Sunday 3 September 2017

A Policy A Day: Teach First

In the lead-up to the election, we are examining a policy a day. We're exploring a variety of policy areas, explaining the background and analysing some of the policy options, with a mixture of technocracy and values-based approaches. Inevitably, some opinion will make its way in and we make no apology for that - after all, we're voters too. A list of all the articles is available hereEnjoy!

Today's post is by Simon Johnson

There are many things that our next Minister of Education could do if they wanted to make education better:
  •   Give more children access to high-quality ECE
  •   Reform NCEA
  •   Refurbish or rebuild our many failing school buildings
  •   Encourage schools to pool their human and financial resources and collaborate to improve learning
At the same time, the Minister must help schools to re-orientate themselves to prepare students for a world where:
  • The average learner will live to be 100 years old – and will need educating for many different careers over their lifetime [1]
  • Mid-skilled occupations, of the sort that the average Kiwi relies upon, will increasingly be automated. More and more workers will need to interact with AI as a core part of their job.
  • Students will have to compete with increasing numbers of English-speaking workers from developing markets who demand a substantially lower wage than their Kiwi counterpart.
Clearly, the next Minister will have their hands full – and they only have three years in office. They have to prioritise. What one thing could an ambitious but realistic Minister do, which would have the greatest possible positive impact on the future of young New Zealanders, whilst also being achievable in the next three years, requiring little additional money and is within the political realities of NZ today?

I suggest that our ambitious Minister should focus on one thing: encouraging more high-quality teachers to teach in New Zealand’s poorest schools.

Why does this matter so much? The evidence is quite clear: improving the quality of teaching – and of the teachers doing the teaching – is the single biggest lever for improving student outcomes [2]. In particular, the effect of being taught by a particularly poor teacher is almost catastrophic for the student involved: the best evidence is that being taught by a teacher in the bottom 5% of effectiveness will reduce an average’s student performance by ~40%, compared to being taught by a teacher of average quality [3]. It is clearly utterly imperative to reduce the number of students being taught by such teachers.

There are (simplistically) two ways that this could be done [4]. The Ministry of Education could find ways to ban these failing teachers from the profession (In the jargon, this is ‘deselection’ [5]). But any Minister who followed this course would be in a political fight of the highest order: no teacher (and certainly no teaching union) would like to work under the threat of ‘deselection’. Moreover, whatever your opinions on firing poor quality teachers as a principle [6], it is a silly policy in New Zealand’s current situation: we struggle to recruit enough teachers to work in low-decile schools as it is [7]. Out on the front line, the ‘pool’ of available teachers is shallow – few head teachers in low-decile schools have the luxury of choosing between several maths teachers to fill their vacancies. Introducing the threat of later firing is not going to encourage many people to join the profession. Instead, we must compete these low-quality teachers out of the profession. We must increase the depth of the recruitment pool – increase the number of high-quality teachers – and hence provide headteachers with more genuine choices in recruitment [8].

As we do so, we must also change part of the incentive structure. At the moment, the pool of teachers is not evenly distributed across the system: high-decile schools find it easier to recruit for vacancies and, on average, attract higher quality teachers [9]. The reason for this is simple: teachers are paid according to a national pay scale. There are no financial rewards for choosing to work in a low-decile school. Indeed, there is an implicit financial reward to teaching in high decile schools. A decile 10 school has about 60% greater income than a decile 1 school, as a result of parental donations (‘school fees’) [10]. Although a high decile school can’t spend this on teacher pay directly [11], they can spend it on the things which make a teacher’s life more attractive: the better classroom facilities, the new swimming pool, the additional school trips. Compounding this, high decile schools have fewer children with the most extreme forms of educational need, and have the pick of high-quality school leaders and supportive parents. Although some teachers are motivated by the challenge of teaching in low-decile schools, many are not: given the imbalance in incentives, it isn’t really much surprise that many high-quality teachers choose not to work in the schools which need them most. Any attempt to improve the depth of the teacher recruitment pool must be closely tailored at ways to encourage more teachers to work in low-decile schools.

Enter Teach First.

Teach First New Zealand recruits teachers in a completely different way to existing training institutions. Modelled on similar programmes in other countries, its basic premise is that teaching is a craft that can be learnt (to some degree) ‘on the job’. Accordingly, they recruit graduates directly from university (without completing the normal year-long teacher certification) and, after an 8-week training course, find them work in low-decile schools. The scheme to date has been small, with only a handful of new teachers being trained each year, and has not been without its controversies: Teach First and the Ministry of Education got into legal difficulties when they failed to comply with various pieces of education and employment law which regulate how teachers are recruited. However, there has been relatively little controversy about the quality of the programme: an initial qualitative study conducted by NZCER and the University of Auckland found the quality of teachers produced by the programme was generally very good, which is in line with international experience where Teach First teachers tend to be as effective (although not more effective) than conventionally trained teachers.

One particular feature of the Teach First programme is worthy of our particular attention: it seems to attract teachers who want to teach in low-decile schools. Once they complete the 2-year training programme, Teach First alumni are under no obligation to continue teaching in low-decile schools. Indeed, with the brand-stamp of ‘Teach First’ on their CV, they would be in a good position to take teaching jobs in other schools. And yet they continue to teach in low-decile schools in significant numbers: about 50% of those who complete the Teach First programme remainas teachers in low-income schools aftertheir programme.

This is remarkable and could – if done at scale – make an incredible difference to low-income schools. Any programme that increases the size of the ‘pool’ of people willing to teach in these schools, despite all of the incentives which encourage them to teach in higher-decile schools, will increase the competition for a fixed number of teaching roles and will reduce the number of sub-standard teachers who are hired merely to fill a vacancy. Teach First to date has shown that there is a pool of socially minded potential teachers amongst our university graduates who are willing to ignore their rational interests (financial and non-financial) to teach in low-income schools.

The first order of business is to ensure we are fully recruiting in this market: we should step up our efforts to identify these potential teachers and recruit them into teaching. For all the difficulties that Teach First has as an organisation (including, most recently, the withdrawal of the University of Auckland as its academic partner), it seems to be the organisation that is best placed to find these teachers. Our new Minister of Education should sanction an expansion of Teach First. It should be allowed to recruit as many teachers as it can find – on the proviso that it manages its programme in a way that keeps the proportion of its alumni who teach in low-decile schools stable (at around 50%), whilst also maintaining its existing quality and performance standards [12]. Further, the Minister should work with Teach First to remove any bureaucratic barriers that are preventing them finding a new academic partner. A critic of Teach First will rightly point out that Teach First is expensive (it costs about 50% more to train a teacher through Teach First versus conventional training channels [13]). However, I suggest that this a price that – ultimately – we can afford to pay, in order to get more high-quality teachers in schools where they are most needed [14].

So far, so easy. The problem is that Teach First is tiny. Under any conceivable expansion plan, only a small sliver of New Zealand’s teachers would ever be taught by Teach First. This is probably right – the on-the-job method of learning that Teach First uses may not be right for every potential teacher. Instead, the Ministry of Education should learn from Teach First to improve its own recruitment efforts for recruiting teachers to low- income schools. Teach First recruits teachers in a fundamentally different way to traditional institutions: it offers a sense of ‘mission’ and purpose, as well as a structured set of mentors, coaches and peers who remain with the teacher as they begin their teaching career, offering advice and encouragement as the new teacher encounters the difficulties of being a new teacher. In an ideal world, the Ministry would launch its own recruitment strategy incorporating these elements which Teach First have shown to be successful, with the addition of a financial incentive, possibly a student loan rebate, to increase the pool of people willing to teach in low decile schools. This would be a new role for the Ministry - it means they are actively intervening in the teacher labour market in a way that they haven’t done before – but it is an idea which would have real benefits for low-decile schools by increasing the size of the pool of teachers they can recruit from. Doing so would not be the most eye-catching initiative for a new Minister but it has the advantage that it might just have a significant difference, quickly, for the young New Zealanders who need it most.

Simon Johnson is a former Treasury official and management consultant. He now lives in India where he is working on teacher quality and retention. He writes in his personal capacity and has no party-political affiliation, nor any connection with Teach First NZ. [Editor's note: Simon is still very tall, and very British.]

[1] Gratton and Scott (2016) ‘The 100 Year Life’
[2] Hattie (2009) ‘Visible Learning’
[3] Hanushek (2009) ‘Teacher Deselection’
[4] A third alternative would be to improve the quality of the lowest performing teachers through in-service professional development. I am not discussing this here, not because I consider it unimportant (it clearly is vitally important) but because I am assuming that it has limited potential to help the least effective 5% of the teaching force.
[5] I am not clear why Hanushek invents this euphemism for ‘firing’ [Editor's note: if anything, 'deselection' sounds far worse, almost robotic in its clinical coldness].
[6] And, to be clear, I am in favour it, all other things equal (which they definitely are not).
[7] I really wish that the data exists for me to show this empirically. However, at the time of writing, my OIA request has not been successful and I cannot show this more than anecdotally.
[8] This assumes headteachers can distinguish between a high and low quality candidate at interview – a vital point which is worth much fuller discussion than I can spare here.
[9] See note 7
[10] For more, see our article from the last election cycle:
[11] This would be ‘bulk funding’.
[12] The MoE should insist on this being written into the contract with Teach First.
[13] See, for example,, conforming with the wider literature.
[14] That is not to say that we should write a blank cheque. For example, I would be very happy to introduce some form of bonding, where Teach First participants have to pay back a portion of their costs if they leave teaching (or low-decile teaching) before a certain number of years. My point is that a short-term additional cost in improving the size of the teaching profession will have long-run advantages if it is can remove low-skilled teachers from the classroom. 

No comments:

Post a Comment