CEP LSE RSS Contact Us YouTube Facebook Twitter

Recent work on Schools

What works? The effectiveness of government policies


Academies and school autonomy
The programme has a series of work on the effect of the Academies programme and school autonomy more generally. Andrew Eyles and Stephen Machin evaluated the impact of the original New Labour academies on student intake and achievement. The analysis shows that this first round of academy conversions that took place in the 2000s generated significant improvements in the quality of pupil intake and in pupil performance. Analysis of mechanisms points to changes in head teachers and management structure as key factors underpinning these improvements in pupil outcomes. This paper is forthcoming in the Journal of the European Economic Association. Stephen Machin and Matteo Sandi have recently explored whether the improvement of academy schools might have been partly generated by higher exclusion rates. They rule this out as a mechanism.

Andrew Eyles, Claudia Hupkau and Stephen Machin have also evaluated the longer term impact of these academies. They show that that academy conversion also had a causal effect on the probability of degree completion at university. The paper is published in Labour Economics.

 However, these results should not be taken to infer that the subsequent wave of academy conversion (which was much larger) had similar effects. The characteristics of schools that became academies in the context of the post-2010 policy were very different. This is documented by Andrew Eyles, Stephen Machin and Olmo Silva in another discussion paper, recently published by Fiscal Studies.

A team of researchers at CEP have worked with the Education Policy Institute to evaluate secondary schools that became academies since 2010. The results are published in the EPI Report, The impact of academies on educational outcomes. The main conclusion that academies have not provided a panacea to school improvement. With the exception of outstanding convertor academies, there is no visible, positive impact on outcomes amongst any other type of academy (both sponsored and convertors).

The situation is no better amongst primary schools, which first became academies in 2010. Andrew Eyles, Stephen Machin and Sandra McNally show that there was no effect whatsoever of academy conversion on pupil performance within primary school. The discussion paper is now published in the Journal of Public Economics.

Of course, the signal of ‘academy conversion’ might mean something for parents. This is considered for a sample of schools in Birmingham by Marco Bertoni, Stephen Gibbons and Olmo Silva. They find that parents rank converted schools more highly than other schools. The patterns suggest that families combine the signal of conversation with prior information on quality, popularity and proximity as a heuristic for assessing a school’s expected future performance.

The ongoing work of the programme in this area will consider questions such as: How effective are school networks for raising attainment? How does choice and school competition influence pupils’ outcomes?

Synthetic phonics

What helps improve children's reading? Again our Learning To Read project brings the data to this hotly contested question. In our Learning to Read project Machin, McNally, Viarengo study a national change in policy and practice in England that refocused the teaching of reading around “synthetic phonics”. This was a low cost intervention that targeted the pedagogy of existing teachers. We evaluate the pilot and first phase of the national rollout (i.e during and after the Rose Review of 2005/06). While strong initial effects tend to fade out on average, they persist for those with children with a higher initial propensity to struggle with reading. As a result, this programme helped narrow the gap between disadvantaged pupils and other groups. A revised version of the Discussion Paper is forthcoming at the American Economic Journal: Economic Policy.

School resources

What is the link between school resources and pupil outcomes? The programme has a number of publications on this topic, including the research paper by Steve Gibbons, Sandra McNally and Martina Viarengo which shows a strong positive impact of school expenditure, especially on improving the performance of disadvantaged students. See: Does Additional Spending Help Urban Schools? An Evaluation Using Boundary Discontinuities; or the CentrePiece Article for a summary.

Steve Gibbons and Sandra McNally have reviewed the literature on the causal effects of school resources on pupil outcomes. Overall, research suggests that spending matters and that spreading investments evenly across the various stages of education is preferable to interventions at any one stage. For example they find that whilst early age investments potentially offer higher returns, their benefits erode during later phases of childhood unless they are topped up with subsequent investments. See: The Effects of Resources Across School Phases: A Summary of Recent Evidence; or the CentrePiece article for a summary.

Peer effects and student outcomes

Student rank in the classroom

How much does a student's rank compared to his or her classmates matter? Richard Murphy and Felix Weinhardt find that a student's higher ranking compared to his or her peers leads to continued higher performance later on, and that this appears to work through improved confidence. This challenges conventional wisdom which says that it is always best to place children with higher-performing peers. See: The Importance of Rank Position ; or the CentrePiece article for a summary.

Non-native English speakers

Does school mixing of non-native English speakers and native pupils depress the outcomes of the latter? This has become a big policy question as the number of non-native speakers of English in primary schools in England has increased substantially in recent years. Using two different research strategies, Charlotte Geay, Sandra McNally and Shqiponja Telhaj find no negative effect on native pupil outcomes. See: Non-Native Speakers Of English In The Classroom: What Are The Effects On Pupil Performance? ; or the CentrePiece article for a summary.

Influence of neighbours

How does a student's neighbourhood affect their educational outcomes? Steve Gibbons, Olmo Silva and Felix Weinhardt find that the characteristics of neighbourhood peers make no difference at all to how well children perform at school: their test score progression in secondary education is unrelated to changes in the social composition of the place where they live. Neighbours seem to have some impact on children's attitude towards school and their propensity for anti-social behaviour, but the effects are very small and weak. For a summary of this work, see the CentrePiece article or the full article: Everybody Needs Good Neighbours? Evidence from Students' Outcomes in England which was published in the Economic Journal .

