Salaries, bursaries, and fees: too few, too much, too painful?

Justine Greening’s conservative party conference speech included an announcement of a pilot student loan repayment scheme for MFL and science teachers in 25 target local authorities. Details were initially a bit sketchy but have now been clarified. How much this will be worth depends on career progression but for a teacher working FT for the full 10 years, and moving steadily up the scale and picking up a TLR at some point it could average out at getting on for £1000 per year.

This announcement follows a pledge in the Conservative election manifesto, comes along with changes to bursaries for maths teachers, and also echoes a similar but different suggestion from Rusell Hobby, new CEO of Teach First. I thought it might be time to think out loud about the issues around funding, fees and bursaries for ITT. So here is where I would start if this were my decision.

I think that the bursaries for secondary shortage subjects are too high. Paid over 10 months, with no tax or NI deductions, a fair number of my trainee teachers are taking home a fatter monthly pay packet than anyone else in the department where they are training including the HoD. Regardless of the impact of the bursaries on recruitment, that just isn’t right.

What’s worse is that we have almost no evidence on the impact. How many extra trainee teacher applications are made because of the bursaries? How many of those are of decent quality? How many more trainee teachers do we actually recruit? What is retention like for these additional teachers? How many train with only half-hearted intentions of teaching in the maintained sector (or in this country, or at all…)? How many stick out the training through concern procedures/teaching module referrals, to the detriment of children, even though they are no longer intending to teach? These questions have been asked and the DfE and NCTL have been founding wanting.

Those are the obvious questions about the impact of bursaries. The wider question is what else might make an impact on the recruitment and retention problem? There are various people and organisations chipping away at the task of bringing some decent evidence to bear on this but I think we are probably still in the realm of informed hunches. These are mine.

There has been a lot of discussion about having bursaries with strings attached as a way of increasing their impact. Singapore has this kind of system but I don’t think it is right for the UK. I think the spectre of a massive financial penalty for failing to qualify and/or take a teaching job in the maintained sector would be really off-putting to a large number of prospective applicants in shortage subjects. This is the wrong image.

I think, if we really want to improve recruitment, then ITT needs to feel equivalent to the graduate training schemes that attract large numbers of high calibre applicants to high-profile companies, and I’m sure that sort of positioning is one element of Teach First’s recruitment success. This means all teachers should just be paid a sustainable amount during their training.

Some already are paid (SD Salaried, Teach First, and tiny numbers of Troops to Teachers and Researchers in Schools) but it is difficult to just extend this because universities are not in a position to start employing their own students, and the recruitment pattern from SD clearly shows that if all ITT recruitment is done by schools, the shortage subjects will fall woefully short of national targets. However, whilst universities are probably going to have to still work with bursaries rather than salaries, there is potential for considerable re-branding.

The level of this basic salary/bursary ought to be set in relation to qualified salaries and I think the current unqualified teacher rate is probably about right (in relation to qualified salaries only – the real terms cuts to salaries since 2010 are a different issue). However, there is a clear difference in the opportunities outwith teaching for science and maths graduates compared to history, PE and primary, for example, and I think there is a place for a shortage-subject uplift, but this still ought to be in line with likely NQT salaries. These are surprisingly varied but evidence suggests the average in science is maybe £3K more than basic M1 so to me that seems like a justifiable uplift during training. Caution suggests making this change gradually and keeping a careful eye on recruitment, though.

The DfE currently spends about £170 million on bursaries. I think the additional cost of paying a basic salary to all trainee teachers with just an uplift for some would be about £200 million – wallet out! On the other hand, something over £400 million is spent on ‘wasted training’ due to failure to retain these teachers (based on IFS figures). I think the evidence from other countries suggests that the image and status of teaching as a career can have a huge impact on both the quality of the pool of recruits, and their subsequent retention. I think gimmicky bursaries are not helping this image; I think bursaries with strings attached would almost certainly damage it; I think a clear commitment to paid training might help.

Fees for ITT are the other issue. Tempted though I am by the idea of fees being paid for all trainee teachers, I don’t think we’re ready for that yet. With student loan forgiveness or repayment, I think it’s reasonable for trainee teachers on non-salaried routes to use Tuition Fee Loans to cover that cost. However, the pilot student loan repayment scheme, with its 10 year maximum extent, would probably only just about get this £9000+ debt paid off, leaving undergraduate loans untouched. That’s not very tempting compared to choosing a different career and never having the extra debt in the first place.

