Why isn’t Engelmann in our library?

Following the little kerfuffle over the new BPP University PGCE, there has been a bit of discussion about whether or not Theory of instruction: Principles and applications by Engelmann and Carnine, should be available in university libraries for PGCE or other Education students.

It isn’t in my university’s library. I’m open to persuasion that it should be but here is what I think at the moment. I would be interested in any corrections or constructive comments.

Firstly I think it is important to be clear that Direct Instruction (DI, with capitals) refers to specific programmes of scripted lesson sequences. The original DISTAR programmes were designed for EYFS and KS1. More recently programmes for older primary children and an intervention for struggling readers have been developed. The approach is very specific, to the point where there is a rubric for deciding if a programme is genuinely Direct Instruction or not. It is worth quickly skimming this just to see its length (133 pp.) and level of detail. However, if you want to understand DI you might be a lot better off looking at this really excellent interactive tutorial for psychology students at the University of Athabasca.

By contrast, direct instruction (without capitals) tends to refer to any teaching that involves the teacher setting out the expected learning clearly, as opposed to providing learning activities that help the learner to construct their own learning. Typically, this is narrowed further to a situation in which a teacher is engaged in whole-class interactive teaching: laying out clear explanations, asking questions, modelling, and then getting children to practice independently, perhaps initially with scaffolding that is gradually withdrawn.

DI is clearly a good example of direct instruction, but the overwhelming majority of direct instruction is definitely not DI. Use that 133 page rubric to check if you’re not convinced!

Project Follow Through provided pretty conclusive evidence that the DISTAR programmes were very effective for disadvantaged young children in the USA in the ’70s. Arguably that’s a bit dated but I think, given the strength of the evidence, it is highly unlikely that a similar evaluation now would not produce a similarly positive result. Is that also evidence in support of direct instruction generally? Yes, I think so, but only weakly. Why? Because it’s not clear which elements of DISTAR made it so effective.

Here are the features of DI:

  • Attention focused on the teacher.
  • Active, cued, responding as a group and individually.
  • Frequent feedback and correction.
  • High pace.
  • Children taught in small groups; grouped by ability.
  • Scripted presentation.
  • Meticulously designed instruction
  • Faultless communication achieved through application of logic
  • Sequence of examples and non-examples, leading to generalisation

It is these last four, and arguably the grouping, that mean the DI is not conclusive evidence in favour of direct instruction generally. Firstly there is other evidence that suggests expertly designed, scripted lessons may be more effective than lessons planned by individual teachers. Secondly, most direct instruction does not use the logical system of examples and non-examples that is central to DI.

Much better evidence in favour of direct instruction is available. The two main sources I make use of are the ubiquitous Kirschner, Sweller & Clark (2006) (or the lighter professional version of this paper Clark, Kirschner & Sweller (2012) ) and Muijs & Reynolds (2010). The latter is a very good overview of the evidence on effective teaching and learning from two important figures in the Educational Effectiveness Research movement. I should state that Daniel is a colleague and David used to be, but some time on Google will probably establish their credentials and some of their thinking.

But if DI and direct instruction are both effective, why are my students not reading about both? Well, I think that DI is really about curriculum design and I think it is at completely the wrong level, with it’s incredibly meticulous approach, for anyone planning sequences of lessons as a trainee or early-career teacher, or a HoD squeezing out a new SoW in their evenings and weekends. There are things that have worked for me in the past, that are similar to some features of DI, and I share these as part of the PGCE I work on. There has been work done elsewhere (I’m particularly aware of Kris Boulton) but I haven’t come across anything fully developed, nor have I come across any of the NFIDI programmes in use over here. If I’m missing something then do please let me know. I think there should probably be more comparison of examples and categorisation exercises in teaching, and therefore in ITE, but until I’ve read an accessible overview that my trainee teachers can make properly make use of I don’t think there is very much they can usefully take from reading Theory of Instruction.

There is another argument I’m aware of. As a Russell Group university with one of the leading schools of education in the country, perhaps this book should be available for students engaging with the traditional/progressive debate, and maybe part of this is to be aware of the way in which DISTAR was sidelined despite the quality of evidence in its favour from Project Follow Through, but Theory of Instruction isn’t going to tell them much about that. Possibly a later publication deals with that well – if anyone thinks that there is a book specifically on this topic that beats the journal articles that the students can already access then let me know and I’ll pass on the recommendation to colleagues whose modules head in that direction. However, as far as the PGCE goes, I’m going to upset some people by saying that I don’t think the progressive/traditional debate is very useful and I don’t think I’m the first person to suggest that focusing on evidence-informed ideas about effective teaching is more helpful to trainee teachers than getting into the sociology of education. I therefore intend to stick with direct instruction, and will park Direct Instruction until I am convinced of its utility: convince me!


Clark R., Kirschner P. & Sweller J. (2012) Puttting Students on the Path to Learning: The case for fully guided instruction. American Educator, Spring 2012

Engelmann S. and Carnine D. (1991). Theory of instruction: Principles and applications. Eugene, OR: ADI Press

Kirschner P., Sweller J. and Clark R. (2006) Why Minimal Guidance During Instruction Does Not Work: An Analysis of the Failure of Constructivist, Discovery, Problem-Based, Experiential, and Inquiry-Based Teaching. Educational Psychologist. 41(2) pp.75-86

Muijs D. and Reynolds D. (2011). Effective Teaching: Evidence and Practice. London: Sage Publications Ltd.


The announcement of the new BPP PGCE in the TES yesterday generated a lot of twittering, most of which struck me as knee-jerk in nature, unduly negative, and rather uninformed. Now, given the nature of Robert Peal’s polemic against the educational establishment – which ended by describing all those people who (whether right or wrong) had dedicated their working lives to trying to provide a good education for our children, as a persistent national embarrassment – I’m not overly bothered that criticisms that he only trained 6 years ago, hasn’t even been in the classroom for all of that time, and therefore doesn’t have the necessary experience to run the course, are a bit personal. However, I do think trying to tar the subject tutor team with the same brush is unfair since some of them can count their teaching experience in decades and all those I know anything about have clearly been effective teachers and have thought long and hard about teaching within their subject specialisms.

However, my main issue with a lot of the reaction is the misunderstanding of the nature of the PGCE in relation to QTS and the role of the university and the SCITT. I can forgive anyone for not being able to understand the current complexity of ITE in England – I regularly fail to explain it clearly to prospective trainee teachers myself – but a lot of the negative commentary and questioning clearly assumes that the BPP PGCE is all the training that is provided and I very much doubt that’s accurate.

Now, I don’t know exactly how this new PGCE integrates with the Pimlico-London SCITT so instead of speculating, this is what I do know, based on working with various SCITTs and HEIs.

A SCITT is an accredited ITT provider, based around a school or group of schools and approved by the NCTL to recruit trainee teachers and recommend the award of QTS. As such, it is their responsibility to ensure that all the Teachers Standards are met, that trainee teachers get suitable training (including things like professional expectations, subject knowledge, planning, behaviour training, understanding of SEND issues, assessment and so on) and good support, and that the quality of NQTs at the end of the course is appropriate. Ofsted’s view on this is that all NQTs should be exceeding the Teachers’ Standards at the point QTS is awarded and if Ofsted inspect and conclude that the training has omissions or the NQTs are not good enough, then a Grade 3 (or 4) effectively shuts them down.

All of this relates only to QTS; a SCITT is not a university and cannot award a PGCE, which is a post-graduate qualification, usually at Level 7 with associated Masters credits. Typically the PGCE consists of two or three assignments totalling 12000 words. These will usually include relating practice to theory and may include collecting data from lessons but, importantly, the PGCE is an academic qualification and the grading has absolutely no link to the actual quality of classroom practice of the trainee teacher.

Some SCITTs therefore just do QTS but most have a contract with a university to provide the additional PGCE. This gives the training more kudos and makes it more portable e.g. to Australia. How much input the university has depends entirely on the nature of that contract. Here are some examples that I know exist:

The university train the SCITT tutors to deliver the PGCE and never work directly with the trainee teachers. The university sets the PGCE assignments but the SCITT tutors do the input and mark the work, and the university just moderate the marking and make the award.

The university provide all the teaching and assessment of the PGCE assignments. This might involve the trainee teachers going to the university, or university tutors coming to the SCITT.

The university provide all the teaching and assessment of the PGCE assignments and also some subject-specific training. This is what we do at Southampton. The advantage for the SCITT is just related to economies of scale. A small SCITT may only have 20-30 trainee teachers. Pulling an experienced teacher out of the classroom for every subject, for a dozen days a year, to do subject-specific training, is difficult, whereas a university has enough trainee teachers for there to be someone (like me) in each subject for whom this is the core of a full-time post.

Set against this background, the BPP PGCE model looks perfectly acceptable to me. The university are supplying some training, which is likely to be more subject-specific than the SCITT can provide on their own, and the PGCE assignments, which the SCITT can’t provide at all. The only thing that is a bit unusual is that BPP itself is not an accredited ITE provider and it doesn’t have a track-record, or any intention, of producing education research. I’m not convinced that’s a problem. In some universities, the QTS and PGCE work is done by tutors who have a research element to their contract. In others, like Southampton, there is some overlap but most of the tutors are on teaching-only contracts. I don’t think that reduces the quality of what we do at all.

Finally, PGCE tutors are typically qualified at Masters level, and I think that’s not entirely the case for the BPP PGCE. On the other hand the intellectual capability of the team is clearly strong and it is the university’s job to ensure standards are appropriate, which presumably includes having an external examiner from another university.

So, it’s different in flavour, and it’s sort of starting from scratch. I think the content of ITE generally, and this course in particular, is a legitimate, and very important, area for debate (and helpfully some clear indication of this is already available for English and history). I think knee-jerk responses of “this must be rubbish” because of who it is and how it is set up, are unhelpful. We train 40 000 teachers each year in this country. In that context BPP is a tiny but potentially interesting innovation. Let’s either engage with what they are actually doing (or ignore it, if preferred), and see how it goes.