Sort out physics teacher preparation courses

The next couple of years likely to see the best recruitment levels to physics ITT courses for more than a decade. As a result, there might be a risk that everyone concerned with teacher preparation breathes a huge sigh of relief, and put the problem of the shortage of teachers of physics in the ‘job done’ bin. In my view that would be a big mistake.

Now is the time for someone, perhaps the Institute of Physics, NfER, Nuffield or Gatsby to consider a research project that looks at the pipeline of physics teachers from school to school, and notably from university to teaching. Do different courses produce different numbers of teachers of physics that stay in the profession, and become the leaders of tomorrow or just provide short-term additions to the teaching stock. How important is a middle leadership cadre?

 Mapping these outcomes both geographically and as between public and private schools, and within the public sector as between 11-16; 11-18 and post-16 institutions might create an understanding that could then lead to a debate about how every child could access high quality physics teaching on a regular basis up to Level 3.

With the improvement in mathematics in schools over recent years, there should be the possibility of increasing interest in physics, especially amongst girls. The percentage of girls taking physics is still lamentably low. This is despite 30 years of programmes such as WISE. How far has the lack of management of the scare resource that is teachers of physics held back the encouragement of more girls to study the subject?

Teaching has always looked to be a profession where there is basic pay equality. That’s fine when there aren’t shortages, but there have always been incentives and rewards from golden hellos to additional payments for working in challenging schools. What incentives work to keep teachers of physics in the profession. Is it non-pay matters, such as not having to teach ‘all sciences’ or some mathematics that is as important as pay?

If gender is an issue, what about ethnicity: of both teachers and those that study physics at school? Then there is the issue of what percentage of pupils on free school meals have access to high quality physics teaching? Is it different from those small numbers on free school melas in schools in affluent areas, compared with schools where a large percentage of pupils are on free school meals. In the latter schools, attracting a physics teacher means access for some pupils. In the former, even if there is a physics teacher do the pupils on free school meals have access to physics?

And what about pupils with SEND? What is their access to physics teaching like?

Physics could be a template for other subjects to ask the questions about, ‘what can we do to ensure we have the best system for preparing teachers, recruiting them into schools, and ensuring that they stay in the profession.’ The alternative is that we could carry on as before, and rely upon market forces to provide the Nobel Prize winner of the future.

Is discipline worse in schools?

It was interesting to hear Laura McInerney and Tom Bennett on the ‘Today’ programme on BBC Radio 4 this morning discussing whether or not behaviour was worse in schools these days than in the past. Both are experienced commentators, and Tom led a review in 2017 for the then government, about behaviour in schools. It is also interesting to see the BBC taking an interest in schools. The World at One last Sunday (also BBC Radio 4) devoted the whole of the programme to an analysis of the SEND issue. Interestingly, there was no government spokesperson available on Sunday, so they had to make do with the chair of the Education Select Committee.

The discussion this morning was around whether or not behaviour had worsened in schools, and if so, why? The usual suspects, covid and mobile phones were trotted out in support of discipline being worse in schools, along with families facing multiple challenges, but there were precious few facts.

One way of measuring the state of discipline in schools is by looking at the number of permanent exclusions each year by schools.  The largest single reason each year for these exclusions is always ‘persistent disruptive behaviour’. So, this might be seen as a good proxy measure for how schools are faring in relation to discipline in the classrooms. Of course, this measure doesn’t pick up low level disruptive behaviour, but it is reasonable to assume that there is a correlation between the different levels of behaviour in schools.

Looking back over the past 30 years, the level of recorded permanent exclusions was 10,440 in 1998/99. The level fell to 5,040 in 2010/11. In the latest year, 2023/24 there were 10,885 permanent exclusions. On the face of it, discipline is getting worse again, but is only back to levels last seen at the end of the last century.

I would like to suggest to causes not mentioned on the ‘Today’ programme: teacher supply and school funding. Is there a causal relationship between the fact that permanent exclusions were at their lowest when schools were fully staffed, and had experienced a period of several years of significant funding by government.  By contract, permanent exclusions seem to rise when there is difficulty staffing schools, and when funding is less than might be expected in a civilised society.

So, is the answer as simple as proper funding and staffing if you want fewer exclusions? The age and experience of the teaching force might also play a part. More experienced teachers, as I can testify from personal experience, are much less likely to face discipline issues then new entrants, especially if they are unqualified.

In the latest statistics on exclusions, 13 of the 25 local authorities with the lowest rates of permanent exclusions were London boroughs. This just adds more evidence to my thesis that if the rest of the country were funded like London, schooling would be in a much better place across the country.  Although I was also pleased to see Oxfordshire in 10th place overall for the lowest rate of permanent exclusions.

Ethnicity issues remain for new teachers

Data published by the DfE for the 2023/24 postgraduate cohort of trainees achieving QTS (Qualified Teacher Status) showed that differences still remain in regard of entry into the teaching profession between different ethnic groups on two counts. Firstly, the percentage trainees achieving QTS, and secondly, the percentage employed in State-Funded schools.

EthnicityPostgraduate
2023/24
Number of trainees
TotalAchieved QTSDid not achieve QTS
Total Teaching in State-Funded SchoolTotal
Asian / Asian British2,9312,7041,868227
Black / African / Caribbean / Black British1,2061,106821100
Mixed / Multiple ethnic groups87281461158
Other Ethnicity41038026830
White16,08215,04811,5701,034

Initial teacher training performance profiles, Academic year 2023/24 – Explore education statistics – GOV.UK

The outcomes were more favourable for the dominant White group. 94% of this group achieved QTS, compared with 93% for the Black and Mixed Group’s trainees and 93% from the Asian Group.

Teachers from the White group were also more likely to be working in a State-Funded school by the time the data was collected. 77% the White Group that had achieved QTS were teaching in a State-Funded school. This compared with 75% of the Mixed and Multiple ethnic groups; 74% of the black Group and just 69% of the Asian group.

Despite the number of teachers from the Asian group not working in State-Funded schools, this group still accounted for 12% of entrants. The black group accounted for 5% of entrants and Mixed and Other ethnic Groups together made up another 6% of entrants. The White group, of some 11,570 teachers, accounted for 76% of entrants. 

Sadly, it isn’t possible to track the whole journey from application to train as a teacher through to working as a teacher with QTS for each ethnic group. It is also not possible to see whether certain routes are more or less favourable to certain ethnic groups. Both those sets of data would help illuminate possible areas of concern within the process of how graduates become a teacher.

The DfE has also included some experimental regional data about outcomes that would also be useful to see by ethnic groups because, historically, the Asian and black groups applicants and trainees have tended to be concentrated in specific urban areas.

The DfE data suggested that fewer trainees achieving QTS were employed in State-Funded schools in the north of England than the south. The region with the highest percentage employed was the East of England, were 82% of those achieving QTS were employed in a State-Funded school. 

Another indicator where there are no published data on trainees by ethnic groups are in respect of new graduate and career changers. For the purpose of the employment data for those with QTS, the DfE splits the trainees into two groups; those under 25 that might be assumed to be mostly new graduates and those over 25 that will predominantly be career changers. Those in the over-25 group were less likely to achieve QTS, but there was no difference among those with QTS in terms of the percentage working in State-Funded schools.

Less than 400 teachers of physics entered service in 2023/24

As a result of the latest data from the DfE, it is now possible to start to see the consequences arising from the collapse in ITT recruitment to secondary courses in 2023/24.

It is interesting to compare the turnaround from the improvement in recruitment brought about by the covid epidemic with the poor recruitment figures for just three years later.

Year Achieved QTSAchieved QTSDid not achieve QTS% Did not achieve QTS
2017/182549013045
2018/192640213485
2019/202754211914
2020/213070616065
2021/222971522417
2022/232243718408
2023/242121015507

Initial teacher training performance profiles, Academic year 2023/24 – Explore education statistics – GOV.UK

The reduction in numbers achieving QTS (Qualified Teacher Status) from 30,706 in 2002/21 to 21,210 in 2023/24, a reduction of some 8,500, or more than a quarter in just three years, goes some way to demonstrate the depth of the problems with recruitment schools faced in September 2024 and even more so in January 2025.

Hopefully, these numbers represent the lowest point for secondary trainee numbers over the next few years, after adjusting for the changes in targets that result from falling rolls in schools.

The data on the percentage achieving QTS might suggest that when it is difficult to recruit trainees, more marginal candidates are offered a place by training providers, and, as a result, the percentage not achieving QTS increases.

Because the DfE only records those with QTS teaching in a State-Funded ‘school’ it does not count those employed in a Sixth Form College or other further education college and, does not count any teachers from the cohort with QTS working in the private school sector. This latter omission might account for why only 49% of those trained to teach Classics were teaching in a State-Funded school.

SubjectPostgraduate
2023/24
Number of trainees
TotalAchieved QTSTeaching in a State-Funded School (of those achieved QTS)Teaching in a State-Funded School (of those achieved QTS)
Classics69673349%
Physical Education1,4851,43896567%
Primary9,3788,7126,27172%
Business Studies22720114673%
Art & Design40138228675%
Computing41337428275%
Total22,76021,21015,92175%
Modern Foreign Languages1,02497073976%
Other32731023776%
Drama24622717577%
Physics54048537377%
Secondary13,38212,4989,65077%
Geography82277160178%
Chemistry76469154479%
Mathematics1,9001,7621,38579%
Biology88581865280%
Music23722317980%
Design & Technology57654043781%
English2,2102,0621,66181%
History97791974481%
Religious Education27925821282%

It is interesting to note that the percentage of those trained to teach physics teaching in a State-Funded school was the same as the overall average, at 77%. However, that meant that there were only 373 new physic teacher entrants into State-Funded secondary schools from this cohort. It would be interesting to know the routes these 373 took before taking up their initial posts? How many of these 373 were from the PG Salaried High Potential ITT route? (Better known as Teach First).

The relatively small percentage of physical education trainees recorded as working in State-Funded schools may well be a result of the numbers recruited into training compared with the target.

Where trainees are required to pay a tuition fee for a course, what responsibility does the government and the course provider have to the trainee to ensure that there are not way too many trainees for the number of vacancies likely to arise? If this means that the student debt is less likely to be recovered, is this a waste of public money?

In a future post I will explore some other aspects of this dataset.

It’s a funny old world

On the day when nurses look as if they will join resident doctors in demanding more pay, figures about applications from graduates to train as a secondary school teacher hit decade high levels, even after removing the degree apprenticeship numbers from the totals. This month, according to DfE data, 58,880 candidates have submitted one or more applications to train as a teacher. This compares with 46,696 list July and 45,000 in 2108, before the pandemic. Initial teacher training application statistics for courses starting in the 2025 to 2026 academic year – Apply for teacher training – GOV.UK

This July, there were 36,283 candidates applying to train as a secondary school teacher, compared with 17,997 wanting to train as a primary school teacher.

By comparison in July 2018, 26,060 women had applied, whereas in July 2025 that had increased to 31,439. However, applications from men had increased from 12,680 in 2018 to 18,904 this July

Traditional higher education and SCITT courses still account for the bulk of the routes into teaching selected by candidates. However, candidate numbers on traditional salaried routes were down this July, from 8,927 to 7,636, but that may be partly the 7,332 candidates that have applied for the Postgraduate teaching apprenticeship route, up from 6,433 last July.

The new Teacher Degree Apprenticeship route that has attracted 1,079 candidates so far this year. This is a new route and, presumably isn’t open to graduates.

Although applicant numbers from the ‘rest of the world’ group are down this July, from 9,586 in July 2024, to 8,563 this July – this number still represents nearly 20% of all candidates.

Some subjects, including art, physical education, physics, mathematics and computing have recorded their highest level of ‘offers’ this year since the 2013/14 recruitment round. How many are multiple offers or from candidate’s not able to fulfil visa requirements won’t be known until the courses start in just over a month’s time.  Interestingly, offers for English courses are below the number of offers made in July 2024.

Despite the significant increase in candidate numbers, some subjects will not hit their targets set by the DfE this year. Subjects most likely to miss their targets are business studies, drama, religious education, music and design and technology. In English, it looks touch and go at this moment in time as to whether or not the target will be hit.

In some subjects, such as physical education, where the target is 725, there is a risk of a significant overshoot in offers. Such a situation might leave large numbers of trainees with additional debt and little chance of a teaching post in England next summer. The DfE will need to be alert to this issue, especially if the growth in ‘AI’ changes the labour market for those with degrees in physics and mathematics, so as to make teaching look like an interesting career at current salary levels.

It would be a funny old world if incentives to train as a teacher had to be switched from mathematics and the sciences to English and the arts.

The crisis in physics teaching

NfER has published some interesting research about the distribution of physics teachers A widespread lack of specialist physics teachers persists due to recruitment and retention challenges – NFER The most alarming statistic in the report is that 26% of state-funded secondary schools that responded to the School workforce Census had no qualified physics specialist in their science department.

However, there is a caveat to making too much of the data. This is because it is taken from the School Workforce Census. As this is a self-reporting census, the data must be regarded with a degree of caution, as there could be some under-reporting.

As the School Workforce Survey is conducted by the DfE each autumn term, it should have a degree of reliability. However, the NfER report only contains data from 2,296 of the 3,456 state-funded secondary schools in England.

Even so, assuming all the remaining schools have at least one qualified teacher of physics that would mean at least 12% of schools were without a qualified teacher of physics, and more than a third of schools (36%) has either no teacher or only one teacher.

Now some of these schools are 11-16 schools, and a few the remaining middle schools classified as secondary schools. These schools don’t need a teacher for ‘A’ level courses. But who is teaching the GSCE physics courses, and how many pupils from these schools go on to study physics at ‘A’ level?

For the 11-8 schools with no qualified teacher of physics, what arrangements are being made for pupils that want to study the subject at ‘A’ level. If it is matter of having to change school, then what are the costs to the pupils and their families. This is another example of where transport costs may affect choice of courses post-GCSE.

Do schools support each other? This was easy when all schools were maintained schools. In the 1960s, the local girls’ school where I lived could not support Chemistry ‘A’ level, and those girls wanting to study the subject joined the ‘A’ level class at the school I attended. This must be more challenging to arrange these days with competing Multi Academy Trusts.

Interestingly, if you add up all the qualified teachers in the table in the NfER survey it amounts to more than 3,500 qualified teachers of physics: enough for one for every school. However, our distribution system for teachers is based upon open market principles, with teachers free to apply for any post, and teach where they like. Is this the best system for the education of all children, if it means that some are deprived the opportunity to study subjects such as physics because there is no qualified teacher?

Hopefully, the present position marks the bottom of the staffing cycle, and improved interest in teaching, as reported in this blog and on my LinkedIn pages, means more trainees with emerge into the labour market over the next few years.  The issue then will be how to create teaching posts for them. Wil schools be required to either redeploy an existing member of staff or make them redundant? Those schools with falling rolls and a stable staff might find the former difficult. What is needed is a national plan for physics, and perhaps other subjects where there are teacher shortages. But, sadly, I doubt we will see such a radical idea from this government.

Class matters more than ethnicity

The end of the summer term is a curious time to announce an inquiry into White working-class kids in schools. The inquiry seems to be funded by private finance, but with government backing. Members revealed for white working-class kids inquiry

Two former Secretaries of State will be on the board, along with a DfE official, as well as many others representing the great and the good in schooling, but not perhaps either the churches or representatives of the under-fives lobby.

As SchoolsWeek pointed out in their news item, this is not the first such inquiry into the achievements or lack of them, of this group in society.  Indeed, the House of Commons Select Committee has had two goes at the issue, in 2014 and 2020. HC No

As well as the Select Committee’s reports, and the evidence submitted to the Committee, The inquiry might also like to read the DfE’s Report on outcomes by ethnicity Outcomes by ethnicity in schools in England – GOV.UK published before the pandemic.

I am sure the inquiry will focus on what works, and no doubt discuss issues about what is being measured and over-reliance on Free School Meals data. They will also need to discuss the issues around definitions, as society has become much both more complex, and more polarised. The measurement of children – I prefer the term to kids – of mixed heritage has added many more sub-categories to the original list.

However, I cannot help thinking that the focus of the inquiry is wrong. All the evidence suggests that of the three factors of race, gender and class, it is the third one that really matters. Yes, they are often inter-related, but looking at socio-economic data it is often schools in deprived areas, regardless of the ethnicity of their pupils that fare less well in school performance table.

Is this due to the funding arrangements. Some areas, notably London, are better funded than other parts of England. Is it down to teacher deployment and the market system. Do the best teacher seek to work in the most challenging schools or those with the best outcomes. How much does support from home matter. Can poor teaching be overcome with support and tutoring from home. All these were issues considered by the Select Committee. Then there are issues such as school attendance and what happens at the Foundation State if pupils miss vital building blocks in language and mathematics. Does the class teacher system help or hinder these children?

In terms of funding, what effect has the Pupil Premium had on outcomes, and is there any evidence that where academies can pool the funds of all schools and move resources between schools whereas local authorities cannot do so that this arrangement can boost outcomes in traditionally under-performing schools?

I guess one measure is the percentage of pupils on Free School Meals across the country that pass the tests for selective schools. Will the inquiry suggest a fully comprehensive secondary school system? If not, how will it address this injustice.

I am disappointed that it has taken this Labour government a year to start the process of addressing this issue. What were they doing in opposition? After all, the Liberal Democrats pushed the Pupil Premium right for the start of the coalition in 2010, as it had been in their manifesto.  How much does this government really care about those children that don’t achieve their full potential for whatever reason.

Falling rolls

The projections for pupil numbers up to 2030 were issued by the DfE this week.

At the same time the Office for National Statistics (ONS) looked at the likely size of + the school population  aged 5-15 National population projections – Office for National Statistics both have implications for the demand for teachers in England. This data can help determine how many teachers will be needed to staff schools and the attractiveness of teaching as a career.

The DfE concluded the following for the remainder of this decade;

State-funded nursery & primary schools

  • The overall pupil population in these school types is projected to be 4,205,000 in 2030. This is 300,000 (6.7%) lower than the actual population in 2025 (4,505,000).
  • The revised projection for 2028 of 4,319,000 pupils represents a 5.4% fall from 2024; this is 38,000 lower than the previous forecast, as last year’s projection showed a 4.5% fall over the same period.
  • The nursery & primary population is therefore still projected to drop and is doing so at a faster rate than previously projected.

State-funded secondary schools

  • The secondary school population is projected to be 3,135,000 in 2030. This is 97,000 lower than the actual school population reported in the 2025 school census of 3,232,000.
  • The revised projection for 2028 of 3,208,000 secondary pupils represents a 0.8% fall from 2024, and is 55,000 (1.7%) lower than last year’s projections which reverses the previously projected 0.9% increase over the same period. 
  • The pattern of change in the secondary school population seems to indicate that it plateaued between 2024 and 2025, will remain at a similar level until 2026, and is then projected to start declining slowly. This suggested plateau is earlier than the peak in 2026 projected last year but will be subject to change if pupil numbers bounce back, fluctuate, or continue to plateau in future years.
  • The actual number of secondary school pupils in 2025 fell slightly as more pupils moved out of the secondary phase than moved in. Changing historical birth rates and trends in net migration are likely driving factors. National pupil projections, Reporting year 2025 – Explore education statistics – GOV.UK 

According to the ONS projections, there are expected to be fewer children in the UK by the middle of both 2032 and 2047, compared with the middle of 2022. The ONS have based this on a view that the assumed fertility rates in the 2020s and 2030s will be even lower than those around 2001, when UK fertility reached a record low.

By mid-2032, the number of children (those aged from 0 to 15 years) is projected by the ONS to have decreased by 797,000 (a fall of 6.4%), from 12.4 million to 11.6 million. By mid-2047, the number of children is projected to remain around the mid-2032 levels. Of course, if the birth rate changes, these numbers could be either too high or an underestimate. Either way planners are no expecting the need for more teachers due to changes in the birth rate.

However, these numbers are just assumptions, and don’t take into account other changes in the numbers in the 0-15 age groups resulting from any net balance once both emigration and immigration have been taken into account.

Other factors, such as the departure rate from teaching, due either to the popularity of teaching or the availability of alternative employment opportunities, as well as the age-profile of the profession and the number of teachers retiring can affect the demand for teachers. Policy changes concerning how children are taught can also affect the demand for teachers. How will AI and attitudes to issues such as home schooling are two imponderables that might affect the demand for teachers.

These days, as well as the traditional modelling, the DfE has access to up to the minute data on vacancy rates for teachers, and how posts are filled. This new data, allied to more traditional methods of estimating the demand for teachers, should help to ensure teacher unemployment can be kept to a minimum. After all, there is little point in spending money training teachers only for them to be both burdened with debt and unable to find a teaching post.

1,500 posts and counting

When I wrote my first post on this blog, on the 25th of January 2013, I little though that I would reach 1,500 posts. However, despite stopping posting for 18 months, between the autumn of 2023 and May this year, while I was otherwise occupied as a cabinet member on Oxfordshire County Council the blog has now reached the milestone of 1,500 posts, including 40 so far this year since I started the blog up again this May.

Since one of the features of the blog has been commenting on numbers, here is a bit of self-indulgence. The blog has had 175,983 views since its inception, from 93,875 visitors, and has attracted 1,459 comments. The average length of a post has been between 550-670 words, although there have been a few longer posts in response to consultations and Select Committee inquiries.

How much holiday do teacher have? is the post with the most views – more than 6,500 and rising. Some posts have had no views, but are still an important record of my thoughts. The United Kingdom has been responsible for the most visitors: not a surprise, as most posts are about education in England. However, the USA comes second, with more than 15,000 views. Apart from some former French speaking countries in West Africa, Greenland and Paraguay, almost all other countries have had someone that has viewed the blog at least once.

Later this year, I will be publishing a book of the 2013 posts from the blog, and at that point they will disappear from public view. If you want to register for the book, check on Amazon after August 2025 or email dataforeducation@gmail.com for publication information. Alternatively, ask your favourite bookshop or library to order a copy.

I am sometimes asked about my favourite post. With 1,500 to choose from, that’s difficult, as many haven’t seen the light of day for a decade or so. However, Am I a blob? From 2013, was fun to write, and the posts about Jacob’s Law finally brought about a change in the legislation over admissions in the current bill going through parliament.

Most posts have been written, as this one is, in one session from start to finish, with editing just to tidy up my thoughts. Some are more passionate than others, and many are about teacher supply issues, where I am also researching a book on the subject covering the past 60 years of ‘feast and famine’. Much of the recent history has been well chronicled in this blog.

Thanks for reading, and for the comments. Who would have thought that someone that failed ‘O’ level English six times would end up writing a blog!  Funny old world.

Where should Teach First recruit its trainees?

There have been some interesting discussions recently on the LinkedIn platform about Teach First, and its possible extension beyond its original scope of recruiting from the Russell Group of universities after SchoolsWeek revealed this condition might be altered when the contract is re-tendered for the scheme. Teach First: Labour plans recruitment scheme revamp

Two points are worth making about the discussion. Firstly, the universities within the Russell Group have not remined the same since Teach First was established more than twenty years ago. Secondly, when faced with challenges in filling its target for recruiting teachers, Teach First does seem to have already extended its reach beyond the Russell group. In it 2024 annual report to the Charity Commission it said that:

‘Increasing the proportion of trainees from Russell Group universities compared to the previous year and sustaining the proportion of trainees with a first-class degree despite a decline in the number of firsts awarded.’ (Page 10, 2024 accounts with Charity Commission)

However, it didn’t provide any details of the number of non-Russell Group trainees recruited, and in which subjects. This is an important issue because of the schools where Teach First place their trainees. Historically, schools within the M25 with high percentages of disadvantaged pupils were the main focus of the programme, although in recent years it has spread more widely across the country while keeping its core mission.  

An analysis of, for instance, the percentage of new physics teachers recruited through Teach First and the schools they were placed in, and subsequently went on to work in, would be interesting, especially if compared with the distribution of new teachers of physics across all schools with similar levels of deprivation in the parts of the country not covered by Teach First.

Another interesting issue with regard to Teach First is the cost of recruiting their teachers. I saw a comment that surprised me about ‘needing to interview applicants because of AI generated applications’. I thought that all qualified applicants would have been interviewed as a matter of course.

This caused me to look at the cost of recruitment to the Teach First programme. Their accounts with the Charity Commission suggest that in 2023 the charity spent just over £7 million on recruitment and then £6,587,000 in 2024. Now, in 2023, it recruited 1,417 trainees, including to the pilot SCITT programme. In 2024, with the development of the SCITT programme, some 1,419 trainees were recruited. If the financial data is correct, then that would mean more than £4,000 to recruit a trainee in 2023, falling to £3,800 in 2024.  I wonder whether other ITT courses spend anything like this amount on recruitment?

Of course, some of the expenditure is offset by donations to the charity, and during a period when recruiting new entrants to teaching is a challenge, recruitment costs would be expected to be high. Although when recruitment to teaching is buoyant, as it may well be over the next few years, the overall cost may be higher because there are more applicants to process, especially if Teach First is opened up to a wider range of graduates seeking to become a teacher and interviews more applicants. How much should we spend on recruiting trainees teachers and how good are we at obtaining value for money on recruitment overall, including the national TV advertising campaigns?