I assume there is a reason why the pilot is covering loan repayment costs rather than actually writing off debt – I guess it is easier to administer that way. In the long run. though, I think that writing off debt is the preferable option. Partly it just somehow feels better, and as I’ve already said, I think a lot of this is about the image. I also think covering loan repayments is regressive – it is worth £1350/year for someone on £40K, and nothing at all to someone on £23K. That could be seen as leveraging career progression and TLR payments, but it also favours those negotiating higher starting salaries (more often men), and anyone working in London or London fringe.

So I think the DfE ought to look at moving from covering loan repayment costs to writing off debt. This would allow ITT tuition fees to be written off quite quickly – perhaps over three years (although it would be worth looking more closely at the SWC data to see if there is any kink in the retention data that would suggest a particular career point that it ought to cover). Meanwhile, other student loans could be written off at the slower rate. This would maintain the attraction of undergraduate loan repayment, whilst also making the tuition fees less of a burden, provided trainee teachers go on into a teaching career.

I assume there is also a reason for targeting whole local authority areas rather than specifically schools that have particularly difficult recruitment and retention problems. Again, perhaps it is simpler, since a school is clearly either in Portsmouth, or not, whereas targeting schools above a certain FSM or other threshold might leave teachers in borderline schools dipping in and out of the scheme as random fluctuations change which side of the threshold the school sits. Perhaps more generally, a focus on FSMs might push recruitment towards large cities and leave out some coastal and rural schools with tremendous recruitment difficulties. I don’t know what’s best but it should be relatively simple to model. If the DfE haven’t already done so, that needs to happen alongside evaluation of the pilot.

My final thought relates to Teach First and a couple of tweets exchanged with @russellhobby the new CEO of Teach First. I think Teach First has brought some tremendous new ideas into ITT, has demonstrated that teaching can be an attractive career at a time when that was proving difficult, and has undoubtedly produced some very effective teachers. However, it’s important to remain aware that this route is significantly more expensive for the taxpayer than other main routes, and retention over 5 years is a fair bit lower. My concern was that any loan reduction based on FSM thresholds would immediately start benefitting all Teach First participants as soon as they started employment at the beginning of their first year, whilst university-led trainees were racking up more debt. And whilst there are obvious reasons for targeting schools on the basis of FSMs, that also neatly guarantees Teach First participants a benefit whilst leaving it only as a possibility for other routes. I may just be biased! My suggestion of accelerated write-off for PGCE loans might be unfair in the other direction. The current pilot seems equally fair to all routes and I’m cautiously hopeful that it’s a step in the right direction.



Learning Styles, Universities, and ITT

Sharks have teeth that slope inward; once they get their jaws on something, it isn’t getting away. I don’t think @tombennett71 has teeth that slope inwards (although, like me, he grew up in a land and era of pragmatic dentistry, far from the pearly white, braced and straightened paradise of the Home Counties) but there is no doubt that he has had his teeth into Learning Styles for some time now, and he’s not letting go. Mostly this is a good thing but I think the recent little flurry of tweets about the failings of university-led ITT involved tarring us with a rather broad brush.

So I thought it might be useful to flag up a few things, partly to try to explain why this is (or has been) an issue in ITE, and partly to explain why Googling “x university” + “learning styles” is a poor way of finding out about what is happening.

The people working on PGCE courses in universities have a mixture of backgrounds. Nearly all will be qualified and experienced teachers. Some will have moved straight from a school job into a ‘Teaching Fellow’ post – one with no research element to their contract – and some will be Lecturers or more senior academics with both a teaching and a research role. There is, I think, an assumption that we all have a lot of time to stay abreast of broad swathes of academic literature but for most of us that is not really any more true than it is for a teacher or senior leader in school. Those with a research element to their contracts of course have an obligation to be right up-to-date with their narrow area of expertise but that’s quite time consuming; Teaching Fellows have a full-time job teaching and looking after trainee teachers, and probably other students.

And so what happens in universities is, in my limited experience, the same as what happens in schools. We draw on our personal experience, the reading we have done, and everything else we pick up along the way – particularly from colleagues – to plan and teach as well as we can. Learning styles was so ubiquitous in schools that, without enough critical thought, it was bound to end up in some teacher training. Hopefully we have all learned our lesson about taking on ideas that sound convincing without actually checking the supporting evidence, and will pay more attention to critical analysis, such as Sharp, Bowker and Byrne (2008) – from colleagues at @BGUlincoln (ironically), @UniofExeter and @SotonEd. However, the rush to embrace Mindset in schools suggests ResearchEd, and anyone in a position to influence schools’ thinking, have their work cut out.

Direct confirmation from recent trainee teachers is one thing; trawling Google for university webpages that mention learning styles is quite another. If you Google “Soton learning styles” you will find this page on ‘picking your profile’. Feel free to read (if you must) but it is a classic example of the genre. How can we square this with a research-informed PGCE? Dig a bit deeper and you’ll see this has nothing to do with the Education School, never mind the PGCE. This is not a primary or secondary school where the SLT oversees all teaching and learning. The Education School makes up less than 2% of the university. Holding us accountable for this webpage, or implying it affects our PGCE, is like holding the literacy lead in a primary school responsible for the way literature is taught in a secondary that happens to be in the same MAT.

For what it’s worth, here is our position statement on learning styles. If anyone else wants to make use of, or adapt it (or offer constructive criticism in the comments), then provided due credit is given, you are welcome to do so.

University of Southampton Education School Initial Teacher Education: Position Statement on Learning Styles

As a leading school of education, with an internationally recognised research profile, and a long and successful history of initial teacher education, we are committed to ensuring that our teacher education programmes are fully research-informed and reflect current best practice.

There has been considerable confusion, and a significant amount of poor practice, associated with learning styles. In particular, the matching hypothesis suggests learning is enhanced when the mode of delivery is matched to the preferred learning style of a student. This is a widespread belief but is not supported by the evidence (Pashler, et al., 2009). There is a plethora of different, disconnected models, many based on incoherent theory and/or tests which are neither reliable nor valid (Coffield, et al., 2004) but it is the Visual, Aural, Read/Write and Kinaesthetic model (VARK), often simplified to Visual, Aural, Kinaesthetic (VAK) which is typically encountered in school settings (Sharp, et al., 2008) to the potential detriment of children.

However, there is evidence that presenting new ideas in several different ways is helpful to learners. This is not because of simplistic differences in learning styles but because learners benefit from multiple opportunities to relate new ideas to their existing knowledge (Nuthall, 2007), and benefit from multiple modes of presentation. This is particularly true when information is presented in both verbal and graphical form at the same time (Pashler, et al., 2007).

It is therefore our policy in initial teacher education at the University of Southampton Education School to avoid presenting simple learning styles models (including VAK), and the matching hypothesis, as useful. Since the VAK model and the matching hypothesis are still so prevalent, this misconception should be actively addressed. However it is appropriate to encourage teachers to present new ideas in a variety of ways that are appropriate for the material (Willingham, 2005) and to emphasise the particular strategy of combining verbal and graphical formats where relevant.

However, it is also recognised that learning styles, now more often referred to as cognitive styles, is an area of active research. More sophisticated models are being developed which better reflect the evidence from cognitive psychology and neuroscience (Kozhevnikov, et al., 2014) and it is clear that there are differences in the way in which people approach problems and learning opportunities. Evidence suggests that, whilst different people will have different capacities (Kohzenikov, 2007), most will adapt their approach depending on the situation, and most people can benefit from developing their ability to apply different approaches to best meet a particular challenge (Zhang, et al., 2012). This is often described as a meta-cognitive skill. Work has been carried out to try to provide a framework for teachers to make use of this understanding (Evans & Waring, 2015) but our view is that this is not yet adequately tested in classroom settings, and may not be simple enough to be of practical benefit to teachers working in complex school and college settings. In accordance with our commitment to draw on the best available evidence from research, we will keep this situation under review.

Coffield, F., Moseley, D., Hall, E. & Ecclestone, K., 2004. Learning /styles and Pedagogy in Post-16 Learning: A Systematic and Critical Review, London: Learning and Skills Research centre, LSDA.
Donker, A. et al., 2014. Effectiveness of learning strategy instruction on academic performance: A meta-analysis. Educational Research Review, Volume 11, pp. 1-26.
Evans, C. & Kozhevnikov, M., 2013. Styles of practice in higher education: Exploring approaches to teaching and learning. London: Routledge.
Evans, C. & Waring, M., 2012. Application of styles in educational instruction and assessment. In: L. F. Zhang, R. J. Sternberg & S. Rayner, eds. The Handbook of Intellectual Styles. New York: Spring, pp. 297-330.
Evans, C. & Waring, M., 2015. Understanding Pedagogy: Developing a Critical Approach to Teaching and Learning. Abingdon: Routledge.
Kohzenikov, M., 2007. Cognitive styles in the framework of modern psychology: toward an integrated framework of cognitive style. Psychological Bulletin, Volume 133, pp. 464-481.
Kozhevnikov, M., Evans, C. & Kosslyn, S., 2014. Cognitive Style as Environmentally Sesitive Individual Differences in Cognition: A Modern Synthesis and Applications in Education, Business, and Management. Psychological Science in the Public Interest, 15(1), pp. 3-33.
Nuthall, G., 2007. The Hidden Lives of Learners. Wellington: NZCER Press.
Pashler, H. et al., 2007. Organizing Instruction and Study to Improve Student Learning, Washington DC: National Center for Education Research, Institute of Education Sciences, U.S. Department of Education.
Pashler, H., McDaniel, M., Rohrer, D. & Brork, R., 2009. Learning styles: concepts and evidence. Psychological Science in the Public Interest, 9(3), pp. 105-119.
Sharp, J. G., Bowker, R. & Byrne, J., 2008. VAK or VAK-uous? Toward the trivialisation of learning and the death of scholarship. Research Papers in education, 23(3), pp. 293-314.
Willingham, D. T., 2005. Visual, auditory, and kinaesthetic learners need visual, auditory, and kinaesthetic instruction? American Educator, Volume Spring 2005, pp. 31-35.
Zhang, L. F., Sternberg, R. J. & Rayner, S., 2012. Handbook of intellectual styles: Preferences in cognition, learning and thinking. New York: Springer.


Times tables, retrieval practice, and what I’ve learned from Y3

My son and I have just made a major breakthrough with his times tables. I think it’s taught me something about retrieval practice that I hadn’t appreciated before.

My son attends a lovely school, as far as I can tell his teachers are great, and he was “working at greater depth” across his KS1 assessments (though I believe it was quite tight for writing). He is a boy who would rather be tearing around outside or playing a noisy board game than doing something calm and studious like colouring or reading but he was one of the first children in his peer group capable of sitting still through a film at the cinema, safely crossing roads, and getting to school on his own.  He was widely considered to be the noisiest baby anyone had ever come across – babbling and shouting away to himself, with or without an audience, and he rarely stops talking now, but seems to have good self-control at school and his teachers always describe him as thoughtful, co-operative, and responsive to feedback. At home, I have to say it’s sometimes more like rude, obstinate, and apparently deaf, but he is a great little boy and I’m very proud of him.

He seemed to pick up counting and basic number sense pretty easily, one more, one less, place value, number bonds, adding and subtracting 10, counting in 2s, 5s and 10s and coin values from the bits of practice we were asked to do at home. However, in contrast to the quick grasp of the 2, 5, and 10 times tables, 3 and 4 seemed to be difficult. He could manage them in order but random – no. Now, I know almost nothing about teaching or learning any of this, and although I can see that 3 and 4 lack the same patterns as 5 and 10, I don’t know if they really are a lot harder. Regardless, he didn’t seem to be getting them, and he seemed to be getting frustrated, so I thought it would be a good idea to make a concerted effort to support him. I didn’t think this would be very hard. Surely a brief daily (spaced) practice, interleaving 3s and 4s, would soon crack it. But nay! It was like herding cats.

Each day, he would remember some and forget others. 1x, 10x and 11x were okay, and usually 5x but the others came and went. Even more frustrating, if he got stuck on say, 7×3, I would give him the answer and then do maybe 10×3, 7×3, 11×3, 7×3, 5×3, 7×3, 2×3, 7×3 and he would still be no more likely to get 7×3 on the last go than the first, however many times it had been “21”. Sometimes 7×3 would be good for a few days, then gone again, like a fart in a car. What was going on. “Hold your nerve” I said, to myself. “It’s just practice, it’ll come” I said to him. But after about a month, it still hadn’t. So I had a re-think.

If he wasn’t getting it, maybe it needed simplifying. We put a hold on the 4 times table and just did 3s. No joy. What was the problem? Maybe still not simple enough, so I tried just doing 1×3, 2×3, 3×3 and 4×3. And we finally cracked it! He got those four to a pretty automatic level in two or three days – it may have been faster but I needed to see it hold to know. Then 5×3, 6×3, 7×3, 8×3, with three of those having been a problem for a month. No problem now. Then the last four multiples and within just over a week we were there – the whole table completely at random with speed and accuracy. The 4 times table took another week or so, even with practice on the 3s to get in the way, and now we’re on to 6s.

Even better, it was like pulling teeth even to get him doing just a few minutes each day, and now, whilst he would certainly rather be playing Crossy Road, he is a lot more amenable to the whole thing.

It’s not perfect yet. If he isn’t fully concentrating then he has a tendency to give answers like 6×4=18 through paying more attention to the 6 than the 4 but it’s continuing to improve, even as we get stuck in to 6s, and I’m pretty confident we will have them all completely solid by the end of Y3, and maybe sooner.

So what have I learned? With retrieval practice it seems that biting off too large a number of items to learn can, at least for my son, prevent any of them from sticking. And when it doesn’t work quickly, don’t hold your nerve; reduce the number of items until some success is experienced. I have no idea how that relates to the evidence on retrieval practice – will have to ask or do some more reading.