t mentioned things like SEN, which is quite generic, and EAL which is a bit of both.
    One of the issues with SD, in my experience, is the difficulty small alliances have had in getting subject-specific training to work really well. One of the adjustments our partnership has made is to allocate more of the university-based training to subject-specific sessions for exactly this reason. I also feel that the three M-Level assignments in the PGCE are something that should be looked at – I would prefer one M-Level assignment to focus attention on education research with the other two completed in following years. My post http://wp.me/p44DHA-2E covers this and some other ideas for improving ITT.
    Your comment about science I would like to elaborate on. Science graduates usually have one science to degree level, one to A-Level and one to rusty GCSE level. Reading your recent posts I’m struck by the sophisticated level at which you are working with trainees. You do have an issue with different historical periods but my impression is that your trainees are capable of filling in these gaps through independent study, allowing your training to focus on the subtleties of teaching the subject. We do as much work as we can on flagging up common misconceptions during the PGCE but I’m very conscious of the contrast between our trainees, still struggling to feel confident with GCSE questions in their weaker sciences, and those in other subjects for whom the subject knowledge is a given. I would like to see as many science trainees as possible doing a 24+ week SKE prior to the PGCE – this is the course which would address the problem and it is already fully-funded, with a generous bursary. It just requires the NCTL to recognise that this course should be about improving the overall quality of science teaching rather than just converting some biologists to physicists or chemists. There is a model being developed whereby undergrads complete QTS as part of their degree (your idea about a teaching stream) – Imperial are already piloting this – however I very much agree with you that many people underestimate the challenges in making this work well. The Troops to Teachers 2-year training programme is another example of extended training but also another example of how new training routes need time for problems to be ironed out.
    Best wishes

You can offer your thoughts here:

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: