Open Policy Experiment 1: School Direct and Initial Teacher Training (UPDATED 25/7)

One of the things I wanted to do in the Department for Education was open up the policy making process and run things like wikis in open formats in order to a) start off with better ideas and then b) adapt to errors much faster than is possible with normal Whitehall systems.

Obviously this was ‘impossible’. In the DfE, one is not even allowed (officially) to use GoogleDocs. (Why? Officially – ‘security risk’. Reality – Whitehall’s fundamental operating principle is ‘obedience to process‘. It is not – have a good product, service, or idea. Decentralised collaborations are inherently threatening to Whitehall’s core principles. Hence, for example, why they hate the model of: incentivise a goal and be neutral about method. Although this model has been a success throughout history, it obviously flouts the principle of ‘obedience to process’).

So we could read what was happening in the outside world far from the 7th floor of DfE, and occasionally email or get people in, but we could not interact it with it using modern tools.

But I have a proposal that costs nothing… I’ve planned to do it for a while but today’s twittering on School Direct prompts me to do it now.

I will pick a topic. Today, School Direct & ITT.

And I invite people to enter comments explaining –

What does not work with X?


What specific things would improve it?

The more specific complaints and recommendations are, the better. A curse of being in the DfE was generalised whining and when we asked ‘what SPECIFICALLY do you mean, what SPECIFIC regulation is causing trouble?’, <1% of people had an answer.

I specifically INVITE criticism of what we did. Not abuse, not praise, not general whining – but specific criticism that can be used to improve things.

The ideal comment would be something like –

‘The following specific regulations XYZ and guidance ABC say on pages X the following Y. This is damaging because X. The evidence for this is X. What should Charlie Taylor do? Tell Marcus Bell and his team to eliminate pages X-Y, and rewrite Z to make everything much simpler and the incentives better aligned. The whole of document A should be withdrawn apart from para B on page C, which should be added to D. The funding system causes problems by XXX. If you simplify it by doing YYY, it will eliminate 90% of the problems with A but won’t solve B. B could only be solved by changing primary legislation XXX…’

You get the drift. This is the sort of advice that approximately never is given to ministers or spads. If the people on the ground dealing with the consequences of Whitehall decisions could give them such help, then it is possible that lots of small improvements could be made quickly. I often made small improvements / corrected our own errors  in response to emails from the front line but this was very sporadic – not systematic – and the process left me screaming at my computer that we were, because of the insane Whitehall structures, so disconnected from reality.

Why would you bother?

DfE ministers, spads, and officials watch this blog. They might change things if you help them by explaining SPECIFIC things they can do. They might also think ‘if we do X, then education world will complain Y, so let’s not do it’.

Gove will read the comments (this is not a promise based on discussion but a prediction based on character). Gove is going to be involved in writing the next Tory manifesto. Therefore if you can show why something is wrong / stupid, you have a chance to influence him and give him ammo to head off the appalling stream of gimmicks that are, as we speak, being cooked up. Others in No10 will also read it. (A plus is that this process can influence No10 even though everybody in No10 will deny they even read it.)

Labour’s team read this blog looking for information to harm the Tories therefore will happen upon useful information that may also nudge them in useful directions. If they become the next government – which betting markets think is reasonably likely – you will have helped educate them.

The media read this blog looking for ‘news’ so also will see worthwhile information.

I will try to answer questions (from those interested) about why we made certain decisions, relying on memory, emails, papers etc. But my goal is not to ‘defend what we did’ – it is to discover what we did wrong so others can improve it. Also, NB. I left DfE partly because I was desperate to have as little involvement in the election as possible and I plan on implementing this by being abroad for its entirety so I don’t have to listen to a word. From recent interviews etc, it ought to be clear that this experiment is not designed to help Cameron or any other political force win an election.

Nothing will be censored or edited by me other than abuse/swearing/obvious frivolity etc, so that hopefully reading the comments will be worthwhile.

If nothing comes of it, then I’ll stop and nothing has been lost apart from a little bit of my wasted time. If someone comes up with a better technical solution then I’ll ditch this and transfer whatever has been done to it…

So, School Direct.

What do you think, why, and what should be done. SPECIFICS PLEASE.

I’ve texted Charlie Taylor so you know he’s going to be reading…

UPDATE 1: Acronym glossary.

Someone reasonably pointed out in comments that non-specialists don’t want to have to google all of the acronyms so here is a quick list of the most common used in comments below.

EEF = Education Endowment Foundation:

HEI = higher education institution.

IP = intellectual property.

ITT = initial teacher training.

NLE = national leader of education.

NQT = newly qualified teacher.

PGCE = post-graduate certificate of education.

QTS = qualified teacher status (a Whitehall-defined certification process for new teachers).

R&D = research and development.

SCITT = school-centred initial teacher training.

SD = School Direct. (A post-2010 programme in which schools recruit people before they do training (unlike PGCE), then train them, then often give them a job. Controversy over the flaws / merits of this programme is one of the reasons I did this blog.)

SLE = senior leader of education.

TS = teaching schools.

UPDATE 2: next steps.

To those who have commented…

I am going to leave this thread as it is until Sunday/Monday, then do another blog summarising / clustering the comments and publish that (Monday), in the form of a note to ministers / spads / officials in the DfE. Then people can send corrections / additions etc, and I’ll redo it, then post a final (for the moment) version.

Thanks to all who have contributed so far. I know many of the relevant people in the DfE have read your comments so hopefully some good will come from your efforts…


134 thoughts on “Open Policy Experiment 1: School Direct and Initial Teacher Training (UPDATED 25/7)

  1. Shut down all the PGCEs. Teaching is a very practical subject that you learn by doing. The Academics at University try and impress their colleagues by being as progressive as possible rather than teaching what actually works. Also, a lot of cosy, middle-class jobs depend on pretending that teaching is really complicated and needs years of study led by them. It isn’t. Teaching is an incredibly difficult job to do well but it isn’t complicated. You teach the kids some stuff and then you help them do some work. Practice for a lifetime.


    • I’ll OK this comment this time but only to use it as an example of what I do NOT want to see.

      It does not help anyone improve anything specific.

      This experiment will only be useful if it involves specifics & not the sort of comment that clogs up other blogs and is an alternative to useful discussion.

      I am not interested in moderating ‘random unsubstantiated political discussion’.



      • Without wanting to necessarily continue the opening flurry from Ollie (who is, as I know from discussions I’ve had on the subject with him on Twitter, vehemently against PGCE / university route into teaching), are you interested in hearing about the advantages of PGCE and university led teacher training or is that not a line worth pursuing in this discussion on ITT? I want to know, before I put pen to paper as I don’t want to waste precious holiday time arguing for something no one in government is interested in retaining.


        • Yes I’m very much interested in the pros of PGCE.

          The key to this experiment is specifics.

          So I don’t want a reply to Olly in the same style but evidence and specifics that people in the chaos of Whitehall can act on.

          Similarly if Olly provides specifics, good.

          All arguments need testing. We can do this in the open and those making decisions will read…



  2. Allow and then directly encourage ITT in Special schools to place Primary trainees in Secondary age classes where there is curriculum content being delivered which falls into the Primary areas of teaching. This would significantly increase capacity to deliver training placements in Special schools and go some way to begin to alleviate the major recruitment problem in this area of education.


  3. Ok Dominic, I’m having a go with limited/intermittent wi-fi and no access to some of the documents I might find useful but as a starter we need to recognise that there is a distinct difference between ITT provision at university level and educational research – they are often separate departments and in some cases there is little communication between the two so what might be useful on the ground doesn’t always get through. Bringing ed research and teacher training closer together might help – this means thinking about the research funding processes and how we value the impact of research on the profession.

    Secondly we need to think about access to research for teachers. Journals are protected by, for example, Athens and so it is hard for teachers in the ground to access them. This means that the profession is dependent on middle man interpreters which can lead down the VAK route!

    It is my personal view that all routes should lead to the same quality of outcome and wider reading if research and theory is crucial to the professional integrity of a teacher. So the standards have to be the same for all. This is where there is difficulty with school based routes. I was entirely opposed to SCITT and SD for the simple reason that it would be almost impossible to secure consistency. I’ve recently been invited to work with and advise a SCITT group in Greater Manchester and have been impressed with the quality and rigour of their thinking, but this is largely because it is led by a dynamic and thoughtful teacher who herself was trained at Cambridge and who has a clear idea of what is required. Elsewhere this is not the case and I think consistency is the key here.

    Finally I would urge any SoS to consider the impact that the NQT survey and Ofsted have on the behaviours of ITT providers in universities – not unlike that in schools. The priorities of Ofsted drive T&L behaviour and programme structures and so we see provision sway in policy directions rather than focusing on what research suggests. This is a major problem.

    I’d also suggest that whatever the route, all teacher trainers should have recent experience of teaching, if only to have empathy with the realities of the job. And equally, it should be a requirement of all teachers to keep up to date with reading. I disagree that teaching is an entirely practical process – it is also an intellectual one and we should have a professional responsibility to keep up to date. For this reason I am strongly in favour of Alison Peacocks notions of professional development monitored and approved by an independent body such as a Royal College of Teaching.

    I know this is not as as specific as you would like, but I am in holiday and half way through a glass of ouzo. I hope it offers something to the debate.


  4. Oh, sorry one last VITAL thing. It should be a professional duty to take trainees into school. One of the biggest problems PGCE courses face is having to beg schools to take on students. This is not the case in the NHS – you must commit to training the next generation of professionals there. The same should be true of education.


    • I completely agree with this. I have just completed a PGCE and had to find my own placement (ending up in my Uncle’s school which was really not ideal but necessary seeing as I was only given 1 week to place myself!) as the university were really struggling to find schools to take us. One fellow course mate had to place herself twice… Added a lot of stress around assignment hand in time!! I believe that a school should only be able to get ‘Outstanding’ in OFSTED if they are involved in training future teachers.


  5. I’m probably breaking your rules by being too frivolous, just wanted to say good idea, great stuff, we need more like this, topics of govt policy I know more about I would welcome a chance to contribute like this.
    As an aside on this stuff, I think the wording is something like “Academics who have completed (or are finishing) a doctorate can become qualified to teach through the Researchers in Schools course” I would change this. In practise the qualification for teaching in university is a Masters, and there are many good uni lecturers who have done a great job teaching at uni with just a Masters and some help from their more senior colleagues in the early part, not so many fancy lectures as per qualified teacher status. I myself (sorry no evidence, but a few short experiments could test) feel that some uni lecturers would be great teaching at secondary school level, and should be allowed to do so, even if just for one year placements, or subjects where schools need to do better.
    Indeed I think one year secondments of the right kind of people from the outside world, without qualified teacher status, would do schools the world of good. One of the problems in some schools being teacher burn out, no matter how good a teacher you are at the beginning many years working your socks off in a sink school does take its toll even on the best. I also think there should be a mechanism for teachers in leafy suburbs to spend some time in a sink school, without the real risk to their life and career that they could not escape back at the end of it.


  6. Take away the necessity for Teaching Schools to have an Outstanding OFSTED rating. Any designation should be based on the quality of the bid and the prospective provision. The reliability of the OFSTED Outstanding/Good judgement is insecure. For example, the OFSTED judgement for a selective school is more likely to be Outstanding, but there isn’t much correlation with quality of SD provision.

    Teaching schools are overloaded with too many strands alongside ITT. Without moving down the road of a designation for every schools, like the specialist school designations, you could have national Research School hubs, like the recent ‘Maths Hubs’. The 32 Maths Hubs sounds like a good number for a Research Schools.

    These Research School Hubs would be responsible for the dissemination of research evidence that feeds SD ITT programmes; leading on the knowledge mobilisation of the work of organisations like the EEF etc. These schools can each be linked to a HEI, openly and with a transparent bidding process and contract. They could train Research-leads (I am currently beginning an RCT for this very model: They should marshall a corpus of educational evidence & theory that best informs ITT. This can be regularly updated and challenged.

    We should rigorously analyse the impact of SLEs and share good examples of practice. My understanding is that the quality and impact of the role is erratic and without a systematic approach. The DfE should encourage an RCT (with some qualitative data), independent of the NSCL, to be undertaken based on this initiative.

    Teacher trainers should still practise in the classroom, or, at the very least be based in schools, with very recent experience of teaching. Even two years away from the classroom can create a crucial knowledge gap between existing practice and training – for example, the application of new technology etc.

    Ensure all core literature from ITT courses, such as lesson plan templates and course outlines, can be accessed publicly to ensure transparency and quality. I have looked at a couple of ITT lesson plan pro formas from SD schools and I was sorely disappointed. Why change the whole national model if it isn’t going to be significantly better than what has come before.

    More generally, transparency about recruitment of trainees and retention etc. should be shared by all SD schools.


    • Thanks.

      When you say Teaching Schools ‘are overloaded with too many strands alongside ITT’, what exactly do you mean? Sorry if it is obvious.

      I’m sure Tim Leunig in the DfE will have noted your comment re RCTs.

      Re making materials transparent… Before I left I was working on a plan to get all ITT providers to put their materials on the web. There was a lot of resistance and bleating about IP from HEIs but I was hopeful it would happen. It has not. I’m told that lawyers have, as usual, kicked up a fuss. Do you know why HEIs are reluctant to make their materials transparent? To me, transparency about basic materials is vital if there is to be a self-improving training system, but obviously this can cut across commercial incentives.

      When I organised the maths and physics projects with Cambridge University Maths & Physics departments, we agreed that all materials would be made public so everyone could study and learn from them. The maths & physics professors involved not only did not object but thought it obviously the right thing to do, as well as to have proper evaluations. Their attitude seems to be rarely shared by administrative people (similarly on the issue of exam quality but that’s for another blog).

      Best wishes


      • Surely the same rules apply as something like the Creative Commons system – share alike unless something becomes a commercial proposition and then it is possible to change the licence although it may not be revokable:

        To avoid disputes use something like a TimeStamper to record when, where and what was issued. There are also such things as sui generis database rights, as well, that could trump these – so the lawyers are right to be wary in such cases.


        • That’s a good idea though I know from previous arguments that Govt departments think of CC as a bit newfangled/California and like the old system of everybody being subject to Crown copyright. But I think you are right that this is the direction ITT shd move in.


          • Government departments need to a) read the stuff the cabinet office has put out on open standards and b) value getting up to date. Crown copyright has limited value if you want open standards to be the norm. If we are expecting teachers to be up to date in what they are teaching why do we have lower expectations of government departments?


  7. By too many strands I mean the big six of being a teaching school: ITT; CPD; school to school support; developing new leaders; SLEs; and research and development. One school doing all of these brilliantly and continuing to run a successful school is no doubt a stretch.

    R&D is sixth in the list (perhaps symbolically!). It should be number 1 – underpinning each other strand. I know many teaching schools are alliances and some are part of a larger chain, but even then the specialism of being a world class research school rarely comes to fruition. I know of no school doing this (neither do the DfE given their sharing of best practice document on teaching schools – it has only one exemplar school in the latest report). Clearly, the DfE values evidence led policy and practice (the EEF is perhaps the best legacy of the last decade IMHO). At this current point of systemic change, R&D gets relegated behind the exigencies of immediate action by teaching schools. Mandating Research School Hubs in the next decade would be a bridge to an evidence led system.

    I can’t think of many valid reason why making documentation like lesson plan pro formas and basic course outlines (including what evidence and theory is being explored as part of the course) would be invalidated by lawyers. Schools have to publish lots of documentation for OFSTED inspectors on their school website. The same should be said for ITT providers. It could be easily done in this way. I don’t imagine teaching schools are in direct competition to the degree that they have a secret recipe for brilliant teacher training (if they do, they should execute their public service and share it!). There are some excellent HEIs who would and could do all of the above, but it is in no way systematic.


    • Ahh, re your first point. I remember a senior official mentioning this last year and suggesting TS be given flexibility, not have to do all 6. And this was agreed. And I thought this had happened and been communicated!

      I know from some emails that officials are already watching this thread, so I look forward to an answer! Have I misremembered / got confused?



      • If it is the case I am not aware of it. I don’t know how it would work unless you have real strategic oversight regarding where the strands are being met in a locality. If it is the case I worry that research would become a very niche practice. It doesn’t pull in the punters like ITT or SLE strands!


        • Re the 6 strands, I’ve had an email from a middle-ranking DfE official who works in this area saying – TS don’t have to do all these 6 strands, most don’t – it’s an “unenforced requirement”. But DfE IS still looking at removing it altogether.

          Also, I know that a) Gove has been working with officials on a plan since before Xmas for a ‘new QTS’, b) Clegg was blocking its announcement because he did not want the QTS issue to vanish as a political factor pre-election, but c) Clegg had recently dropped his objection as part of some deal, and Laws WAS in favour of announcing the new QTS package (I’ve been critical of him since I left but ‘good for him’ here). I do not know whether it is likely that NM will announce this herself now or whether it will revert to ‘blocked’.

          The new QTS deal involved some of the things mentioned by commentators here – eg. awarding QTS after X yrs, not straight away so that it is not just a mark for getting through a course.

          (NB. Tristram take note, you should outline your plans sharpish to push No10 into thinking ‘close this issue down’! And NM and the new spads – ALSO take note that you shd get this issue resolved pre-election, as this would not only keep Lynton happy (QTS closed down, less ‘noise’) but much more importantly it would help schools and teacher training make progress. Laura T will be a useful ally here…)

          When I get better information I will post it.



  8. As somebody who has just completed the PGCE School Direct route for Secondary English I feel the main issue is lack of consistency. I was in a school where three of us were training within one department and, even in the same department, the quality of mentoring was inconsistent. This had a noticeable impact on final grading and, more importantly, attitudes towards teaching. Discussing issues with peers on the course at university, it was clear that there was no level of consistency in the quality of mentoring/timetabled teaching hours/responsibilities within school and, other than the three ‘quality assurance’ visits from the University tutor, there was no way way of monitoring this. I was very lucky with the mentors I had in both of my schools however I know a number of people on my course (University of Reading, English) were not so lucky and this really impacted their progress.

    Similarly, there was little support in completing academic assignments – 2/3 of these were due in whilst we were in school. Within school we were not given support with these and many of us were in a school too far from the University to be able to easily access facilities, such as the library and ict suites, out of school hours.

    On the plus side, 18/19 of us have all secured jobs for September – the one person who hasn’t has decided they want to take some time out of teaching. A large number of us secured jobs in our placement schools.

    I am currently in Oman typing this on a phone so apologies for not being able to be more specific.


    • Quality of mentoring is an essential component of successful training, but you’re right in identifying the lack of consistency. There is no training given, to my knowledge to mentors, without even a guide. They are most often good teachers, but that, of course, doesn’t correlate with being a good mentor. It is particularly an issue when mentors are pushed for time – which is probably a universal issue.


  9. Dominic you say “Re making materials transparent… Before I left I was working on a plan to get all ITT providers to put their materials on the web. There was a lot of resistance and bleating about IP from HEIs but I was hopeful it would happen. It has not. I’m told that lawyers have, as usual, kicked up a fuss” – I don’t ever recall (I work in a University) such a request but what did you want and what did you want it for? If anything the changes to a market driven approach have made this worse as people become protective about their materials (and this includes schools). Equally if what you say is true this isn’t unique to HE (even though it is almost inconceivable to run a PGCE without sharing information). I have walked into schools as they are taking down displays as they don’t want visitors from another school to see what they are doing.


    • I just wanted everyone being funded by taxpayers’ money to make the materials they are using for training purposes to be put online so that there is transparency. It seemed to me that this would a) help institutions swarm towards the best materials, b) put pressure on the duffers to pull their socks up, c) open up the REAL world to proper scrutiny, instead of having a lot of speculative and ‘my friend on a PGCE said’ type comments.

      Lots of people attack PGCEs now, including some of the best heads, because of what they think of as poor quality materials. But most of this debate is anecdotal – heads and Academy chains like ARK/Harris et al won’t say publicly what they say privately and HEIs fight back saying criticism is unfair. (Similar to the debate over exams.)

      My hope was simply that if HEIs put their materials online then the debate over how good / bad they are would be based on the real materials rather than on stories.

      Of course, there are many dynamics to this that I don’t know about or understand so I’m sure that there would be unintended consequences, but discussions in the DfE boiled down to ‘the HEIs say they don’t want to do it / refuse and it’s legally a minefield’. (I thought it would nevertheless happen in Q1 2013 but it has not, ‘for legal reasons’ according to officials.) Unfortunately I did not have time to follow up personally so I did not discuss this with HEIs. Comments here suggest some already are making materials public.

      In a nutshell: is there a reason why those materials paid for by taxpayers should not be public?

      I have a general attitude – much influenced by Michael Nielsen’s brilliant book ‘Reinventing Discovery’ and Tim Gowers’s campaign for open access to scientific journals – that ‘open is better’, although I know that there are inevitably some exceptions.

      Best wishes


      • I don’t see why in principle, any employer can not specify that any materials produced by an employee is made available under a Creative Commons License as part of their contract. Further, if that material is donated to a web site and the conditions of putting stuff on that site or editing it is that the content has to have the same license it means everything on that site is licensed for sharing and in the same way. That means we get rid of all the complexity of trying to find out who owns the copyright and and exactly how it is licensed. Such a site is being created for computing at also the learning site at has around 55,000 pages and 20,000 users (mainly children) all contents licensed for sharing. Without wanting to sound rude, the DfE is still collecting data in word processing documents so I doubt there is much real understanding of how to build on-line communities there. It certainly isn’t easy but you need a can do attitude to it rather than identifying all the reasons why it won’t work.


        • Yes in the DfE there are a few problems with such things. 1) Their own ICT is rubbish. 2) They are not set up to run these sort of collaborations which generally cut across Whitehall processes. 3) Legal advice pushes everyone towards very conservative positions. It can also sometimes come from the Cabinet Office as part of a general reactionary ‘don’t change’ directive, then officials feel obliged to do as they are told unless a minister forces the issue (and often even then). 4) Although many junior/middle people want to do things, the general dynamics of being part of Whitehall strongly push against a ‘can do’ attitude and senior people are incentivised to be very risk averse.

          I must have raised the ‘publish everything using some sort of Copyleft system’ at least 12-18 months ago. Still has not happened despite lots of people – including Gove and Charlie Taylor – wanting to. This happens on zillions of things and often can only be overcome when one of the most senior people makes everybody’s life hell in determined fashion until the system ‘breaks’ and gives what is wanted: e.g. a SoS says ‘I don’t care if the CabOff legal advice says X, I’m doing Y.’ Obviously senior people have limited bandwidth and can only do this on a few things. If I’d stayed I would have made this a priority but I also thought it was just about solved just before I left. Obviously I was wrong but I don’t yet know the specifics of why I was wrong.


          • It is getting better. I had umpteen meetings with BECTA over this and now as an Awarding Organisation deal with Welsh DfE, Ofqual, SfA and DfE. I only use open systems so when they send me something in a Word Document that won’t open in eg Google Docs or Apache OpenOffice, I politely ask them to send it me in an open format in line with cabinet office policy because they should not be forcing me to use specific applications from one monopoly supplier (and send the the link ). Mostly they are getting the message. DfE VQ reform say they are moving to web based data collection next year. So we are getting somewhere even if it is glacially slow. Point is the more people we educate the more likely it is to dawn on people that there are better ways to do it. So maybe this should be part of ITT provision. Certainly understanding copyright and trademarks is important for all teachers in an internet age. One of the reasons to build it into school qualifications with mass take up. It then gets to a lot of people.


  10. This may be tenuously linked to the needed outstanding rating for Teaching Schools, but I do see leaders in “outstanding / good” schools getting more opportunities through Teaching Alliances than leaders in “requires improvement” schools. I’m not sure if this is an agreed part of the initiative?

    This view is exacerbated in selective areas where intake is an increasing factor in the Ofsted rating. To get systemic improvement these schools, and the effective leaders in these RI schools should also have the opportunity to be involved, or the existing divide is maintained.


  11. School direct had given primary schools the opportunity to develop the teachers we really need. Our students get qts, pace and master credits. They are also taught by practicing teachers and 100% were snapped up this year by schools desperate for properly trained teachers who understood inner city challenge and had the right moral purpose to succeed. What’s not to like?


  12. A common theme in these posts is that one of the challenges of a School Direct is a lack of consistency of provision. I would agree with this. I also think there are some other consequences of SD which have not been highlighted yet, and which will affect the overall quality of ITE as SD expands.

    Firstly, learning to teach is an emotional journey as well as an intellectual one; student teachers need good support. In my experience of working as an External Examiner the aspect of their PGCE course that student teachers particularly value is the support they receive from their subject specialist, University tutor. The quality of mentoring is variable and the university tutor often helps students to weather the problems that poor mentoring can cause. Crucially the tutor is external to the situation and works with the student for the whole year. In many SD programmes, this role no longer exists and students are reliant on their mentor and other people in the same organisation. The mentor may be excellent in every respect – but they might not be, and that is the point: the issue of support is too important to be left to chance.

    Secondly, universities are very good at teaching subject pedagogy. Student teachers value the opportunity to work with other maths students, music students or science students. Most of the professional studies teaching on school direct programmes is generic and the amount of subject specific pedagogy that student teachers learn on SD programmes in the ITE year is often limited, and again, very variable. It is often down to individual departments to provide this input. Again, the provision might be excellent, but it might focus on operational issues rather than on ideas about pedagogy: it is too important to be left to chance.

    3. There has been under-recruitment for two years now. Schools recruit people into SD who they would like to employ. Universities recruit people who they believe have the potential to be good teachers. There has always been concern about wastage in the system – people who train but don’t go on to work as teachers. I wonder if there is a new sort of ‘wastage’ emerging – people who would like to be a teacher and have the potential to be a good teacher, but don’t get the chance because they do not come across as employable before they start to train?

    4. University providers put the student teacher at the centre of their work. If issues arise and decisions need to be made, the student teacher and their learning is the priority. Schools, quite rightly, put their pupils at the centre of what they do. I have encountered a number of instances in which decisions that schools have made have not necessarily been in the best interests of student teacher learning. This is particularly evident in the ‘squeezing’ of the placement in a different school, to a few weeks.

    My advice to potential students would be to spend as much time as possible in a school before you accept a place and ask of a SD provider the following questions.

    1. What support will I get? What happens if I don’t get on with my mentor or if I find the course more difficult than expected. (A failing ITE student takes up a lot of time and energy. In my experience, schools are quicker to give up on someone than universities – and many people who experience problems are eventually very successful).

    2. How and where will I learn to be a (subject) teacher. Where will I learn about …..(eg Robin Millar’s work on practical ( science) , Gary Spruce’s work on critical pedagogy (music)).

    3. Will I be allowed to try out different approaches as I develop my own teaching personality? ( In my experience student teachers are increasingly being told ‘ this is how we do this’ and are always being encouraged to try alternative approaches.)


  13. Hi Dominic,

    Firstly this is a great initiative. I’m not remotely an expert in ITT as I focus much more significantly on the ongoing and continuing development that happens in schools. However, from what we see through our lens here at the Teacher Development Trust, I’d suggest that there are a number of key issues around ITT and early-career development: the content and balance of theory & practice, and consistency (in many areas). I’m sure others see this differently, I merely offer these thoughts to add to the debate and work toward some professional consensus.

    In terms of theory/practice, it seems to me the biggest issue is a lack of common understanding or agreement over what a new teacher needs to be able to know and do. While the old standards were both fuzzy and bureacratic, the new standards are just even fuzzier albeit less bureacratic. I think it’s laughable that the expectations for what a professional should know and be able to do are written on four pages. No wonder QTS is not universally valued if all it means is an entirely subjective adherence to these.

    ITT cannot be improved until we have a clearer, collegiate and collective view of these expectations so that we can then both support our training institutions to achieve this while also holding them accountable for doing so.

    Solution: don’t wait for a Royal College of Teaching to do something about this, start the work around new standards. Any form of QTS should not happen at the end of year 1, it should be year 3 or 4, accredited jointly between HEIs and schools – both would need to sign off. Given the importance of subject knowledge and pedagogy there should be different standards in each starting specialism. You might consider commissioning the Expert Subject Advisory Groups to begin considering this for each area.

    ITT Standards need to include more specifics about the theory that teachers need to know and then we need more common mechanisms for ensuring that teachers understand these – e.g. move toward common exams. The content needs to be determined through debate, research and consensus and then each standard needs evaluating through, e.g., RCTs to ensure that having the standard does lead to improved outcomes. I’d tentatively suggest that we should be looking at areas such as:

    – Some basic psychology (e.g. motivation, mindset, self-efficacy, attachment)
    – Some basic ‘learning science’ (e.g. the basics of memory, retrieval, encoding, spacing, testing, etc)
    – The principles of effective assessment
    – The principles of effective research and evaluation – becoming research-literate, critical consumers who can start to consider how to evaluate effectively
    – Common learning issues in the specialist subject area: e.g. misconceptions, proven ‘best practice’ models and approaches.
    – The characteristics, diagnosis and approach to working with common SEND issues

    Once we’ve come to more agreement about the theory, and decided minimum expectations through a more consistent examination, we should also consider more carefully what sort of dissertation we expect. I think there should be at least one more theoretical dissertation – two or three over the course of the first few years, ideally. This should be accompanied by a more clearly structured ‘teacher enquiry’ which has clear impact on pupil learning outcomes. All of these dissertations, enquiry projects, exams and other more informal learning and reflection could go in to an extended portfolio which can be submitted when teachers are going for interview and also built upon for the rest of the teacher’s career. Clearly this portfolio has elements which are accredited by HEIs, others by schools, others by both – this is as it should be and mirrors the most successful school systems internationally.

    Another huge issue is around consistency of experience and opportunity in the ITT route. The quality of mentoring is hugely variable – I’d suggest we consider whether there should be a qualification/accreditation either for mentors or, perhaps more practical, for those who oversee mentors. These people would be supported to improve the quality of mentoring while also held accountable for consistency and impact. Again, we need to be clear, consistent and well-aligned on what ‘good’ looks like and then plan from there.

    Consistency of practical experience and opportunity is problematic. We should be clear that every teacher should experience an exceptional school for a reasonable amount of time, and also, heavily supported but not in the first year, a stint in a struggling school. I heartily agree with previous posters that every teacher should spend a decent chunk of their time in another phase – e.g. primaries spend time in secondary, and vice versa.

    In the first 3 to 5 years every teacher should have an opportunity to lead a whole school or even cross-school project with leadership involved – this should come with expectations about what support and training for leadership is given, along with how to work effectively with colleagues around coaching. Through this process we can more systematically develop leadership.

    We are also now finding that there are certain technologies that may have huge impact in teacher training. Use of video and in-ear coaching, for example, should be more thoroughly tested and then rolled out in a more consistent manner. Going further, EEF should include a remit for researching teacher development, both initial and ongoing, and this should be built in to a regular review-cycle of standards and induction. Eventually this would be taken over by a Royal College of Teaching.

    Finally (for now), we need to consider retention and morale. Day and Gu’s work, among others, points us toward the key reasons that teacher stay within the profession or leave it. We should be clearer that wellbeing of teachers is a key requirement of early career training and induction. Schools who ‘burn through’ a huge number of teachers, sending them out of the profession, should be penalised even if they are achieving extraordinary results for their own pupils – the consequences for the rest of the system are far too serious to consider this acceptable.


  14. Very briefly, one flaw I would note with the ‘on the job’ model is that new teachers do not get a chance to see what different school contexts look like. On the 4 year BEd course I did, I got to see a breadth of different types of schools, systems, methods, age groups, behaviour approaches, etc. This was hugely valuable. There is a danger that teachers get trained in a particular style/approach and then find it hard to adapt to other contexts, or to understand why what works well in one place might not work so well in another. So as David says above, a specific requirement to spend significant time in another phase (or ‘style’ of school) would be useful.


    • This makes sense to me but I would be reluctant to have hard rules set by DfE on such things as they will often get them wrong then not want to change for fear of ‘u-turn’ accusations etc.

      Isn’t this the sort of thing that needs experimentation and rules/norms/customs will emerge from successes and failures?

      Thanks for your comment and best wishes



    • I agree with Sue. Many student teachers show a ‘dip’ in progress when they move to another school. It is getting through that ‘dip’ that often enables significant progress. Many SD programmes offer their student teachers a wonderful range of short experiences. In my view it is 8 to 10 weeks engagement in another setting that produces deep learning.
      I do think ‘substantial’ needs to be defined by Government (see below) because the reality is that this requirement can place demands on a training programme. If it is difficult to organise then it is dropped, not for sound educational reasons, but for practical ones. The student teachers are the ones that loose out.


  15. 1. ‘School Direct’ is proving to be too confusing to the majority of potential applicants. ‘PGCE’ has a common currency which applicants understand. SD can be anything from working as an unqualified teacher via the salaried route to a ‘traditional’ PGCE with school-based application process and placement arrangements. There needs to be much greater differentiation in the use of terminology which is reinforced through central marketing of routes in to teaching which focus on enabling potential applicants to make better choices. (The confusion about the availability and value of salaries and bursaries and the different level of awards MUST be overcome centrally – it cannot be left to individual provider to attempt to explain the differences).
    2. UCAS TT is not fit for purpose – a multi-provider training system cannot operate effectively where information is so poorly communicated. With 900+ providers the search engine must be upgraded to allow applicants to find providers local to them (rather than in a particular, and poorly defined, region). The uploading of information to UCAS TT by providers is overly complicated and fails to take account of individual secondary providers working with different HEIs for different subjects (for example).


    • I agree with (1) up to a point – I think DfE’s primary responsibility in marketing teaching should focus on the various benefits, both financial and as a career in general, of the teaching profession. We need a persuasive and pithy pitch to graduates of the wisdom of choosing a career in teaching over the other options available to them. The specifics of routes come a little later, I feel, once the graduate has been ‘hooked’ by the initial sell – though yes, it is better for prospective trainees to a central, authoritative point of contact explaining these technical aspects (upon which school direct partnerships can draw).

      I also agree that the different routes under the School Direct could do with differentiation (not to mention SCITTs).

      One systemic point I’m aware of, albeit *sounds alarm* anecdotally, is a perception that schools want a more ‘oven ready’ applicant; lack of experience in recruiting trainees means that they have less of an idea of how an applicant can be shaped and improved. The universities, on the other hand, are more willing to take on applicants who aren’t yet the finished article (though conversely they are less worried about being lumbered with a dud). I wonder if this problem will be resolved over time, or whether the (legitimate) self-interest motivation will be a long-term problem for the School Direct approach. It is a microeconomic solution to a macroeconomic problem (though principally sound in placing schools themselves in charge of the supply of new teachers). And of course systems never change willingly.

      To point (2), I wonder – does a perfect application system exist, even hypothetically? To say it isn’t fit for purpose does seem to overstate the case, unless there is a crisis in trainee numbers that can be directly traced to inadequacies in the application process. I agree that it does particularly disadvantage School Direct providers though, not least for the example given.

      Anyway, my specific recommendations to counter the problems identified.:

      1) A central campaign promoting teaching remains a sound idea, indeed increasingly necessary as a competition for graduates increases in a recovering economy and a demographic bulge in youngsters requires more teachers to be recruited into the system. HOWEVER – School Direct partnerships themselves need to receive more/better support in selling themselves to prospective teachers. Some, I’m sure do a sterling job; they need to share their practice with their peers. All need to align their individual campaigns with the national pitch (at least making sure the facts presented are consistent).

      2) School Direct schools should either all offer a PGCE, or none of them should. Either we want this qualification or we don’t. As it is, I believe the fact that ‘PGCE’ is synonymous with ‘traditional teacher training in a university’ puts prospective teachers off School Direct – when in fact they probably would get a PGCE with School Direct anyway.

      3) Look again at the application process. From an applicant’s point of view as well as that of the School Direct school, a single portal containing both the persuasive argument to choose teaching, the information on the routes and funding available available and the application system itself would make the most sense. This would mean taking the process away from UCAS and either placing it with an executive agency of the DfE or with a separate body such as the mooted ‘Royal College of Teachers’.


  16. My concern with the rise of SD and SCITT provision is that schools can have a short collective memory with a high rate of turnover of staff in some subjects. This can lead to inconsistency of provision. SD also leads to a huge increase in the amount of time that a school must devote to advertising and interviewing candidates. Returning the responsibility of this to the universities would cut down on the admin. It would be great if the teaching schools were involved in the recruitment, but the admin would be better done regionally by the universities.

    It is also my understanding that the current SD allocations appear to be skewed. Here in the North East there have apparently been very few English SD places allocated this year – and yet schools are looking for more English teachers. A better overview of the training needs of the whole country would be helpful. Having the majority of places administrated by the Universities would also mean that there were fewer smaller places which would help with planning.


    • Thanks. One of the problems the DfE has in its central planning model is that its numbers are ALWAYS wrong. Like most central planning systems, there is an abstract rationale that does not seem stupid at first glance. However, as soon as one digs into the entrails, it is obvious that it can’t work as constructed. In particular, it is always behind the game in spotting trends and its data is unreliable for planning purposes. Theoretically one can imagine improvements to a central planning model but it seems to me that it would be much better for supply/demand to be balanced in a decentralised fashion like the workings of the immune system. I admit this is a general preference of mine but having seen the central planning approach I think it is worth trying a bottom-up approach! Though also NB. limits on maths/physics/computer science and some other subjects were removed last year so there are no longer any targets/limits – HEIs can recruit however many they can persuade…

      Re regional level admin – that sounds sensible but again I would urge that people organise this on the ground, NOT ask DfE to set up regional admin or you will wait for years and it won’t work well.

      Best wishes



  17. Hi Dominic,

    Great to see this blog from you. Hurrah! As you have stated, those in ivory towers are unable to make such a move due to Whitehall restrictions placed upon them. Pleased to see this and guarantee this will not be wasted time on your part. Looking at the Twitter feeds and comments above, you have already started a movement!

    Having led ITT in several schools in North London over the past 15 years, here are my thoughts:

    1. Huge disparity between ITT provision at university level and educational research which filters into the classroom.

    2. School Direct (very confusing – reminds me of Specialisms & Diplomas) led through hubs of Outstanding graded schools in LEAs. It can be decisive if Good schools believe they have something to offer re. CPD. Take away this requirement to have this badge of honour. There is much good practice going on in Good schools. Identified hub schools that receive funding charge schools for ‘services’, which is fine and dandy, but not for the sake of ‘making a buck’.

    The other issue is one of trust. What applications do you and see and what do you not see. Hubs schools control applications. This can be divisive. Allow all school based lead coordinators access to online CVs for trainees. Otherwise, there is a danger of ‘creaming’ off the trainees for specific placements. How do we ensure this is transparent? There are other ITT providers who also do the same. There is no robust mechanism in place as far as I see it. Also agree with Alex: ‘lesson plan templates and course outlines, can be accessed publicly to ensure transparency and quality.’ The stipulation by some is unnecessary bureaucracy when in fact, it could be more streamlined to save everybody time and place ‘trust’ back into the hands of the teacher, the mentor and the ITT provider.

    Let us also not forget, the the quality of mentioning varies considerably from department to department within schools. Something must be put in place to ensure ‘experienced’ teachers are the right people for the job and can narrow the consistency and quality control margins.I think this is key and is the responsibility of each school. How do we address this?

    3. The paperwork required to be completed by those in charge of School Direct / SCITT trainees within their schools is hideous. It really is! We do not have time to do this in and amongst all the other requirements of our jobs. The standards required and quality assurance also varies significantly from institution. As Debra Kidd has said above, consistency is shoddy.

    4. I’ve managed to engage with SAGE publications and BELMAS to make academic research more readily available to teachers in the classroom. There is much to be done here for research education. I fear the bridge between academia and the classroom is far too wide until there is more joined up thinking and schools give teachers ‘more time’ to engage. This answer. More money to release teachers from their 90% timetables to complete action research. On that note, give teachers funding to complete research projects during their career to ensure they keep up to date and can implement school-based action research to feed back up to academia. I cannot see many headteachers giving time to teachers, because of funding, staffing and priorities dictated by you know who! Ofsted goalposts.

    5. QTS – yes please!
    6. College of Teaching – yes please!

    7. Trainees having direct experience in the classroom – yes please. Ignoring quality for the moment… my own BAEd route gave me four 8-12 week placements in various schools whilst I was training. This was invaluably and 20 years later, I firmly believe, the best form of teacher CPD is observing lessons and visiting other schools.
    A PGCE route offers a more meaningful route in 2-3 schools over roughly the same amount of time. I am a huge advocate of TF and other ITT models, but the trainees need to spend MUCH more time in the classroom, before being ‘left alone!’ It should be a requirement that schools host trainees for a ‘set period’. ENSURE ALL trainees see as many schools as possible before the start working full-time in post. I see this as a vital part of the process.

    8. Ensure ITT are fully up to speed with T&L dialogue. Danger is myths are filtered down to trainees and can be damaging. e.g. 20 min progress in a lesson. AfL/Differentiation in every lesson. Solution: consider tutors to be up to speed with classroom practice. Many may observe countless lessons, but have they taught x30 screaming year 8s? Not sure on the validity/solution to this. Getting it wrong, can be emotionally damaging to trainees. We forget that what we learn in our early trainee-years can shape us for the rest of our career. It’s important to get it right from the start with great pedagogy and good/strong/consistent mentoring.

    9. Address recruitment and retention issues across England and Wales.
    10. Please can we see an outcome of this blog? A meeting, a paper, a roundtable meeting. The DfE and all political parties will definitely *prick their ears.

    In a nutshell; consistency, transparency, time and unnecessary paperwork.

    p.s. I’m glad to see you set the ‘comments’ standard from the beginning of the time-line. Pertinent and polite!

    Sorry for bombarding your time-line. The opportunity was too good to miss out.
    Ross – @TeacherToolkit.


    • Quick thought – invite an HMI to speak to trainees so that the inspection process can be described and some of those worrying Ofsted ‘myths’ can be ‘explained’ before they take hold.


  18. Always prioritise the subject and be wary of anything generic. Although there is something to be gained from genericism (e.g. what role does practice play in learning?), ultimately these discussions are fairly meaningless unless considered in the context of a particular subject discipline (e.g. what role practice plays in learning maths / German / literature?) Generic discussions about assessment, planning, pupil learning, ‘critical’ thinking, questioning, and so on can be superficially interesting, but – in my experience anyway as a trainee and mentor – these discussions can never be as useful, purposeful or practical unless they are addressed within the subject domain.

    Generic, non-subject specific feedback on lesson observations is often poor. For example, I observed a lesson as a mentor where my trainee was teaching a lesson on the causes of the English Civil War. The trainee had got the pupils looking at ‘ship money’ as a cause of the conflict, but the pupils were saying banal things such as ‘the king was taking money and so people were angry’. To improve, this trainee needed to show the pupils that there was a tradition of taxation in place since the late middle ages and that Charles’ use of ship money was a break away from what was considered reasonable taxation at the time. A non-subject specialist would probably (a) not notice that children saying ‘the king was taking money and so people were angry’ was actually a problem and (b) not know enough about the subject to help the trainee improve the quality of the lesson. Instead you get banal comments about ‘engagement’.

    So whatever format teacher training takes, it needs to be highly subject specific. I think it also has a number of flaws, but the impression I get is that the German / Scandinavian tradition of ‘Fachdidaktik’ is much stronger than our ‘subject pedagogy’.


  19. I was a teacher who trained through the traditional route of degree & PGCE. Personally I really appreciated being given the time to think about education & learning. I found the PGCE year really helpful. The space it gave you was a luxury, but a useful one, and one you don’t get once you are teaching. I feel it gave me a really solid foundation to build on. There were excellent bits and less useful bits, but that’s life really isn’t it?


    • I don’t mean this rudely but this is not the most helpful of comments.

      The purpose of these comments is not to give feelings but to provide actionable information for those making decisions amid uncertainty and chaos.

      Best wishes, D


  20. For clarity and for the benefit of interested non-experts, can we spell out acronyms etc at first point of use….


  21. Combine training with something routine all teachers need to do so it doesn’t end with the initial training. As an example use routine coursework assessment as a focus for feedback. This is what we have set up with the computing baseline testing project with NAACE. It’s free. Around 400 secondary schools so far have made accounts >10% take up in 2 months with only social media for advertising. 850 teachers in the group mailing with more joining each day. We provide on-line tests and we can give them feedback on which questions a lot get wrong and a lot get right. We link these to assessment criteria accredited by Ofqual and derived directly from the new NC POS. We provide guidance and this will increase as we get more feedback. All creative commons licensed so they can contribute back, remix and re-use. The assessment criteria can be evidenced with on-line management all built from open source software and all free to all schools if they want to do formative assessment. We have shown can run a national testing and feedback program inexpensively enough to do it free. We can test any student or any group and compare it to a nationally representative sample. We can use this to track progress of groups and individuals and train teachers how to use the data objectively to inform what they are doing.

    Once there is a focus and a critical mass engaged in it, dissemination of information becomes possible and active research can start within a growing pool of peer reviewed expertise. It’s not about money, it’s about doing things better and using technology intelligently – like many open source communities all over the world.


  22. For those who make it this far, at the weekend I am going do another post summarising the comments made so far, then I will post that and ask for corrections/improvements. Thanks to all who have commented so far, I’ve already had interesting feedback from DfE. D


  23. I’d agree with the comments above about the variability of quality in school-based mentoring, which was a central theme of my recently-passed (last week!) PhD thesis (“A Study of Mentoring in the Teach First Programme”). If I may, I’d like to include an extract from the conclusion, which covered the implications of my findings for system leaders. If anyone wants to know more they can get in touch, I’m happy to discuss further or share the thesis.
    PS. There are plenty more thoughts in the conclusion about Teach First specifically, which may be of interest to DfE/NCTL/TF but are not directly relevant to this thread.
    All the best from sunny Germany.

    “There are a number of implications for those overseeing teacher recruitment and development in England at a national level, including the Department for Education, NCTL and Ofsted. These focus on how the transition to school-led ITT and CPD can be managed most effectively. Firstly, this research has shown that not only is there a diversity of approaches to mentoring amongst different schools, but that significant variability in quality is a persistent feature of mentoring provision. System leaders should seek to build and support the capacity of the school system to facilitate high-quality mentoring on a national scale. There are a number of possible approaches to resolving this long-standing challenge. This research has shown that mentors in schools perceive there to be a deficiency of time and resources available for mentoring in schools. Allocating additional funding to schools for mentoring is probably necessary but may not, by itself, represent a complete solution. System leaders should also seek to share and propagate those cultures of excellence in ITT and CPD which already exist in schools; the obvious vehicle for this is the network of Teaching Schools. The shift to a school-led model of ITT creates the tendency for a diversity of approaches to develop, and the devolution of needs and initiatives to the level of individual schools; to counter this tendency, the community of Teaching Schools needs to be made aware of its responsibility to national priorities. When considering the number and specialism of teachers to recruit for school-based ITT, these schools need to look further than their own immediate staffing needs. Concerns about the allocation and supply of teachers with particular subject specialisms through the School Direct route have been already been expressed (Ward, 2013); this is arguably the result of individual schools responding to their own needs rather than having a national view. System leaders should therefore consider how Teaching Schools could develop this awareness and responsibility, perhaps by devolving the allocation of training places to a ‘parliament’ of Teaching School leaders mediated by the NCTL, or assigning this role to the recently-proposed College of Teaching (PTI, 2014).”


  24. All I have to go off is personal experience so I will try to explain what I think are the areas that need improving from the SD placements we have had this year.

    We have had a SD trainee with us from a lead school this year and the biggest problem we have had is lack of information. When we have asked, the lead school blame the partner university and the university blame the lead school. This lack of clarity of what exactly is being asked has lead to the trainee being unclear of her responsibilities and us being unclear what we were supposed to provide. After lots of chasing people, we eventually tracked down what we were supposed to be delivering. Training/briefing needs to be given to any partner schools so they know exactly what their roles are and what is required.

    Our trainee came to school with very little experience in schools and I think this needs to be addressed. The trainees should have had quality experience in schools, not just listening to readers and washing up paint pots. We are a primary school and teaching 6-7 year olds can be quite challenging. Just being able to communicate with them on a level they understand is a huge obstacle that can make trainees stumble at the first hurdle.

    We also take PGCE and BEd trainees from a different university and they are much better prepared as they have time in university to study some pedagogy before putting it into action. This is vital and prepares the trainees well for what they may encounter in their placement school. Our SD trainee started on the first day of the year and had no input from the lead school of university and was expected to just teach. We had a lot to discuss! Key areas such as AfL and differentiation needed to be addressed and explained before we could go any further.

    The larger role of the school in Schools Direct (SD) training also means that the trainees need to have schools that contrast in their placements. Our trainee had two schools in similar areas with similar intakes. This does not prepare the trainees well for what they may encounter when they qualify. The lead schools must ensure that a variety if placements in a range of schools is experienced by the trainees.

    I also agree with what has been written previously regarding mentoring. A national standard would be helpful to ensure consistency throughout the country.

    Finally the paperwork we had to complete was immense. The responsibility of teaching a class and training a new teacher was almost a stretch too far!

    I think the key things I would change are:
    Key information and training given before trainees arrive;
    Trainees to have had more quality experience in school before they arrive to teach;
    Trainees to have had time to study some pedagogy before arriving in a class to teach;
    Schools in which trainees are placed should be contrasting;
    Release time given for the completion of the assessments of the trainee.

    A one size fits all approach will not work. Primary schools are very different places to secondary schools and this should also be taken into account when making improvements to the SD ITT.

    Gaz Needle – @gazneedle


  25. Initiate school direct in effectively the Assessment Only route only.

    The big issue is – students paying £9000 for training and no salary = poor recruitment. This also forces weak training as schools and providers are scared to upset the trainees and fail them. SO……

    Schools to have rigorous selection procedure, before appointment – then pay UQ rate or similar and train people yourselves (There might be a need to go through a teaching school or uni to keep standards of mentoring high, and stop abuses. Currently we just do it ourselves and get uni to rubber stamp at end of two years – and pay £2000 for the Assessment only)

    Depending on the initial strength of the trainee detirmines how much teaching they do, but they are immediately put to use – eg teaching small groups / intervention. They are immediately an integral part of the school – not an add on as in most training schemes. They are being paid!

    If after a 3 months / 6 months it becomes apparent they are not fit for full classroom teaching (This is where you need good in house systems) you either keep them as a HLTA etc or they are told they will finish at the end of the year. This could be open to abuse – so would need some thought.

    What I am suggesting is effectively the old school funded GTP. The selection at the beginning is the key. Rigorous interview leading to 1 week trial – paid, then 2 year programme. Currently – but this could be cut to one year)


  26. Several observations

    (1) The PGCE is a qualification, not a course; it’s possible to secure a PGCE through a number of routes: a relatively conventional university-led partnership; a SCTT-programme accredited by an HEI etc. By the same token, School Direct is not a training route but a mechanism for managing training routes. So the first observation is that, following David Weston, we won;t get clarity about what we want until we are clearer about the knowledge base for effective professional practice – debate about routes, however well-informed (like some of the comments here) or ill-informed places a second order concept (organisation) first.

    (2) The market is over-supplied and over-regulated. There are approximately 80 university departments of education in England. By comparison, there are 25 medical schools. Current NCTL pracxtive is to accredit more providers – with new SCITTs we are well on the way to 150. We are constructing a market which is simply over-supplied, and where the economics really cannot work. We should be seeking to reduce, not expand the number of suppliers. At the same time, tight regulation (e.g. on number of days in school) means that programmes all look very similar to each other: there is little scope for innovation. A far more radical approach would be to specify more tightly the outcomes (as in law and medicine) and competencies (David is right that the 2011 Standards are just too feeble) but then de-regulate on process.

    (3) Michael Fordham overstates the case against generic – there are some generic issues (legal issues, behaviour management), but this is tightly connected to issues of scle of provider. The assumption under-pinning whooly-school led approaches to SD is that – at its extreme – a Physics teacher who graduated 25 years ago and has mentored a Physics student every couple of years is more likely to be able to train high quality new Physics teachers than a specialist. This is not an argument for a highly conventional approach but it is a rationale for getting real about the economics of teacher education. It should be possible to evolve relatively quickly to a future with 20-30 high quality university departments of educationworking each with a network of teaching schools, developing locally-structured programmes within a national outcomes framework. Delivery of the programmes would (rather like GP training) be managed by the university department but substantially delivered by ‘clinical teachers’ – essentially, practitioners with part of their FTE in the training system. This would be a much stronger CPD system, a much better ITE system and secure a stable structure with realistic economies of scale.


  27. Dominic

    This is an interesting experiment. Thanks for initiating it. If it works well we may look to do something similar for FE on the Education & Training Foundation website. We have already run a more contained wiki-type exercise to flush out what practitioners think the leadership challenges are of the next 10 years. It produced some interesting results.

    There is, as you say, no good reason for Government not running wiki-discussions. Social media are beginning to influence national policy and practice, and Govt should be shaping and harnessing that power, not being a passive observer or reluctant participant.

    (Nothing to contribute on School Direct; sorry.)


  28. There are a number of advantages to being part of a university–school partnership.

    First, university departments are ideally placed to help build social capital with and between school partners. This can take the form of improving and developing teachers’ subject knowledge. Universities can provide access to real experts in the field who can
    work with science educators, novices and school-based teachers to make recent ideas accessible and available in school classrooms.

    Second, collaborative relationships between schools and university departments are underpinned by a shared understanding of how research knowledge and practice knowledge intersect to inform practice about, for example, helping new teachers to engage pupils in learning how to learn. Effective teachers are constantly called upon to make deliberative judgements about practice. This is learned best when experienced in both a school and a
    university. For example, learning in the university about the latest ideas as to how children learn and then finding out in the school how to teach specific students in particular classroom, new teachers can develop practice and research knowledge through undertaking small-scale school-based research guided by university staff.

    Third, university-school partnerships are able to set up opportunities for novices to hone their practice in different schools. At Cambridge novices benefit from two major practicum experiences in at least two very different schools. Furthermore, the opportunity and time to reflect upon practice between placements may help the novice to develop thinking about practice, as there is little time to reflect and think on-the-job. Additionally, working
    collaboratively with groups of schools and university departments cultivates a sense of identity where novices feel secure, supported and trusted which would go a long way to helping to retain more teachers in classrooms.

    Finally, teacher education institutions also serve as key change agents in transforming education and society. Not only do such institutions educate new teachers, update the knowledge and skills of existing teachers, create teacher-education curricula, provide professional development for practising teachers, contribute to textbook production and consult with local schools, they often also provide expert advice to international
    ministries of education. Because of this broad influence on curriculum design and implementation, as well as policy setting within educational institutions, faculty members of teacher education institutions are perfectly poised to promote teacher education in the longer term. Indeed it could be argued that short term policy responses based on perceived teacher shortages in urban areas might jeopardise existing good practice. By working with faculties of teacher education institutions, governments might be better placed to bring about systematic change.


  29. I’m a humanities lecturer at a big university. My contact with PGCE issues comes from a) writing references for students who go into teaching and b) dealing with incoming students and what they’ve learned at school.

    Primary and secondary schools do need more contact with university staff, but not with the education departments. There are some fantastic education lecturers. But most of them are ex-teachers who wanted an easier life and never had a real interest in research.

    Education researchers and lecturers should have backgrounds in specific school subjects, and also in psychology,neuroscience, public policy, social science etc. Maybe education shouldn’t even be a university subject any more, but a cross-disciplinary centre for PGCEs.

    Teachers should get more of their training from subject specialists in universities. Trainee history teachers should get their desk based training from history specialists who can give them the latest research on topics which will be taught in schools. Teacher CPD should consist of this as well.

    Finally, as a parent, teacher training still hasn’t got to grips with discipline and bullying. Again, proper training in the social psychology of large numbers of children sharing a space would be more helpful here than empty statements that eg ‘we have zero tolerance of bullying’ by people who know nothing of whether or how ‘zero tolerance’ approaches to rule enforcement work.


  30. If anybody is eagle-eyed, you will see that a comment appeared then disappeared from someone saying they are a humanities lecturer at a University who deals with PGCE issues.

    The person who left the comment then emailed me saying that they disapprove of anonymous comments critical of colleagues and wished they had not left it and asked me to take it down.

    I approved it before seeing the email, then saw the email. Having thought about it, I have taken it down. It was not removed because I disapproved of it.

    The heart of the point, stripped of the criticism of colleagues, was:

    ‘Education researchers and lecturers should have backgrounds in specific school subjects, and also in psychology,neuroscience, public policy, social science etc. Maybe education shouldn’t even be a university subject any more, but a cross-disciplinary centre for PGCEs. Teachers should get more of their training from subject specialists in universities.’



    • “Teachers should get more of their training from subject specialists in universities”:

      This is EXACTLY what happens in Spain. As a recipe for: (1) driving a wedge between theory and practice (2) divorcing subject based training from the realities of the classroom and (3) generating student/trainee dissatisfaction about unjoined up training, it is an almost perfect device.


      • My experience too. While there is a lot of folklore and black art in the pedagogical base that would not stand up to proper scientific scrutiny that does not mean experience counts for nothing. With respect to HE subject specialists they are good at what they do but their effectiveness in teaching methods obeys some inverse power as children get younger 😉


  31. I was at different times, the course leader for a PGCE ICT at two Midlands Universities. At the start of each academic year it was routine at one of them to send out emails to schools pleading for places to host trainee teachers – thankfully the schools obliged. It could be argued that this was a deficiency in our planning, and that would not be entirely unreasonable. However, staff in schools do move, they change syllabi and it takes time to get the delivery of new courses right before hosting a trainee: A case in point being the new computing PoS, the mentor might be mentoring an NQT and not have the capacity to take on a PGCE student at the same time. There are many reasons why schools might not be able to host a trainee. This impacts on the quality of provision and the trainees subsequent experience. I’d suggest that in order for schools to be graded outstanding/good, there should be a requirement that they engage in the training of new entrants to the profession.

    Mentors should not be NQT’s but teachers with at least two years classroom experience. Similarly, teacher trainers need to have recent experience of the classroom.

    Subject knowledge enhancement courses should be standardized and the content benchmaked at an agreed level. Individuals can be accepted onto a PGCE course without a relevant degree, but a short GCSE type course can be considered equivalent for subject knowledge purposes. GCSE equivalent courses in mathematics and English should be scrapped. If an applicant does not have a GCSE in either mathematics or English they should sit and pass a GCSE course.

    Accessing research journals is difficult (costly) for most teachers unless they have access to a University library. Professional membership of (say) a National College of Teaching could allow for online access to most relevant journals. This assumes that a NCT subscribed to them in the first place!


  32. To follow up on a couple of comments on R&D above.

    We have almost no robust evidence about the best methods of teacher training. We will never get any credible evidence unless the DfE invests significant amounts in research. Whilst the funding for the EEF is welcome, its remit is limited and overall research spending by the DfE has fallen significantly in real terms. Similarly, investment by the ESRC has significantly fallen in real terms.

    Without investment to find out what works, these questions will never be resolved. Gove and the DfE appeared to acknowledge this in their response to Goldacre’s report. However, I’ve seen no evidence that any of Goldacre’s recommendations have been followed up with funding.

    So I’ve a question for you and your readers, what proportion of the DfE’s budget should be invested in research every year? Some context: the DH spent 0.94% of their budget on R&D in 2012-13, the EU has a target of 3% of GDP for R&D. In comparison DfE figures suggest research spending was £12.2m in 2012-13, that’s 0.02% of a budget of £52bn.


    • Good point.

      I’m sure Tim Leunig will be considering this.

      Unfortunately, because of decisions taken in the ‘quad’ (DC/GO/NC/DA) about political tradeoffs, the DfE has been forced to cut its budget recently in ways that are clearly irrational. Eg. pushing billions into childcare programmes with no RCTs while cutting 16-19.

      Hopefully the strengthening trend towards research will help the next government deal with the issue you raise. External evaluations of research are vital to stop ministers picking ‘evidence’ as they want – i.e. using evidence as advertising for political choices. We need mechanisms in education so that political choices are explicit rather than implicit.



  33. To those who have commented…

    I am going to leave this thread as it is until Sunday/Monday, then do another blog summarising / clustering the comments and publish that (Monday), in the form of a note to ministers / spads / officials in the DfE. Then people can send corrections / additions etc, and I’ll redo it, then post a final (for the moment) version.

    Thanks to all who have contributed so far. I know many of the relevant people in the DfE have read your comments so hopefully some good will come from your efforts…



    • A quick postscript; I agree that ITT colleges should not have to beg for placements from schools. I understand that schools in difficulty may not want to take trainees – indeed some years ago I mentored three PGCE students who I had to “rescue” from a class where the teacher was literally having a nervous breakdown! However trainees now are being put into unsuitable provision which cannot enhance or help their professional development; our primary PRU recently took a lady on who wanted some experience of challenging behaviour before her final year. She had recently had a placement in a private religious school which was in the front room of a terraced house with no outdoor space and no facilities. She was expected to buy educational resources with her own money and her pleas to the college fell on deaf ears; they lamely proclaimed there was nowhere else!!


  34. Thank you for the invitation, particularly welcome as there doesn’t appear to be a mechanism for formal evaluation of the impact of SD on schools so far. SD has delivered better than expected outcomes for us, but at a price. This cannot be explained by the inadequacy of specific regulations alone, but the failure of organisations and different bodies to interact with each other positively towards achieving better outcomes in school, if indeed that is the aim of all stakeholders, and it is sometimes hard to tell. Example: unresolved issues around intellectual property rights over course materials and whether schools or training providers are accountable for recruitment are unhelpful distractions from the core purpose of getting good teachers into schools.

    SD is a very bureaucratic and resource intensive system, considering it was intended to open up possibilities. Choosing a training provider is a big decision for a school, yet there is little emphasis on clarity from providers about what they are actually offering. Some appear to have got away with offering their PGCE courses as School Direct, with few changes, or even none. The ITT criteria are vague about how many days in school an employment based route should have. Schools have been expected to take a lot on trust from providers. There should be a minimum requirement for SD provision which ensures adequate time in school and does not allow providers to simply rename their PGCE as SD without the necessary changes. Instead of a focus on this from the NCTL the emphasis has been on telling schools they should not change their providers.

    Training providers should be required to be explicit about their School Direct programme offer and make this information available BEFORE the bidding process begins. NCTL’s role should be to ensure this happens, not tell schools they can’t change/ shouldn’t change provider.

    SD is meant to be about recruiting trainees directly to the school. However, training providers often still see the trainees as belonging to them. This is fine for PGCE but not SD. NCTL should be intervening to ensure that the direction of travel for SD is providers being commissioned by schools to provide training for their recruits. Schools should be responsible for final assessment of trainees and not subjected to pressure to ‘mark up’ by providers looking to boost their own Ofsted outcomes. Learning to teach is a journey and we can’t all be outstanding from the start.

    UCAS are very helpful and offer guidance and support in recruitment. NCTL should be actively seeking ways to do the same, for example seeking open evaluation of different providers and what they offer and school’s experiences so far. We have had to work out for ourselves the hard way that providers with experience of employment based teacher training are better placed to deliver SD programmes than those with experience of PGCE only.

    The infrastructure and resourcing required to deliver School Direct favours larger partnerships. Smaller partnerships should not be disadvantaged. A more flexible approach to allocation of training places would help schools to avoid under or over recruiting.

    NCTL could play an important role in brokering trust and partnership between schools and providers to ensure SD incorporates the best of employment based training and training provider input.


  35. We have trained 35 teachers through the School Direct route this year and we have recruited 51 for 2014/15. All receive a PGCE.

    We have had phenomenal success which I put down to several things:
    1 – the quality of the recruitment process. We have been conscious not only to consider academic ability along with subject knowledge but also to consider their passion for the profession, their ability to interact with children & young people and their resilience. We have not simply recruited to fill our allocation. We have recruited high quality graduates who want to make a difference.

    2- all our trainees are inducted throughout the summer term and this continues throughout the summer holidays. Feedback from trainees tells us this was invaluable and enabled them to feel more confident in their schools in September. We cover areas such as vision, behaviour management, professional conduct, including dress & lesson planning et al.

    3- all schools in our alliance who take trainees have to agree to their selected mentors bring trained by us. This allows us to QA the mentors and ensures consistency for everyone. This is non negotiable. We have experienced no issues from schools about engaging with this training.

    4 – high quality relationships with HEIs. We work in close collaboration and have an honest and transparent approach. All parties openly acknowledge what strengths they bring to the table and we work to these strengths. We are now looking to some joint appointments with Sheffield Hallam University to ensure we have a presence on university campus.

    5 – we have developed a tracking system for our trainees which allows us to intervene if a trainee is struggling in a particular area. This tracker is used across the alliance and we provide opportunities for mentors to meet each other to discuss progress of trainees. We have had 100% success rate this year with no trainee dropping out.

    33/35 of the trainees have secured jobs. Of the 2 who haven’t, one has decided to travel and the other is relocating in the south due to family commitments.

    I suggest you access the NCTL 2014 Fellowship Commission Report. We covered a lot of teacher training and recruitment issues and made lots of recommendations.

    To finish, schools need to truly commit to teacher training. We are responsible for training the future of our profession & our children deserve the best quality teachers. I am an advocate of SD. We have invested time & effort in getting it right and we have worked in collaboration in order to do so.


  36. Dominic – this is an interesting exercise. It is however telling that this is perhaps the first ‘open’ consultation that has taken place about School Direct and the DfE hasn’t commissioned this?

    I have been involved as a provider of ITE with School Direct from the start and it is possibly one of the most poorly introduced initiatives that I have been involved with in my time in ITE. However despite this I actually believe there are some very interesting developments with School Direct occurring particularly where there is a shared set of values between the HEI and the lead school and where the development of the trainee is central to everyone’s concerns. I believe the joint selection and ‘expectation’ to employ are positive but I don’t believe that School Direct will address many of the aims that have been claimed. For example I am really concerned about how certain types of school get access to the very best trainees. This was very much previously at the heart of ITE delivery as part of a social responsibility to local schools. I fear the supply of high quality trainees to challenging schools may dry up. I could go on about some of these issues, but this is a priority that needs to be looked at.

    However ultimately School Direct was a political and ideological act to disrupt University involvement in teacher education but many of the elements that the new coalition government seemed to want to attack were myths of a bygone era. Initial Teacher education was without doubts one of the highest performing sectors of government and Michael Gove has said on numerous occasions that we have the best generation of new teachers – so where did these teachers come from? The system clearly wasn’t broken. The introduction of market forces is however driving a wedge (perhaps intentionally) between many providers and schools for the wrong reasons and whilst in some areas the feeling is that this was always about an attack on HEI providers – the reality is that I see certain schools being the losers in all of this. The government is forcing schools hands (and has been twisting arms) to be involved in School Direct but schools are now competing against each other for trainees rather than NQTs and the amount of additional energy needed to secure a single trainee via school direct is a concern.

    I could say a lot more and I am happy to expand on any points above however it is important to note that despite some of the early intention to disrupt ITE – School Direct has created some new and positive opportunities which I do welcome. At its best it has enabled productive partnerships to be developed and where School Direct has been developed around mutual values with clear aims it can be very productive and successful – so there is something worth investigating here. But it does need investigating. The disruption and manner of introduction has however been poor and so much more could have been achieved and could still be achieved if a dialogue were maintained with those who perhaps know something about ITE and where School Direct is not presented a divisive strategy but as a genuine alternative route alongside other routes into teaching.


  37. I agree with David Weston that “the biggest issue is a lack of common understanding or agreement over what a new teacher needs to be able to know and do”.

    Except that:

    1. The word “new” is redundant here – this is a much broader problem with the whole of the way we conceive of what the job of teaching entails. Which is why the problem with ITT cannot be addressed in isolation.

    2. I disagree with David’s proposed remedy, which is to establish a Royal College of Teaching. Such an institution would be captured by the orthodoxies that represent the problem and not the solution. Consensus on what works must be established by bottom-up evidence and contested debate, not top-down diktat or premature consensus based on what people suppose to be expertise but almost always turns out to have got muddled up with authority.

    The problem is that we conceive of teaching as a personal (and often personality-driven) craft, which is not something that can really be taught. Although personality will always play its part in the mix, education at scale requires much more emphasis on pedagogy, managing transactions and processes, what Diana Laurillard calls “teaching as a design science”, or what you might unpack as the design and sequencing of learning activities, the monitoring of progress and the provision of appropriate feedback.

    The reason that teachers do not do this well is that they do not have the tools of the trade that will enable them to do it effectively at the scale that modern education requires. Most training programmes start by introducing trainees to the equipment that they will use: these are the different parts of your rifle, this is how you stand on your surfboard or how you hold your violin, this is how the aeroplane that you will be flying works etc. The major problem for teacher training is that we do not have these tools of the trade – teachers are shown a lot of rather tendentious, theoretical texts, urged to become self-critical and reflective, and sent into their isolated classrooms to work things out ab initio, creating their own materials as they go along.

    You can change the place of learning from university to school but you do not thereby escape the damaging effect of a lack of evidence-based theory. Practice unsupported by robust theory becomes, as many contributors to this list have observed, inconsistent and haphazard. Part of the misconception here is to think that better teaching necessarily means better (or better trained) teachers. Israeli soldiers are not more effective than Hamas soldiers because they are better trained but because they have better equipment (in the use of which they are, incidentally, better trained).

    The missing tools of the trade are what the Nuffield Foundation in the 1970s called “resources for learning”, arguing at that time that we would never deliver education on the scale required by comprehensive (i.e. universal) education while we continued to place so much emphasis on staffing and so little on resources. Time has only proved them right. Their proposed remedy was very close to what Liz Truss proposed when she called recently for a return to textbooks – except that what we really need are materials that (a) represent activities not texts, (b) will generally be mediated by digital means, not paper-based books, (c) be re-mixable and adaptable.

    The policy actions required from government all centre on putting in place the infrastructure required to foster a competitive market in digital learning resources (generally software, not expositive files). I have made three specific proposals:

    1. a technical specifications incubator to improve interoperability – a technical issue that represents a fundamental prerequisite for an effective ed-tech market;

    2. a central catalogue of ed-tech products to provide an authoritative reference point for objective information about product certification and pricing, as well allow for the aggregation of subjective information such as product reviews;

    3. discussion with appropriate media groups about how government can support the development of a professional ed-tech and pedagogical press, the lack of which is another sign of the underlying problem – the lack of a coherent and defensible body of knowledge on which front-line practice is based.

    These are key but there are others, such as the decoupling of exam boards from publishers of textbooks and other types of learning resource.

    Note that these proposals represent an answer not just to your call for views on ITT but also to the call for advice on government policy on ed-tech (currently being addressed by ETAG, whose draft report has not been published), and also to government concerns about the lack of effective educational research. The sort of digital tools I am recommending would collect learning outcome data continuously, producing data in the sort of quantities that are needed not only to manage the processes and transactions involved in teaching, but also to underpin research into what works.

    I have produced a 3 page policy paper at and I expand my rationale in a number of essays on my blog at

    You may say that none of this is about teacher training. My answer would be that if you do not address what is the prior problem, whatever tinkering you do with teacher training processes will be ineffective.

    Thanks for the opportunity to comment. I think it is a great initiative and I hope it has some influence.


    • I agree with some of this BUT –

      1. I am sure you are wrong about Israeli soldiers therefore I wonder what else you are wrong about! A classic example is 1870 – the Prussian soldiers had an inferior rifle to the French but they won easily because of a) Moltke’s General Staff system (a global innovation) and b) TRAINING. Training plus unprecedented decentralisation meant that Prussian soldiers adapted tactically to superior French equipment. This lesson repeats in war. Advantages in equipment are rarely decisive – and just look at Russia and US experience in AFG.

      2. The 3 things you mention re Ed Tech are – even if sensible – far beyond the ability of the DfE to organise. E.g. any attempt to rate products on quality is totally impossible because of legal issues. If they are to happen, they must happen without DfE’s help.



      • Hm, I reckon the US cavalry showed rifles were pretty decisive when fighting the indigenous population that had bows and arrows. The atom bomb dropped on Hiroshima was pretty decisive 🙂

        I agree on point 2. The DfE can lead on things like interoperability by adopting the Cabinet policy on open standards themselves and by training their own staff to understand those issues – after all it is central government policy! But as you say I don’t see how the DfE can start endorsing products. it was bad enough when BECTA started inventing bureaucratic frameworks for procurement that simply excluded smaller innovators that did not have the resources to participate in expensive procurement exercises.

        On decoupling the exam boards from text book providers, the way to do that is to use a disruptive service based strategy to provide all the learning content and text books free using a similar fremium model that Google, Dropbox and LinkedIN have shown can be successful. That does not need intervention by the DfE, just imagination and perseverance to do it. Watch this space 🙂


      • Dominic,

        1. I am very surprised that you are so sure of your ground here.

        Of course there are many factors that influence the winning of a war. Wikipedia’s account of the superiority of Prussia in 1870 is that “The German forces were superior in numbers, had better training and leadership and made more effective use of modern technology, particularly railroads and artillery”. So the example you cite against me appears to support my case and not yours. In a war of rapid movement, it is not the quality of the infantryman’s rifle that is of such key importance.

        The problems faced by America in Afghanistan are largely due to the huge difficulty of wars of occupation when fighting against well-motivated irregular troops, particularly when technology places in the hands of those irregular troops weapons that are effective against the enemy (like the rifle in the Boer War or IEDs in modern wars).

        And if you are right that it is all about training, then presumably you think that a Roman legion would still hold its own on a modern battlefield? I think they were pretty well trained.

        2. I do not propose that the DfE should rate products on quality – I agree with you that would be absurd – but that they should encourage the production of a comparison website (exactly the same recommendation made by Sebastian James) where product quality is rated by users. This is a relatively simple thing to do but requires the moral & political leadership that only the DfE can provide in a state-run business like education.



        • …and if user (or indeed professional) reviews cannot even appear on a DfE-supported website, then it is relatively simple for third-party commercial providers to host these reviews on third-party sites, linked to the government catalogue. It is the difficulty in getting access to the data – i.e. the catalogue – that is the key enabler for these services to flourish. The central catalogue would need to be self-supporting, which it could do by charging a registration fee (after an initial, seed-funded holiday), so its long-term future would not be dependent on the vicissitudes of party politics.


          • I doubt the DfE could set up such a website. If it did, it would be rubbish. And even if it weren’t rubbish the lawyers would scupper feedback systems like you’re suggesting.

            I don’t see why others can’t set up such a website. It isn’t intrinsically hard – only for Whitehall…


          • You are right about the DfE not being able to do it themselves. In the case of Charles Clarke’s Curriculum Online initiative, the DfE tried and hashed it up. So the argument is not about whether the DfE should do it internally, but whether it needs arms-length DfE backing.

            Even though it is technically relatively easy to do, the reason why no-one does it in practice are commercial and to a lesser extent cultural. It is against the interests of the larger, more powerful players on the market to create a more competitive market. Much better from their point of view that they continue to leverage their partnership with government (central, agency, local and exam boards) and control the market. Without those players on-board, how can the central catalogue and its associated services achieve credibility with the teachers?

            Meanwhile, the government continues to promote OJEU Frameworks (such as the current Information Management and Learning Services framework) which provide a short list of larger suppliers, explicitly excluding the smaller players, who generally find that they need to bind themselves to the larger suppliers through partnership agreements. The market is uncompetitive to the point of corruption, mainly as a result of the intervention of government in all its forms.

            When I refer to cultural reasons, I mean that (a) there is an underlying belief amongst teachers that theirs is a profession which is unlike almost any other in that technology and process do not matter; and (b) amongst researchers that their job is to produce academic papers and not to support the development of technology that assists implementation. This perception, that teaching all down to the personal craft of the front-line teacher, is one that is hard-wired into the public sector agencies that run the show.

            Almost anyone could produce the website but making it a success is something different. You cannot expect commercial suppliers to undo what government has done and is continuing to do. i.e. undervalue technology.


        • Don’t want to go far off-topic here but to both of you – I was obviously not saying technology is irrelevant.

          These things are a balance. If you look at all professional analyses of 1870, they say the same thing: Prussia won DESPITE inferior rifles (French Chassepot was much better than the Prussian needle gun) because of Moltke / General Staff planning in peacetime (including the use of railways and telegram) / training and tactical adaptation. Prussia did not win because of ‘better equipment’. Why did Prussia exploit the common technologies of railway and telegram? Because of the new GS culture of training and wargaming and facing mistakes. The French army also had railways and telegrams. Why did they not use them well? Because their peacetime training systems were miles behind Prussia. All this is why European states made such huge efforts post-1870 to mimic Prussia’s GS and training systems.

          This is not my view but the consensus of many studies over 150 years. Cf. Bucholz on Moltke and 1870. My confidence is because 1870 is one of the very few things I’ve studied in detail!

          Of course equipment is important. Of course I do not think legions would be competitive. But look for example at the soldiers from Alexander the Great’s armies – they dominated battles even into their 60s, because of culture/training/doctrine.

          I have strongly supported experiments with new IT in schools and am convinced they could help significantly – I will blog on this shortly – but they must be integrated with training that applies useful feedback.

          I don’t think we are all really disagreeing about much…


          Ps. As you can tell I’m working my way through comments to produce a new doc…


          • I agree it comes down to a mix – or even a synergy. In some ways, training *is* the implementation of technology, if you see technology as process and not just kit. Digital software doesn’t really count as kit anyway – it is procedure encapsulated.

            I think, though, that there is a special situation in teaching, in which teachers tend to see their job in terms of personality and ideology (aka values) rather than process and where educational research is so flaky. Unlike other professions, there *isn’t* a recognised, evidence-based body of theory on which professional practice can be reliably based. That is why I think government needs to have a particular focus on the technology of teaching, not just at the theoretical/research end (as it is doing to some extent through the EEF) but also at the practical implementation end.

            I will reply separately to your comment about the practicality of the website.


  38. “I agree that treating ITT as an isolated issue is a mistake. The elephant in the room is the different views of the purpose of education and how success is measured.

    If accountability is mainly about optimising schools’ performance indicators, the approach to training teachers needs to honestly reflect that otherwise you are never going to have the training making the envisaged impact because the performance indicators will lag behind expectations.

    If on reflection we say performance indicators should not figure highly in the way we train teachers why are we letting them determine the curriculum? This is why it is essential that performance indicators represent a weighted balance of what is considered of value or like in an exam where questions are only asked on one section of a syllabus, the rest will not get taught.

    If the performance indicators are a problem, change them, scrap them or whatever, if not focus the teacher training on what is declared to be the highest priority in schools. Eradicating the things that get heads sacked, the things that cause schools to get put into special measures and the things that cause schools to become unpopular and lose their students. at present that will largely mean boosting exam results byt whatever means available.


    • Ian,

      Perhaps you could say that there are three things to sort out before you address training:

      * the ends of education;
      * the way you assess the achievement of those ends;
      * the way to teach, the success of which is inevitably going to determined by your manner of assessment.

      I am not sure that defining the purposes of education is a real problem. There may be many different nuances in the way that different parents and learners prioritise various aims – but ultimately, I do not think there is such a great difference of view. Does anyone really disagree that children need to be taught to read and write, to understand basic morality, and to think for themselves, express themselves and be creative? And is there such a problem in agreeing that parents may choose slightly different ends within a range of acceptable values defined by the state? We can haggle of the precise boundaries of such an envelope, as the Trojan Horse case shows – but even in this case, it seems to me that almost everyone accepts the basic principle.

      As for how we assess, there may well be a problem with the reliability of exams at an individual level and with the extent to which they measure all the aspects of education that we consider to be valuable. But to the extent to which exams measure core academic achievement, and when used in aggregate as a measure of the effectiveness of a particular teaching approach, I do not see that there is much of a problem. We can use them to measure the effectiveness e.g. of teaching basic arithmetic, which no-one will deny is important, without having to claim that we should not teach morality just because there isn’t an exam in it. That means that we can assess the effectiveness of different pedagogies in maths, even if it might be more difficult to compare the relative merits of different approaches to teaching morality. But surely, that is enough to be getting on with?

      So while I agree that there is a problem with many areas of our assessment system, and while I see a more rigorous application of statistical expertise and data analytics as being the way to address these problems, I do not think that the problem is so acute that we need to hold up making improvements to training while we wait for this to happen.

      To me, it is the confusion over what constitutes good pedagogy that is the deal breaker, and the difficulty in addressing this issue through current methods of research, when practice is so bound up with the infinitely variable personality of the teacher.

      Neil Davis makes what Dominic rates as a good point about government’s spend on research. And while I am kind-of making the same point, I am sceptical about the capacity of traditional research to gather data in sufficient quantities to make sense of the complex world of education. This is a point that Dylan Wiliam comes back to repeatedly – e.g. in a recent tweet at

      The Achilles Heel of any data-driven system is data capture, which generally has to be automated if you are to get serious amounts of reliable data. The key to Google’s “big data” is to capture the data we enter ourselves and to match it with data automatically captured by web crawler robots. The key to supermarket logistics systems is the barcode reader. For education, the equivalent is automated data capture through the use of ed-tech.


      • The devil is in the detail. Primary is different from secondary, subjects in secondary are different, SEND is different from mainstream. Lump everything together as ITT for eg reading and writing? Don’t think that will make much difference and could well make things worse. You miss the point about assessment. It’s only tangentially related to methods in that if we put the highest value on what is easiest to measure and put often dubious numbers to it and make people highly accountable to it, it should have a direct implication for how teachers are trained. Nuances are important because there are a hell of a lot of them and the balance between them is not easy to agree with subjects fighting political corners as well as national politicians looking for votes. Most of it is politics not rationality. Gathering data is not going to help if there is little agreement on the relative importance of stuff like eg soft skills that the CBI says are vital for employment yet government weights as zero in any performance indicators. I don’t see any amount of technology changing that.


        • I never said you should lump everything together – of course effective pedagogy will be nuanced and diverse depending on context. But if you have reasonable methods of measurement for a within a particular domain, then you can test what pedagogies and combinations of pedagogies that are effective in different contexts.

          I don’t understand what point you think I have missed. I understand precisely what you call the tangential relationship between pedagogy and assessment – but I explain above why gaps in our assessment coverage (which measures only “what is easiest to measure”) does not distort our development of pedagogy, given an awareness of what we are and are not measuring.

          I agree that the ends of education are political, not technical. That is as it should be and teachers should not complain about it because, as suppliers of a service, it is not for teachers to say what the requirements are that their service is meeting. But I do not agree that there is fierce disagreement about what those ends should be. Just because we do not currently measure soft skills does not mean that anyone does not value them. You would not find any politician in any major party who would say or even believe such a thing. The reason that they are not included in school league tables is that we do not yet have a reliable measure of these things. But by aggregating and correlating data based on teacher judgments, in ways which assess the reliability of the teacher judgments themselves, this is a problem that technology will in my view solve in the future.

          I agree that, at the moment, there are quite a lot of dodgy numbers in education, issued on the basis of examinations that have an almost ritualistic authority, or which are given selective prominence by authorities. Everyone knows that supposedly equivalent A levels apply completely different standards of difficulty. You think that numbers are themselves the problem: again, I think that data-driven technology will solve this problem. The true significance of any metric in education lies in the extent to which it can demonstrate predictive reliability. Data analytics systems will get better at testing the predictive reliability of measurements of learning, understanding the correlations between them, and how they can reliably be aggregated into broader measures of capability.

          In the meantime, the measurements of capability that we have in those core academic subjects where we have them, although not perfect, are surely good enough, especially at aggregate level, to test the effectiveness in those domains of different pedagogical approaches. Are you saying that if I were to test a new pedagogy using an RCT which showed that the pedagogy resulted in a significant increase in performance in a particular GCSE Maths exam, this result would be meaningless?


  39. Hello Dominic
    Perhaps remove the requirement that Prospective Teachers must be placed at a good or outstanding school, and maybe have a requirement that they must have experience at a good / oustanding school AND a poor / satisfactory one?

    This could increase the number of placements available and help their learning process, as it would expose recruits to different philosophies and realities of teaching, allowing them to better develop their own style and philosophy.

    Slightly counter to what I just said, but maybe make it a requirement that outstanding schools become involved in the programe to increase take up and wrench teacher training away from the heavily politisised teacher training colleges. Maybe

    Not sure if this is helpful but Im a big fan of yours.


  40. De-professionalising teachers into unthinking service providers? It might be appealing as an interim step in replacing them with machines I suppose. Someone rather wiser than me once said “Everything that can be counted does not necessarily count; everything that counts cannot necessarily be counted.” Scientists favour empirical data but they also appreciate that its use in a component in a complex interdependent system needs to be very cautiously considered. The road to hell is paved with good intentions. In education that is something there is plenty of evidence to support. Numbers are never a problem, its assigning values to them that go beyond their scope that is.


    • I don’t think we should abuse Dominic’s hosptiality by continuing what I am not sure is a very enlightening argument too much longer, but I do think it is worth countering the misconception (because it is stated so often) that being a professional is incompatible with being a service provider (tell that to doctors and lawyers); or that providing professional technology somehow de-professionalises the user of that technology. I have certainly never suggested that we should *replace* teachers with machines – on the contrary, I have debated against MOOCs at Online Educa Berlin ( and Learning Technologies London precisely on the grounds that that is what they propose to do and that is why they do not work.


      • I think you have taken up considerably more space than me 🙂 Doctors are not simple service providers. They have considerable scope to determine the methods they use and in general the SoS for health does not dictate to the medical profession how to carry out an operation. In fact doctors are free to use quack medicine like homeopathy, acupuncture and chiropractic for which the scientific evidence is that they are at best ineffective and at worst dangerous. Thankfully few do. The content of medical training is largely decided by the professionals in large teaching hospitals, NOT the government as far as I am aware and certainly not a politician based on their experience of their own stay in a hospital. In terms of ITT compared to medicine, the difference is that there is a wide consensus of what is needed to treat specific ailments that does not transfer easily to education which is not about curing but developing holistically. The only way to make education like medicine would be to treat lack of knowledge in any area as a disease and treat it as such. Such a shift in culture does not seem very likely any time soon. In addition, medical initial training is a lot longer. It is all about treating illness whereas a lot of ITT is a year after a subject degree, an add on. As I said before the devil is in the detail and the differences matter.


        • Quite right – I was not intending to limit your further contributions, just to signal my intention to wind down my own, something that I find hard to do in practice.

          My point in saying that education is a service is not to suggest that the SoS (or any other non-practitioner like parents or employers) would normally be expected to dictate to practitioners the *means* (or as you say, the methods) that they should adopt to achieve their objectives, but that it *is* for the non-practitioners to dictate to practitioners the *ends* of their service. If teachers are claiming the authority to determine both means *and* ends, then we are not talking about a profession but an authoritarian priesthood.

          While everyone would normally rely on the expert practitioners to determine how best to deliver the ends of education that non-expert stakeholders demanded, it is clear that (in large part because of a failure to distinguish between value-laden ends and technical means) there is a long history of the experts advocating pedagogies which the evidence shows are ineffective. Doctors are increasingly judged by success rates and where those success rates are poor, the claim that “doctor knows best” loses the authority that one would expect an expert to command. This is precisely the situation in respect of progressive education techniques.

          Rejecting the medical analogy on the basis that it pathologises students is IMO a flawed argument, the effect of which is again to reject the vital distinction between ends and means. I deal with it at between 8:08 – 9:30. In short, the “deficit” in medicine is in respect of a norm (i.e. good health) which suggests that hospital patients are in some respect sub-normal. In education, the deficit is not in respect of a norm but in respect of a learning objective which may be something quite exceptional – e.g. winning gold in the Olympic Games. But it is still a deficit and it is completely appropriate to apply the same general, methodological approach to removing that deficit as is achieved in medicine. We can apply that approach to the deficit without any further implication that the existence of a deficit suggests that the state of the student is sub-normal. All we have to accept is that the state of the student is sub-optimal. Rejecting a “deficit model of education” is to reject the existence of learning objectives, which in turn assumes that the success of educational interventions cannot be measured. If teachers cannot measure the success of what they do, they cannot justify their expertise, and in those circumstances have no right to regard themselves as a profession. Which brings us back to the reason why this somewhat tangential discussion is fundamental to the matter of ITT.


  41. Hello Dominic. Some reflections for you on two unintended consequences of government policy.

    First, I’ll declare my biases and interests. In history education I am passionate about a knowledge-driven curriculum, viewing knowledge as emancipatory, especially for the disadvantaged. In teacher education, my guiding principle, developed from time as history teacher and as SMT, from time running a history PGCE for a SCITT, and now in my present role with the history PGCE at Cambridge, is to empower subject mentors, so that they own and take responsibility for the quality of subject training, so that they are the chief mediators of wider scholarly-professional knowledge (and, where possible, generators of it too) and are able to blend it with the daily, practical, complex decisions of what history to teach, and how.

    I’m privileged to have ended up at Cambridge where there is a long tradition of extremely close working between schools and university. One of our school partners was quoted in our last Ofsted report saying, ‘To talk of school and university is to miss the point’. I feel the same whenever I read others’ comments which, puzzlingly, appear to separate ‘the training course’ / ‘training provider’ and the ‘school placement’. We also have a culture of valuing subject-specificity. Within shared structures and values, a measure of autonomy for subject course teams (mentors and lecturers) allows them to develop in ways responsive to subject issues and subject mentor expertise.

    The two unintended consequences have arisen as unfortunate by-products of how school-based, school-led training was introduced. They have no necessary relationship with school-led models, far from it. These damaging consequences could have been avoided if the DfE had striven to find out much more about ways in which certain already school-managed, school-involved and/or school-led ITT partnerships have long analysed and already tackled some of the problems that SD and related initiatives are partly designed to address.

    The first unintended consequence is a diminution in emphasis on the subject-specificity of training. By this I mean circumstances that actually make it much harder to create, in a new team of school-based subject mentors (a) a central emphasis on subject knowledge and disciplinary rigour in training; (b) an adequate knowledge (for training novice teachers) of how the wider body of subject teachers (nationally and internationally) has, collectively, tackled curricular issues in their subject, learned from past mistakes and built their own subject-teacher-led discourse (in subject literature, conferences, research etc); (c) a properly intricate relationship between this subject/ disciplinary/ curricular knowledge and the development of trainees’ practical curriculum and lesson planning, teaching, assessment, evaluation.

    The second unintended consequence is the danger of re-introducing and even entrenching a theory-practice divide, matching an institutional divide (uni-school), that many of us considered a huge problem in initial teacher education 20 years ago (some wrote about it much earlier than that). SD has the potential to consolidate existing successes in overcoming such a divide. Instead, with important honourable exceptions, I see that divide being re-introduced all over the place.

    These two consequences are potentially damaging both for uni-involved and for school-initiated courses: I hear more and more of their effects, from frustrated history teachers, from all over England, who contact me in confidence desperately trying somehow to introduce or sustain high quality history-specific training, in various structures, and feeling thwarted in so doing.

    One of your respondents makes an interesting generalization, and it brings my two concerns together: “teachers are shown a lot of rather tendentious, theoretical texts, urged to become self-critical and reflective, and sent into their isolated classrooms to work things out ab initio”

    His condemnation of such practice, I share totally; his implication that it is the norm or that certain ITT communities haven’t long condemned and replaced it, I dispute. Policy-makers also appear to have made the mistake that this is the norm, instead of doing a thorough analysis of where it is NOT the case, of where practices have been developed explicitly to avoid that danger of isolation, disconnection and wheel-re-inventing, and, crucially, where teams of subject teachers have worked to build a different kind of initial teacher education, one that prioritises subject/disciplinary rigour and robust curricular thinking and a responsible, cohesive community of subject teachers owning and developing the mediation of such knowledge to new teachers, chiefly in the school context, with uni practice fully woven into it. Many of us, in school and university, long ago realized that there is little point in a trainee reading any text if the mentor you are working with in school can’t help you think about it or use it. Mentors must model a passion for useful professional and scholarly reading and know how to show the trainee how to make use of it. A strong subject-driven partnership is well-placed to make that happen. Debra Kidd, above, is right: experienced teachers, those acting as mentors, need to read. Only thus can they make sure that their subject course (and it should be ‘theirs’, whatever the provider structure) is tapping trainees into the collective, formal and emergent knowledge of the school-based subject community, especially writing by subject teachers, as well as the wider disciplinary community to whom the school subject refers.

    Subject mentors cannot even make judgements about what to read and discuss with trainees if they are not in touch with that wider field of subject-specific practice. Worse, they are in grave danger of re-inventing wheels both round and square, and so, as a consequence, are their trainees.

    Whatever else ‘schools-led’ must mean, it must include some notion of ‘schools-led at the subject level’. That involves building capacity in a subject mentor team. It means having leadership/coordination that regularly reviews the specifically subject quality of the training and outcomes, developing its own robust criteria. It is not enough to rely on a set of ‘Standards’ which are set at a low base line and which separate subject knowledge from everything else, encouraging an atomized assessment and ITT curriculum. As goals for trainees, it certainly isn’t appropriate to rely on flawed measures of pupil progress such as Level Descriptions and current GCSE markschemes (in history, in particular, it would be madness to use either as measures of rigour, and diverse history teachers have long said this). A group of subject mentors needs to be constantly generating, reviewing and scrutinizing its own measures of quality in teaching that particular subject and in its definitions of rigour and excellence for pupils in that subject. They need to be doing that in conversation with wider research and practice in the subject.

    What follows is necessarily anecdotal but I want to share with you the force of the frustration that mounts up in my in-box and hits me with a queue of moans every time I’m in a history teacher gathering. My comments relate to secondary history – my experience doesn’t extend to other subjects. Also, my examples relate to situations outside of ours, here, in the Cambridge Partnership. Here, happily, we (and ‘we’ = ‘the partnership’, not university acting in isolation) are free to provide strongly history-specific training delivered chiefly in schools and we are free to continue working to build increasingly high levels of mentor responsibility. I should add, too, that the points below are, in many circumstances, sensitive. It is hard for subject voice to get out. It isn’t easy for many history teachers to express concerns which could be construed as criticism of their school, chain or alliance. But the helpful informality of your request makes this a useful setting in which to share this kind of thing. I hope it can at least trigger good questions by future policy-makers.

    The problems history teachers report to me are as follows. All relate to a trend towards genericism, away from subject-specificity, in training.

    i) nowhere near adequate time for trainees to continue to build scholarly subject knowledge, to understand major debates, trends and solutions in history education, to read the extensive literature by history teachers who have tackled longstanding problems, and to receive classroom-relevant, practical training that directly integrates all of these.

    ii) a dearth or absence, of subject-specific training for mentors;

    iii) pressure from senior managers on trainees (or on mentors to require trainees) to implement whole-school, generic pedagogies and/or to rely on assessment models that history mentors deem antithetical to rigorous history or which make it impossible to privilege pupils’ knowledge growth or disciplinary rigour in teaching;

    iv) practices in lesson feedback and report writing that are barely subject-specific at all, not showing trainees how to analyse their effectiveness in subject-specific terms, how to shape properties of historical knowledge, how to configure and re-configure relationships of old historical knowledge and new, and no opportunity to do anything about it because mechanisms/time/incentives for coordinating a team of subject mentors, across schools, are inadequate. (Such a comment comes either from history teachers who have been given responsibility for coordinating history mentors in other schools or from history teachers having systems or structures imposed on them by non-subject specialists).

    v) a lack of quality control over who mentors because there is no subject community base of expertise from which the pool of subject mentors is renewed, no reference point of subject mentoring quality by which new mentors might judge if they are ready to mentor or can work out exactly what they need to do to prepare; no clear statement of what is adequate knowledge of wider history education practices for a mentor to be able to teach the course (ie to mentor – a mentor is one type of trainer; a mentor teaches the course).

    vi) using an outside ‘provider’ to ‘deliver’ a self-contained element of ‘the course’. Without shared processes which ensure that mentors see their own practice as part of that ‘course’ or which give them academic capacity and adequate time to share in the review of the content of that course, this can only perpetuate divide across partners.

    This last is an irony, given current policy intentions. It links to my second unintended consequence. It is not unheard of, these days, for a school to ring a university and say, in effect, Could you deliver the ‘theory’ bit for us? as if such a thing could be construed separately. That this mindset even exists is a sign that we have gone backwards 20+ years.

    There are many ways of skinning this cat, and structures that trammel schools/universities into particular ways of developing strong SD models will not be helpful, but the one thing the DfE *can* do is to re-think accountability structures, and through these to factor subject expertise into the mix. Whether through inspection or other evaluative means, through preferment/promotion practices or through inter-partnership competition, we need a higher premium on (a) definitions of high quality subject-specific training construed in disciplinary, scholarly terms (not numerical outputs and not generic measures, untranslatable across subjects); (b) selection and nurture of subject-specialist mentors; and (c) communities of subject-specific mentors who consume and produce knowledge about subject and about subject mentoring, who are given time to do so, whose distinctive specialist knowledge is highly esteemed and who share in responsibility for all parts of a fully blended course.

    In emerging ITT structures, subject leaders need a clearer voice, and a status and authority commensurate with their scholarly-professional knowledge. There is no point in having an intransitive pedagogy; a pedagogy with no object. The object is the curriculum. Pupils must be taught something. That something isn’t just a given that can be neatly packaged with genericist managerial categories. It’s the heart of the intellectual challenge and the central purpose of what teachers do.


    • Hello Christine,

      On the assumption that discussion is good, perhaps I can make a quick response to your response to my point about “tendentious theoretical texts”?

      I accept that this picture of teacher training represents a generalization. I don’t know if this was intended as a criticism – if so, it is worth pointing out that government policy must necessarily be based on a generalised perception of the status quo.

      I did my PGCE in 1990 – so my observation may indeed be out of date. But I think there are still reasons to think that the situation has not changed dramatically since then:

      * classrooms are intrinsically isolated working environments and the quality of much teacher-originated OER is clearly poor;
      * I have not observed a significant rise in the development of professionally-produced learning resources (such as in activity-led coursebooks) of a sort that would replace the expectation that teachers will have to do the heavy lifting in developing their own teaching resources.
      * the same assumptions that I remember from my 1990 reading lists still dominate the professional and political discourse e.g. around “independent learning”, “21st century skills”, creativity (often promulgated in opposition to knowledge by gurus such as Sir Ken Robinson, who enjoy a huge following among teachers);
      * there is still a general assumption in the education community (see my discussion above with Ian Lynch) that it is not possible to measure teacher performance – a point of view which rejects the separation between pedagogy (i.e. educational method) from curriculum (i.e. educational objective);
      * I still see plenty of shockingly poor and politicized work that comes out of many university education departments – though not yours, I am sure!

      I agree with you about the importance of subject-specific pedagogy, by the way, and would support the need for university-led partnerships in delivering ITT, where this training is based on well-researched pedagogy. My point is not that such training should be done away with, but that it should be supplemented by the production of good course materials, which can help formalise and support the sort of good pedagogy that such training would also be promoting to its trainees.



      • I never said it was not possible to measure teacher performance so please don’t imply I did. But if you are going to measure it, state the uncertainty in the measurement or the measurement is meaningless. Assessing performance is a different issue. The quality of much commercial educational resource is poor – a billion spent on curriculum on-line and what is there to show for it? Compare that to Wikipedia which costs a few million dollars a year to maintain. The main difference between poor OERs and poor expensive licensed stuff is tax payers paying through the nose for stuff that could easily be provided much more efficiently using different and more up to date business models aided by generic technologies.


        • I agree that the quality of all educational resources is currently poor – but would argue that the reason for this in respect of OER is a lack of resource and in respect of commercial provision is an uncompetitive market. Which is why my policy recommendations focus on the creation of a competitive market (which would be open to OER as well as commercial providers).

          I’m pleased that you agree that teacher performance is measurable. All I say is that such a position is not compatible with stating either (a) that teaching is an inevitably value-laden occupation, or (b) that disputes the existence of ultimate learning objectives which are “prior”, in the sense that they are not established by the teacher in the course of their work.


    • Dear Christine, thanks very much for such a thoughtful comment. There is no doubt that SD was introduced badly – like, unfortunately, pretty much everything. It is striking that you do not mention any sort of body such as the Institute of Physics who, in history, could do some of the work you describe. Why is this? It CANNOT come from the DfE – so WHO should do it?? Why do you think other ITT providers do not simply clone your approach at Cambridge, rather than reinventing square wheels…? Best wishes, Dominic


  42. There is a danger that this discussion goes off tangentially which whilst interesting may not be beneficial to the original question. However related to this is that there is now appearing to be a body of information within this discussion with some messages appearing.

    I don’t wish to steal Dominic’s thunder as he has suggested pulling something together but perhaps the most surprising overall message is the lack of informed understanding of ITE. I expect this is the same for any discussion but given we are discussing what is incredibly important and a particularly vulnerable aspect of education it would help if contributions were informed.

    So just two points worth noting:

    Firstly research is both underrepresented in this discussion and generally misunderstood particularly given education research is in a largely healthy state. There are 10 British University education departments in the top 50 QS world rankings – which incidentally is better than how medicine performs! As always league tables are fraught with idiosyncrasies however judging by some commentaries education research was invented by excited tweeters and bloggers. I am full of encouragement for teacher researchers but we also have some exceptional world-class education research involved with teacher education, which we should be using more effectively rather than attempting to marginalize. The best ITE draws upon educational research (which includes research from other fields) in a seamless way and engages trainees (and teachers) in research so that they become discerning and develop a level of criticality. For many in ITE we have worked hard to overcome the ivory towers image and have endeavoured to reduce any perceived theory and practice divide. A common feature of School Direct that is therefore unfortunately appearing is a redistribution of roles where the University simply takes on the academic work in a reduced role. This is School Direct on the ‘cheap’ and everyone should be cautious of this emerging practice.

    It is also however sad to note that it is probably only about four years ago when we were talking about an all Master’s profession and the introduction of MATL. Even at that time there were many saying the MATL was insufficiently rigorous however it seems a million miles from where we are now. Therefore something such as MATL would seem a perfect fit in the current climate if someone were prepared to pay for it! There is sufficient evidence that teachers need challenging and developing throughout their careers and whilst social media is finding new ways of joining teachers up there is still a need for teachers to engage with high quality research and practice which is generally but not exclusively located in University schools of education.

    Secondly the other factor that has always been a misconception is that School Direct is about trainees spending a significant amount of their time in schools. The reality is they have always done so with at least 2/3rds of a PGCE spent in schools. This is a significantly higher proportion than in other countries and instead of starting from the position of constantly bashing ITE – there should be the question of how on earth do providers get trainees through such an intensive activity which includes qualified teacher status and a Post Graduate qualification in such a short period of time. When I have explained to academics in other countries what happens on a PGCE they often laugh and say it is impossible!

    Equally even if the government or a minority of headteachers don’t value ITE the trainees certainly do. The most common message I hear from trainees is that the PGCE is the most demanding activity they have ever undertaken but also most rewarding. After a period of time in their NQT year they are also then asked to reflect on their training and overwhelmingly this is consistently positive. In those cases where it isn’t then something usually happens either internally or externally! Whilst the NQT survey has its limitations it does paint a positive picture and when combined with Ofsted reports and the education select committee reports they all indicated that ITE was a success story.

    I have indicated previously that there are some positive stories emerging with School Direct (but also some clear problems) and there really should be a taking into account a picture about what is emerging and build the best onto the best that is already in place (whilst it is still in place). School Direct shouldn’t be offered as a binary to the traditional PGCE – it needs to be offered as an alternative for the right trainees who, now that they are paying £9k for the experience, have a right to be able to choose from different models of delivery in an informed way.

    Ultimately there is a bulging school population and a diminishing pool of talent to draw new teachers from and as such we have all the signs of a recruitment crisis emerging. The last few years have been incredibly demanding for those in ITE with often little educational, but plenty of ideological, justification for such turbulence. There really does need to be a taking stock of a national picture of teacher education as I sense we are in a critical period and the wrong next step could genuinely see a haemorrhaging of expertise from ITE – which may have always been the intention?


    • David,

      Contesting the ongoing validity of subject knowledge is not an activity that ITT providers have distinguished themselves. ICT has been rethought emerging with a new body of knowledge grounded in computer science after external intervention from interested parties. Similarly, D&T has been the subject of ofsted subject repoers highly critical of the state of the subject. The raspberry pi, arduino etc are being adopted by primary and secondary schools. Where is the inter subject collaboration between D&T and computing? Teacher training should lead to enhanced opportunities for school children. If ITT providers are not leading development in subject disciplines, who is?

      ITT educators do not submit much research via the REF, the academic measure of quality. Why is this?

      Where is the research into the use of technology enhanced learning, be it via VLE’s – post the technology enhanced learning strategy or more contemporary use of tablets, bring your own devices etc. If you could cite one study that has been submitted for the REF I’d be very interested to read it and how it applies to children. The literary trust are calling for schools to participate in research relating to the impact of ebooks and the development of pupils reading abilities. They cite a near absence of research in this area as there motivation.

      I don’t think ITT providers are best positioned to undertake the type of academic research that post docs are capable of generating. Action research is useful and assists in developing teachers ability to engage with published research.

      ITT are in the business of training new entrants to teaching. If those trained do not start an NQT year or leave prematurely we should be concerned why. Professor Alan Smithers, Buckingham University has produced a yearly evaluation of ITT providers performance as measured against key criteria in this area. The ‘Good Teacher Training Guide’ should be uncomfortable reading for certain PGCE subjects and by association ITT providers and subject associations.


      • 1) I disagree and the point has been made to NCTL is that the type of subject knowledge development you refer to is potentially being squeezed due to School Direct. Please do contact me if you wish to know more.
        2) I don’t know of any systematic study which has collated how many involved in ITE have contributed to the REF but it will certainly vary by institutional priorities. One unfortunate model of ITE that has emerged in some Universities is that there are those that are involved with ITE and those involved with research – with little crossover. I believe that there should be a symbiotic relationship between research and ITE.
        3) I don’t believe a list of publications submitted to the REF has been released so I couldn’t name one. However anyone can commission research.
        4) Who do you think taught the Post Docs? ITE tutors can also be doctorate supervisors.
        5) The Good Teacher Training Guide, despite its limitations, will provide uncomfortable reading for about 50% of universities (the bottom half) but it will be interesting to see who or how data is being collected on School Direct success in terms of employment, retention, quality, etc. I am not aware of how such data is or can be captured at the level of detail needed so the success of School Direct by traditional means won’t be fully known. From my own analysis there is some positive data emerging about School Direct but also some data which is a concern. Interestingly the culpability and liability for School Direct still rests with the provider which has to be addressed.


    • Surveys of undergraduate students on physics courses (‘how do you rate this course?’) are regarded as essentially worthless. Why would education be different? Students tend to evaluate uni teachers on the basis of whether they like them which is worthless as a method for judging the course’s effectiveness.

      I’m afraid we strongly disagree on the state of education research. You regard it as a ‘success story’. I regard it as very often cargo cult science.

      When I arrived at DfE in 2011 I asked someone v senior in the ed research community about Feinstein’s graph (showing the rich thickos overtaking the clever poor kids). ‘Gold standard, no doubts’ was the reply. A week later, Vignoles’ paper came out saying it was probably regression to the mean!



  43. David,

    You must allow those of us who are already sceptical of the health of the UK’s education departments to remain so, when your main evidence for that health is a ranking system substantially based on the standing of UK education departments among other academic educationalists.

    Would you say that the sector has made significant progress since the days of the 1998 Tooley Report or would you say that the Tooley Report got it wrong in the first place? Would you dismiss Daisy Christodoulou, Tom Bennett, Robert Peal and Andrew Old as excitable Tweeters and Bloggers or would you agree that there is some merit in their criticisms of the educational orthodoxy which has been established by education departments over the last 50 years? Do you think that the sort of stuff highlighted by Andrew Old on the BERA website ( is the output of good educational research, or would you say that this sort of stuff is not representative of much current academic writing?

    I notice that in your most recent blog about your “Evidence Based Teaching” presentation (, you say that “Evidence is not a substitute for our values”. That rings alarm bells for me, confirming precisely my criticism of much of educational academia, that it sees its job as defining the ends of education (often in terms e.g. of “powerful hegemonic discourses of race, class, gender, sexuality, and religion”), rather than defining the means of achieving the ends that the rest of society wants from an education system for which it pays and at which its children are required by law to attend.



  44. Crispin – I am happy to engage in a debate about education research elsewhere however as indicated in my last post there is danger of this topic going off tangentially or being hijacked. Perhaps Dominic should put educational research as the next topic.


  45. State how you will evaluate the uncertainty in your measurement without general agreement over your assertion that teaching is not value laden. Spiritual, Moral, Social and Cultural education are value laden and at the heart of UK education culture. . As for OERs, it just requires the right business model and these take time to gain traction from absolute scratch. Watch this space.


    • Thank you all for your comments.

      1. Kevin Stannard’s review of Daisy’s book. I agree that we do not want to swing wildly from one extreme of the pendulum to another and that memorisation of facts is of little value without that knowledge being assimilated and applied through activity. I refer to these four writers as polemicists who attack the child-centred model of education, for which our academic education departments are chiefly responsible. In this respect, Kevin Stannard’s criticism of Daisy’s book is that the fortress had already been taken – a point that I think is belied by the furore that Daisy’s book has caused.

      When it is generally recognised that the fortress has fallen and it comes to saying what would work better than child centred / independent learning / constructivism, then we can all afford to slip out of polemical mode and look for the nuanced, middle way. While I quote Tom Bennett’s Teacher Proof with approval as a polemic, I have also criticised some of his conclusions (

      2. David on avoiding red herrings. Although I am happy to leave the debate about research till later, the point that I made in my first comment, some way back, is that there is not much point in discussing how you want to train people until you have a reasonable idea about what you want to train them in. So although I accept that the question of research may appear to be somewhat off topic, I think it is actually prior to the question of ITT/ITE.

      3. Ian. At the risk of repeating myself, my position is that the ends of any process are value laden – but establishing the ends of education does not fall within the remit of teachers. Establishing the means to achieve any end is not value-laden (in the sense of being subjective). The relative effectiveness of any means in achieving any end can at least in theory be objectively demonstrated.

      That is not to say that I do not accept that there will be a degree of uncertainty in any measurement of learning – this is something that I am writing about at the moment. But this degree of uncertainty can itself by quantified by the application of statistical analysis, which will in most cases be reducible to a measurement of predictive reliability. There is a big difference between uncertainty and subjectivity.

      4. As for OERs, while I myself am sceptical that OER will do the heavy lifting in the development of ed-tech software, I am all in favour of giving open source a completely level playing field. All I am against is OER being used as a means for supply-side state funding to grab the moral high-ground, while generally producing rubbish outcomes and buggering up the market for everybody else at the same time. I do not see any incompatibility between true, self-sustaining OER and commercial provision. Indeed, I believe that OER will find useful synergies on the margins of a healthy commercial market. But as you say, so long as everyone agrees to keep the playing field level, then we can very happily agree to disagree, while everyone watches this space.


  46. Crispin, teachers will probably stop having views on the purpose of education when politicians stop interfering in operational matters. The main subsidies for education resources over the last 30 years have been for a business model predicated on selling software licenses. It has been a very expensive failure. A billion on curriculum on-line has had very little impact at all compared to using stuff that is largely free from licensing costs. The “specialist ed techs” are the things on the margin compared to the take up of eg Google drive, wikipedia, internet search, dropbox, youtube and other more generic free web based services used creatively to support educational need. The trend is to business models that lower barriers to entry, not very expensive specialist applications that raise them. Smell the coffee. If we learn anything from the last 30 years, its that the English national market is not big enough to sustain highly specialised expensive to develop educational technological solutions. Pretty well all the widespread technologies are derived from generic consumer led innovations that scale globally. If we want tech innovation in ITT it would be far better to look at things that already exist and can be used innovatively by all teachers at very low or zero cost to streamline routine tasks and contribute to their own digital literacy. That would be genuine education innovation as opposed to technical innovation for its own sake.


    • I agree with you entirely on what a properly functioning system should look like – and that is one in which politicians keep out of pedagogical issues, which should be determined by educationalists who back their work on solid evidence and not ideological posturing. Getting to those sunny uplands may involve the odd compromise, but I am more optimistic than you seem to be that it is achievable.

      I completely agree that government should not fund commercial provision through e.g. ring-fenced grants or vouchers, any more than it should fund OER provision. It is essential that ed-tech providers, whatever their business model, should earn their keep by delivering genuine benefits to the front line.

      As for the size of the market, I see two levels in the ed-tech market. (1) is for software that supports generic pedagogical interactions: (1a) at the platform level, assignment management systems, data analytics systems, e-portfolio systems etc; and (1b) at the instructional level, simulations engines, timeline editors (for my subject, History), structured debate managers and educational wikis etc. etc. (2) is for instantiations of those generic interactions which are specific to a particular curriculum and the aggregations of activities (e.g. courses) that comprise sequences of those activities. While (2) is a national market, (1) is an international market and (2) sits on top of (1). In other words, the more important part of the ed-tech market will be international.

      I think it is a mistake to describe the future market for ed-tech by observing the IMO completely ineffective forms of ed-tech that were deployed in the 2000s, which used generic technology (as you describe), and were predicated on flawed academic theories of child-centred independent learning. My view is that the ed-tech of the future will be much more similar to application-specific business software, which models the particular transactions and processes of that business. Business does not run open-source software at this application level – on the contrary, their software systems are complex and expensive. Though any sector-specific technology targeting a vertical market always sits on top of the generic, horizontal technologies that you mention.

      Putting aside the national vs international question, the size of the market is related to the effectiveness of the product. The market for ineffective software is always small. But education, heavily dependent as it is on staffing, is an expensive business, the demand for good education in the knowledge economy is huge, and so the potential size of an effective ed-tech market is also huge.

      I do not think that it is a matter for government to second guess exactly what products education will require or whether they will be provided by commercial or open source business models. I propose that government’s role in this, seeing the transformative potential of technology in education, not least in encapsulating, disseminating and evidencing effective pedagogy, is to establish the market infrastructures that will enable innovation through open competition.


  47. Ian / Crispin

    Perhaps you could each summarise:

    a) What do you think you both agree on?
    b) What are the vital areas of disagreement?
    c) What is the crucial evidence/discoveries that need to be made to resolve the vital disagreements?
    d) What shd the DfE DO given a) and b): ie. what are the uncontroversial things that will bring progress while allowing flexibility to make further decisions in the light of discoveries?

    Thanks for your efforts in furthering the discussion and feel free to continue, but I want to do a first draft of a summary…



    • We seem to agree that government should not subsidise education technology development. We seem to disagree that a big and significant shift is taking place to OERs and service based business models based on generic consumer led technologies as opposed to licensing proprietary specifically educational software. I’d cite Facebook, YouTube, Wikipedia, Dropbox, Google Drive, Android as evidence of a global shift to software as a service. This is not yet fully reflected in schools (or some large public and private sector organisations) because they are locked into a legacy paradigm. I’d say that *any* business model based on licensing content or software is vulnerable to disruption. It is most likely that the legacy licensing model will persist in specialist niches not generic widespread take up. I’m not sure whether Crispin agrees with that or not. Since I’m betting my entire business strategy on it you can assume I have done the proper research and it is working. It’s not easy otherwise many others would have done it already but a lot of that is down to legacy and professional knowledge.

      Uncontroversial things:

      1. DfE get their own house in order and train their staff in the importance of open systems so they stop using unnecessary proprietary protocols and formats that slow down the shift to open standards. (Not that expensive and the DfE above all government departments should value lifelong learning 😉 )

      2. Require all current holders of contracts for DfE web site based services to ensure that the sites work properly to W3C standards and have been verified to work properly with the 5 most popular browsers. Make the deadline say 6 months or their contract will not be renewed. (after all competent design in the first place would mean there should be little or no work to do)

      3. In terms of ITT require new teachers to evidence their own learning using an IT system (we have a free one if they want to use it so need cost the tax payer nothing for the technology, only for training which is an expense already there. Ok you would have to train the teacher trainers but that is a much smaller number to deal with). This would give them the skills and knowledge to transfer the same workflow to their classrooms. It will save them time and they can transfer those skills to the pupils. So there is a strategy then to get everyone to routinely use cloud based technologies to support their work – much in the way the most productive businesses do now.

      4. Provide the incentives for on-going updating by providing certification of teachers every 3 years based on their evidence portfolio that is part of their day to day work (started in their ITT) After all, all teachers plan lessons and document stuff, just shift it to the cloud so they can be given credit for doing it. Make it voluntary, not an imposition, but obviously a teacher with evidence of keeping up to date is going to have a better chance of promotion than one that does not so I think you’d get a high take up at least from ITT.


        • Hello Dominic,

          My account of the conversation with Ian, written before I saw Ian’s account.

          My position is that much university-based education is value-laden and not based on robust evidence. The attempt to circumvent this by placing ITT in schools is only partly successful because, without being underpinned by educational theory, classroom practice is inconsistent and the value of school-based training is likely to be overly dependent on the personal craft of individual mentors.

          My remedy (which I realize may appear a little off-topic) is to encourage the development of a market for teaching resources, which will encapsulate theory in a form that is directly usable and more testable than is personal practice. This will provide experimental justification for the techniques that are taught during ITT, as well as the “handrail” that will help support students’ own practice during training. To echo Dominic’s cut-and-paste comment, these are similar to a tips-and-tricks approach, though in context-specific, concrete form. Such resources are likely to be predominantly digital, which has led our discussion into one about ed-tech.

          My position is predicated on the view that one can separate the subjective, value-laden aspect of teaching from the objectively measurable, technocratic aspects of teaching. I associate the subjective aspects with the ends of education which are ultimately not for teachers to determine; and the objective aspects with the means by which teachers deliver those ends, which is the proper subject of their expertise.

          As I understand it, Ian does not think it is possible to isolate teaching from these subjective values and is sceptical about my claim that data-driven software systems will help us to evaluate the effectiveness of educational processes. He is concerned that data-driven ed-tech will disempower or de-professionalise the human teacher. My answer is that in providing the equivalent of the “tools of the trade”, ed-tech will on the contrary empower teachers, while leaving them free to exercise their core role, which is to manage, inspire and provide at the top end of Bloom’s taxonomy.

          With respect to ed-tech (a field in which we are both involved), I think we agree that nothing has worked very well over the last couple of decades. Ian places his hopes in open source, substantially using generic internet technologies; I think that more proprietary developments will be required, particularly focusing on software built specifically for the education market. We both think that direct government intervention on the supply side has been damaging. Though as I think there is more heavy lifting to be done in terms of software development, I see more of a role for government in building a competitive market, while I think Ian believes government should just leave the whole sector well alone.

          Having read Ian’s account, I have two responses.

          1. I have no strong views on software-as-a-service licensing models, but on the whole I am inclined to agree with Ian that this is the direction of travel. I think in our debate, this may have got conflated with the proprietary/open source distinction.

          2. Ian and I are both interested in open standards and interoperability. But as Ian sees the main change coming through generic software, he places the main emphasis on generic W3C standards. This is an issue that the Cabinet Office has been addressing over the last 4 years, albeit slowly.

          Because I believe in the need for a top layer of education-specific software, I believe there is a need for a set of education-specific data standards, which will sit on-top of the current W3C standards. In some cases, there will be overlap. For example, IDPF has developed the open EPUB e-book standard. One of the specific requirements in the education space is for people to be able to exchange book annotations. So there is some work going on within W3C on open annotations standards, which have generic application but which are particularly important to education. In cases like these, I would agree with what I am sure would be Ian’s position, that the generic standard always trumps the sector-specific standard. Leave it to W3C, in other words.

          That leaves, in my view, a significant requirement for education-specific interoperability standards. How does a learning platform launch a digitally mediated activity in a way that is consistent, regardless of the type of activity that we are talking about (creative exercise, assessment, investigation, collaboration etc)? How does that digitally mediated activity report learning outcome data, also in a consistent way? If the education community does not sort these issues out, then no-one will.

          When participating in the Cabinet Office’s open standards consultation a couple of years ago, I found that there was a very similar situation in Health. Indeed, the whole fiasco of the patients record system could have been sorted if you had a competitive market for such systems, allowing the industry to drive the standardisation of the data that such systems would need to exchange. In other words, you achieve integration at the standards level, not by having a single IT system. Phew! The possibility of creating such data standards is related to the question about whether learning can be measured, which in turn is linked to the current interest (still somewhat premature) in the likely impact of “big data” on education and the work of the EEF.

          So when I say that the government needs to build the infrastructure for competitive markets, creating dynamic data standards processes is part of that infrastructure. The creation of such standards, inevitably tied to industry innovation, is an area that both the Cabinet Office and the DfE’s Information Standards Board have both avoided like the plague. As Chairman of the (equally ineffective) BSI committee in this area (which represents UK interests at CEN and ISO/IEC) I have have done my best to persuade the ISB to offer leadership but without success. No-one else is doing this work.

          I think Ian regards this emphasis on sector-specific data standards as a bit of a red-herring, believing that generic W3C standards, underpinning generic software systems, are sufficient.

          Thank you for listening and I am sure that Ian will correct anything I might have got wrong in describing his position.



  48. This is an interesting comment. I am posting it on behalf of someone who has been deeply involved with teacher training policy since 2010 and knows the DfE wiring but who because of their position can’t comment on the record. I have put my own comments [inside square brackets like so]. It is a verbatim cut and paste other than I have stripped out a few tiny things / pronouns that would identify who it is.

    “1) SD needs a proper evaluation urgently. [Agreed.]

    “2) It seems to work well for academy chains that have got their act together (ARK, Harris, Cabot, Outwood Grange etc…) – they have the capacity to recruit good people and hire decent people to do the training. [Agreed.]

    “3) Works much less well for individual schools or small groups that don’t have this capacity. This was inevitable the moment DfE decided not to have a central recruitment process. They find it hard/complex to recruit and end up handing back places. In addition quality of the training will only be as good as the quality of the school. So it is patchy.

    [Mostly agreed. Some good individual schools are also happy and doing well but it’s clearly true many aren’t. Sam Freedman argued for a central recruitment process in his PX report pre-2010 and in meetings with me and MG in the DfE. We explored doing this in 2011 but the dysfunction of the department made it impossible. A scheme was proposed but like all things involving the DfE running something involving a new IT system it was an obvious car crash. We therefore decided to go ahead without a central process and instead hoped/tried to encourage an organic approach entirely bottom-up. I think this was the right thing given the constraints. The other options were a) drop the whole idea, b) approve a central recruitment process that would have 1) delayed everything and then 2) been a shambles with >95% probability. It is the sort of thing that could be done now in the DfE given it is in much better shape though it would still be much harder to do than people outside realise, given inevitable nightmare arguments about a) EU procurement rules and b) Whitehall approaches to IT that separately and in combination would be a real pain. Two NEDs could help enormously though – Paul Marshall and Jim O’Neill, both of whom could help officials minimise the chances of failure.]

    “4) The system is adjusting and we’re starting to see larger groups of schools come together to manage recruitment etc… but left to happen organically this will be probably take too long to avoid a recruitment crisis over the next few years.

    [The commenter has much better finger tip feel than me for likely dynamics of ‘recruitment crisis’ but I would make one point: the DfE’s data is always late and bad and my experience confirmed me in the view that we must find a way of allowing supply/demand of teacher training recruitment to become a bottom-up process driven by schools/demographics rather than a top-down process driven by attempted ‘wise central planning in the DfE’. Nothing I saw suggested to me that the latter will work, certainly without a transformation of Whitehall that is inconceivable regardless of whether DC or EM is next PM. Also, as Tom Bennett has pointed out, the DfE’s central planning did not prevent previous ‘recruitment crises’. To push the point to an extreme: nobody would suggest a central process to plan recruitment of different types of programmers in Silicon Valley: an adaptable ecosystem is the best solution. Teacher training needs to be more like successful models outside Whitehall rather than resembling a 1950s Whitehall process. I do not mean that the current system is right or that we could not have done it better – it obviously is not right and it could be much better. But the preferred evolution should be mechanisms that allow decentralised coordination/adaptability rather than more central planning based on bad data and the inevitably limited attention span of MPs.]

    “5) So the system could be pushed by either introducing centralised recruitment or, perhaps more realistically, getting schools to cluster together in a slightly more brokered/organised way to do joint recruitment and training (unless of course they’re already in an academy chain or strong Teaching School alliance). [For the reasons above I think the second option is more likely to work and the first option is more likely a) to blow up and b) be used to seize back control for Whitehall. But I also think that IF a SoS made it a top priority AND insisted on involving people from outside Whitehall in the design and implementation AND if noise from EU procurement etc could be minimised – THEN a central recruitment process could enable decentralised adaptability.]

    “6) In terms of ITT itself what ARK, Teach First and a few others are doing is at the cutting edge. Essentially building in more practice of specific technique/routines a la Doug Lemov alongside traditional theoretical/classroom based material. Lots of the university people hate this as they see it as ‘teaching tips’ and taking away from the intellectual value of teaching but it really is a false dichotomy…you can have both.”

    [I strongly agree with this point and I think it is vital. The main hope of Sam Freedman, Gove, Zoete, me was to enable the Doug Lemov approach to be integrated in Academy chains so that direct empirical evidence of what works is built on and failure is quickly adapted to. This is consistent with Feynman’s classic essay on Cargo Cult science and education research. It is very hard not to fool yourself about what works. Much education ‘research’ has spread errors – not knowledge. Physics has long-established tools to limit human propensity to fool oneself – tools that have developed the most accurate quantitative models of the world that we have. Medicine ignored this for a long time but has done better in recent decades. Teaching must now do the same. (Obviously neither medicine nor teaching can approach the accuracy of physics but that is not the point – mechanisms for not fooling oneself can be built.) Here I am very hopeful as I think things like Goldacre’s report, ResearchED and all the other stuff that has grown recently is an unstoppable tide now, despite union opposition and wariness/recalcitrance from many in universities.]


    • However what ARK et al and Teach First do are to emulate the very best practices of University – School based teacher education. Although these top performing HEIs go further and also enourage new teachers to think critically and to steer away from cults and ten top tips approaches. Rather Uni-school partnerships work together to prepare new teachers to be able to make good deliberative judgements in specific nuanced complex classroom situations based on being well educated.

      It is a great waste of energy that we are once again rehearsing arguments that were being raised in the 80s by Shelia Lawlor et al l when I first started teaching.

      Surley we all [HAVE] the wherewithal to work together to make our education system the best in the world and to go beyond partisan polarised discussion?


      • I inserted the [HAVE] above assuming it was a typo.

        I’m not sure what you think is the ‘waste of energy’. If you think good schools have been generally happy with the quality of HEI training, I’m afraid you are mistaken. This is part of the reason for the debate over how to a) determine what is successful/failing, b) channel more money and people to the successful, c) connect training with the very new trend of trying to approach training and education research in a genuinely empirical way – in a scientific way rather than a Cargo Cult Science way (Cf. Feynman).

        This is not to say all HEIs = bad, or that School Direct is an obvious success. Some HEI training is clearly deeply valued by schools. Some aspects of SD have been bad. But these things do need debating and I do not think it is a ‘waste of energy’. You are of course not obliged to waste any of your own energy on the debate!

        I do not think the discussion above has been ‘partisan polarised’. The goal is to produce some agreed specific action DfE can take that would grab low hanging fruit and help the system evolve in positive ways.

        Best wishes


        • I did my PGCE in 1976 have things really changed? I really doubt it except at fairly superficial levels. If you want ITT to grow from grass roots you need methods that enable devolved practice from a central set of standards. The internet shows how successful that can be. In fact everything exists to do it now and with no need to go for EU procurement – a lot of the technology was developed using EU grant funding in the lifelong learning programme in any case, just extend what has been done. They like sustainability.

          I did make a start on defining a teaching apprenticeship for a post graduate qualification based on the teaching standards at but other things have taken over for the time being. This is a competence based assessment model which defines baseline teaching competence that can be linked to academic and theoretical learning. Good teaching needs both so let’s not get into a polarised either or debate. If we have a web-based assessment model that links the defined outcomes to underpinning resources and guidance, the student can provide the evidence of what they have learnt from anywhere and the assessor can verify it from anywhere and external moderation can be formative too, providing constructive feedback across students. Extend it beyond ITT to have a manageable system of CPD. It does not mean we have to abandon university based training or school based practice, it simply makes it easier to compare standards and manage the information using common open systems based infrastructure. It’s also a tried and tested method with similarly constructed school qualifications accredited by Ofqual and endorsed by the DfE in terms of league table points.

          I’m not saying you have to use this exact system, I’m saying the solutions are already available and ready to be deployed. Flexible enough for individual institutional character but with common technical dimensions to improve efficiency. If we want evidence based education, let’s start by making the evidence required to grant qualified teaching status practical and transparent.


      • Thank you for editing my typo! Done in haste.
        I am glad that you have started this debate and don’t mean to appear to be dismissive but I do think we have regressed in terms of our world standing in teacher education in the last four years. It would be good if we were to move forward rather than standing still or even regressing in some areas. I can see that thinking more divergently may be a good way forward and it has to be right that we should listen to all the views represented. Thank you for allowing HEIs the opportunity to voice our opinions too.

        So I have one final question. Is the evidence you talk about based on anecdotal impressions and feelings expressed by some people in ‘good’ schools or is the policy drive to promote SD over University -school partnerships based on sound empirical evidence? If it is the latter please excuse my ignorance and point me to the sources. The ‘evidence’ used in the 2010 White paper was not convincing and the Hargreaves ‘think pieces’ are untested.


  49. […]

    its interesting to note the differences between the suggestions from the teaching business here, and the suggestions from parents in the comments on the recent john redwood blog on this topic.

    personally I think there is a lot to be said for the parents views expressed on johns blog.


    • The […] above contains a link to Redwood’s blog.

      I have allowed this comment but deleted the link.

      The reason is that the comments on that blog are largely a) about different subjects to this debate or b) gibberish/insults, like most comments on political blogs, so I am deleting the link to avoid readers here accidentally wasting time.

      Anybody who wants to look at JR’s view or the comments can google it.

      A summary is: a) allow more selection, b) allow profits.

      Whatever the merits of either point they are not directly relevant to the narrower issue here of teacher training and what precisely the DfE should do about it.



  50. The comment below is cut and paste BY ME from Tom Bennett’s blog:

    “My six recommendations to the [Carter ITT] review are:

    1. Imbed a basic literacy about what research looks like in ITT, and the varieties of methodologies available to education researchers- including their limitations
    2. Provide better guidance about best practise in teacher trainee research, rather than just say ‘go do research’
    3. Warn teachers of use perils of blindly conducting Action Research without governance from an established research body. 20 kids in your class for two terms isn’t research. It’s a punt. Which is fine, but a punt isn’t research.
    4. Encourage teachers to become research literate simultaneous to actually practising in a classroom. Real life often sobers us up when blind theory can obfuscate and intoxicate. And theory can illuminate experience.
    5. Encourage teachers to plan their CPD on a research basis, so that even after ITT their powers of research literacy can be used to guide their futures.
    6. While acknowledging the nuance and subtlety of what research actually says ( for example, the front page of the EEF teacher toolkit makes easy reading, but the devil is in the details within), teacher trainers need to present the big picture of what the best research points towards- and most of all, what it insubstantiates. Or fails to substantiate. VAK, for example, isn’t definitely untrue, merely unevidenced. So there might be something in it. But as far as we know, there isn’t. Teach that.

    “The DfE has been surfing the wave of evidence based practice for some time now, and in my discussions with them I’ve never seen anything other than an honest, ambitious desire to find out what represents the best research in education, and disseminate it. With Gove gone, and an election around the corner, all hats are in the air again. Even the outcome of this review is uncertain. Will it launch like a rocket, or be quietly published and filed? Only time will tell if Morgan will be fey- or will we get Carter?”

    Tom Bennett


  51. There is little in this that I would disagree with and you would anticipate that much of this is bread and butter for most HEIs. the disconnect comes from School Direct potentially increasing the theory and practice divide (see my earlier note). Valuing research and recognising that teaching is also an activity grounded in academic understanding seems to be something we have moved away from. I suspect an unintended (or perhaps intended) consequence of the changes in the last 4 years is that there has possibly been a reduction in some of the activities that Tom calls for – despite the increased noise (but the noise is welcomed). Again I mention the discontinuation of MATL and removal of PPD funding which have been a backward step.
    Unfortunately much of the debate is presented as a binary – when the reality is a mixed economy can exist but if this is to be the case then there has to be a transparency to the diversification of ITE.

    Turning briefly to an earlier (insider) post – there is so much in this that I don’t know where to start. However a central message that is emerging is that there really needs to be a considered approach before the next steps in the development of School Direct take place. There should be time taken to capture the very best practice that is emerging as we can’t simply let it grow organically or let the market dictate its development – there is too much at stake for this to be allowed.


    • Tom Bennett seems to be mainly concerned to inoculate front-line teachers against ingesting snake oil. That seems to me to be a reasonable reaction to current conditions but somewhat defensive as a long-term strategy – sort of like giving travelers training in self-defence before sending them out into bandit country. The better long-term objective would be to encourage the emergence of a professional environment in which even gullible teachers are not likely to be waylaid by quack prescriptions.

      So the question is how to foster the emergence of such an evidence-based environment in which a reliable consensus emerges on good pedagogical practice. David’s second paragraph makes the assumption that the market is inferior as a way of achieving that goal than central control exercised through the authoritative identification of best practices. The truth of that assumption will depend on (a) how much you trust the agencies that are exercising central control, and (b) how the market is set up: e.g. how transparent are your information flows, how flexible are the procurement mechanisms, who are the customers and how they are incentivised. The history of educational theory suggests to me that the picking of pedagogical best practices by central authorities has been extremely unreliable. That appears to many of us precisely why we are in our current mess.



  52. Seems to me that there is a significant issue in science and maths education here. We want research to be “scientific” ie stand up to the sort of rigour that double blind surveys do in medicine. One ground breaking study on that basis in the 1980s was CASE (Cognitive Acceleration through Science Education) at Kings College carried out by science educators. It showed that children that went through the CASE programme improved all their GCSE grades, not just science. Now one would think that concrete evidence like that would take precedence but the National Curriculum scuppered it with retrenchment to a knowledge based syllabus. And more recently all the transferrable skills and knowledge stuff is being called into question by Willingham et al on the basis of the science of how memory works. So even when there is “scientific research”, we end up with conflicting results and polarised politics. The problem is that most teachers and probably many ITT lecturers don’t have the maths or science to know what is and is not scientific and how to work out how reliable their conclusions are. I did an error treatment of the data in my MSc dissertation in Education Management. My supervisor asked what all the complicated maths was about and said the final examiner wouldn’t understand it so it was not worth including. Hmm.

    On VAK I’d say context of learning is a motivator and there is strong evidence that people have different preferences for learning context otherwise every child would choose the same options when given a choice (assuming its not just teacher preference). So probably if we changed from preferred learning style to preferred learning context there might well be a case, and its probably as much to do with the Amygdala and motivation as it is the Cortex and working memory. But maybe that needs a research project 🙂


  53. Further to your comment earlier about ‘good’ schools’ views on ITE. I have just read this blog by the Head at a school which might fit your ‘good’ school category who is very positive about recently PGCE trained new teachers.
    I don’t think the situation is a drastic or clear cut as we are being led to believe.


  54. Paragraph 3? One emotionally laden anecdote in a highly non-representative context is hardly the stuff to engender confidence in rational and research led approaches to education reform. While we have this constant politically confrontational point scoring I can’t see much hope for change.


  55. People are free to leave more comments but… I’m now travelling and may not be able to click APPROVE for a day or so. I’m writing up a summary of the above debate to post next week. I know that relevant officials and others have read all this and are thinking about it. I don’t know what they will do. I’ve also learned a lot from the comments. Thanks again to all who have contributed. D


  56. A few things to note, first this discussion seems to have become dominated by a few insistent voices. This seems to be the norm in discussing education and ITT in particular. There are a few misconceptions flying around. Namely that ITT HE tutors are somehow dissociated from ‘real’ practice in schools, odd as most of them, like myself either came recently from schools as teachers or school leaders, or are still working part time in schools and/or are researching in schools while engaging with their own studies, in my case my Doctoral studies in education. Nearly all HE ITT tutors also work as link tutors in schools alongside mentors so this myth of academic remoteness is rather strange and is rather more indicative of ignorance as to the who and what of ITT academics. Besides for HE ITT tutors/academics teaching in HE is the major part of their work load. Added to this is that most HE ITT tutors specialise in teaching subjects that they themselves are specialists in, unlike many secondary teachers, particularly STEM. Unlike schools HE ITT tutors keep up to date with latest research and are indeed forced to anyway via the mandatory revalidate on of courses. It is true that the PGCE course is too short, many of us in HE ITT have argued for it to be extended,

    with specifically a more focussed mentored NQT year, as others have noted mentoring beyond ITT leaves a lot to be desired and it could be argued that it is this that causes a lack of retention later on, especially as NQTs report high satisfaction with their ITT.

    Click to access DFE-RR306.pdf

    We should also note that there are far more routes into training than being discussed. I myself came via the GTP route, but having now experienced teaching, mentoring and tutoring undergraduate full time, part time, PGCE full and part time including school direct and other modular forms in Primary and 7-14 I realise that we need routes that suit people and schools rather than ones that just suit a particular ideology, either left, right or indifferent.

    Finally, the recent meme about training teachers to be researchers is a red herring, we already do! Admittedly it could be done better by some ITT providers, but we shouldn’t confuse the development of a reflective, reflexive critical practitioner via enhanced courses formed around research with attempts by some commentators to promote certain limited forms or research practice.

    ITT has improved in this country and it is alive and aware to the need to be proactive to current research (most of which it actually does with partners in schools) but divorcing it from that research base as a practice only form of ITT would run the risk of turning a profession into a calcified form of turgid regurgitation laid open to the whims of corporate necessities rather than that of the society it should serve.

    Many thanks for opening this up to discussion in this way.


  57. The quality of the above comments demonstrate (by and large) just how strong a position education in this country is potentially in to continue the trajectory of improvement we are already on. Many thanks everyone for fascinating reading.

    I think these comments demonstrate that the issues are unfortunately not just a case of small, specific changes to the implementation of a policy – for example making the administration process simpler for schools, by making the default option to leave this administration to HEIs who had the relevant admin staff and experience – but actually I can bring it down to a single point. If I were to list what I thought to be the positive and negative points of SD I would be pretty close to whatever kind of consensus might emerge from collating everything above and nearly all of these could have been fixed if the SD model had developed slowly with continuous reference to existing best practice, and adjustments as necessary to build on the successes of the emerging model and to avoid emerging problems. So the single change that was needed to make this policy work better, was the rate at which it was expanded.


  58. To follow those who have already provided their own analysis of this discussion, I would say that there is a sharp divide between:
    * those who believe that the current model of ITT is already working pretty well and steady improving;
    * those who think the current model of ITT (and perhaps even the wider perception of what constitutes best practice) is fundamentally broken.

    I suspect that most people who read Dominic’s summary will already have made up their mind on this point. Nor is there any significant evidence or argument that is presented here that is likely to change their existing views.

    Maybe it would be best if Dominic produced two completely separate reports, which could be written up like one of Ian Livingstone’s Dungeons and Dragons books. “Do you think the current model of ITT is broken? If yes, go to paragraph 24, if no, go to paragraph 267”.


  59. I’d recommend following this Facebook post for feedback from School Direct trainees, as well as perceptions of School Direct versus ‘PGCE’ (a false comparison, of course, given that many School Direct courses offer a PGCE qual) from prospective teachers.


  60. Ian, You obviously have high expectations of Dominic’s output!

    I take your point & I am not saying everything needs to start from scratch, but that our view of what represents good practice should always be provisional in a world (a) where we hope that technology may change much, (b) where existing views sit on a pretty shaky evidence base. When you step back from the content of ITT (much of which I agree will survive) and look at the type of intervention that government should take in this area, I think you start to see the two halves of the report diverge more clearly – in my view, away from prescription and standardisation of practice (which is always implied by “best practicies”) and towards enabling innovation, which includes ensuring that there are good mechanisms to sort the wheat from the chaff.


  61. I have a wife who has been an initial teacher trainer in a University for 7 years after 35 years in the primary sector classroom or as head teacher. i also have a son who did very badly at Uni but managed to get a SCITT placement and he has since become a very good secondary science teacher.

    I would not defend either route into teaching as necessarily the best as it always depends on the quality of the specific scheme.

    My observation is that his SCITT scheme was not woderful but he made it nevertheless.

    Once my wife acquired responsibility for running the primary ITE programme, i believe she did it as well as she could but the University was the problem as it had awful and mind-numbing bureaucracy. It would not recognise a qualified and experienced head teacher as having just as much, if not more status as a PhD holder with no teaching experience so denied promotion to the teacher.

    The result was dreadfully incompetent management and the introduction of teaching and assessment regimes completely dysfunctional to trainee teachers.

    Despite all this I do not think that the university is necessarily the wrong place to deliver ITE.

    With appropriate recognition of the status of experienced practitioners, Uni is probably a good place to do ITE because it should be able to ensure that trainees get a proper grounding in pedagogy and practical teaching skills. This may not really occur in school direct schemes.

    It would be better to continue to offer both training routes as some people would benefit from one but not the other and vice versa.


  62. Pingback: UPDATE DOC – Open Policy Experiment 1: School Direct and Initial Teacher Training | Dominic Cummings's Blog

  63. Pingback: Evidence is not a substitute for our values? | Dr David Spendlove's Blog

  64. Pingback: What should we do about ITT? | Love Learning....

Leave a Reply

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

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

Facebook photo

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

Connecting to %s