Teacher status

Evaluation of Teachers’ Pay Reform

The coalition government implemented wide ranging reforms to the teacher labour market in England. This culminated in the removal of teacher pay scales and pay portability (where teachers were guaranteed their previous salaries when they changed jobs) and the introduction of mandatory performance related pay. Richard Murphy from CEP (with co-authors Simon Burgess and Ellen Greaves) is investigating how schools reacted to this new-found freedom. Findings show that schools mostly kept to the old pay scales with teachers receiving pay increases according to their years of experience, or no increase at all.

 Teaching to the Test: The Long Run Impacts of Standardised Testing on Student Outcomes

A growing concern among parents, teachers and policy makers is that high-stakes testing will incentivize providers of education to focus rigidly on the materials that are tested - “teaching to the test” - at the expense of providing students with a more holistic educational experience. Richard Murphy and Gill Wyness from CEP (with co-author Joe Regan-Stansfield) are working on a project that exploits the period of industrial action in the UK in 2009/10 where one third of primary schools stated that they would not teach towards or administer the national exams of students aged 11.

Teacher Turnover: Does it Matter for Pupil Achievement?

Recent research has established that teachers matter for student achievements, albeit because of dimensions of ‘teacher quality’ that are largely unexplained. A less closely investigated issue is whether teacher turnover directly harms student academic achievement. Steve Gibbons, Shqiponja Telhai and Vincenzo Scrutinio investigate this issue for 16 year olds in England using the National Pupil Database. They find that a higher teacher entry rate reduces students’ test scores, albeit by small amount.

Valuing schools through the housing market

CEP research has consistently found that housing valuations are significantly higher in places where measured school quality is higher, implying a strong parental willingness to pay to get their children educated in better performing schools by moving into good school catchment areas. Steve Machin finds similar results in his critical appraisal of the empirical literature in a number of other countries, using a variety of identification strategies and at different parts of the education sequence that children follow. See: Houses and Schools: Valuation of School Quality through the Housing Market. Steve Gibbons summarises CEP work in this area in his 'Big Ideas' article.

Randomised Control Trials in Schools

Information and Education Decisions (Martin McGuigan, Sandra McNally and Gill Wyness)

How can information affect student decisions on higher education? Against the background of increased university tuition fees in the UK, Martin McGuigan, Sandra McNally and Gill Wyness look at whether information about the costs and benefits of future education has an impact on the knowledge and aspirations of school students. We investigate students’ receptiveness to an information campaign about the costs and benefits of pursuing post-compulsory education. Our results show that students with higher expected net benefits from accessing information are more likely to avail themselves of the opportunity presented by our experiment. Their intention to stay on in post-16 education is strongly affected by the experiment, though not their intention to apply to university. This paper is now published in the Journal of Human Capital. Please also visit the project website.

Sandra McNally has written a review of papers using RCTs to assess the impact of providing information on students’ educational decisions. This is available as a IZA World of Labor paper: How important is career information and advice?

'Lesson Study': Evaluation of teacher training on student performance (Richard Murphy, Felix Weinhardt, Gill Wyness)

It is well recognised that the quality of teachers has a large impact on the academic attainment of pupils. Moreover, it has been found that it is during the first five years of teaching that teachers improve most in effectiveness, but that they remain 'stuck' at a certain level. Effectiveness varies dramatically across individuals. Can effectiveness be improved by policy intervention? In a project carried out with the Education Endowment Foundation we use a Randomised Control Trial on 180 schools in England to determine whether teachers trained in observation and feedback who then observe each other during class have more impact on student outcomes in the years during the training and in subsequent years than those in 'untreated' schools. The findings show that students of observed teachers perform no better on national exams a year after the program ended.

Evaluation of Small Group Support (Teaching Assistants) for Literacy (Sandra McNally, Jenifer Ruiz-Valenzuela)

As well as investigating which methods of teaching reading are more effective we are also looking at whether the way teaching of reading is delivered has an effect. We have undertaken an evaluation of the effects of two interventions to help young children learn to read. Both interventions involve the use of a teaching assistant to work with small groups of children but one is ICT-based and the other non-ICT based. We find evidence that both interventions have a short-term impact on children’s reading scores, although the effect is bigger for the Non-ICT intervention. The effects of both interventions are found to be much bigger for children of lower ability. In further work, we are exploring the lon ger term impact of this programme. McNally and Ruiz-Valenzuela are working with the implementers of the intervention (in Coventry University) and the process evaluation team (in NIESR).

ACTive citizenship to enhance pupils’ social and civic competencies (Sandra McNally, Jenifer Ruiz-Valenzuela)

This is a European (Erasmus) project which we are working on with partners in the UK as well as in France, Greece and Spain. The project aims to find out if and how active citizenship projects lead to increased democratic engagement, improved civic skills, and development of common values in young people. It is built on the premise that education is vital in ensuring that young people are prepared for contemporary life and embrace fundamental values of democracy, the rule of law, individual liberty, and mutual respect and tolerance of those of different faiths and beliefs. Active citizenship projects are a way for students to derive personal meaning from classroom activities and take responsibility for their outcomes. They also present the opportunity for pupils to experience democratic decision making first-hand. The intervention is being evaluated in all four countries as a Randomised Control Trial. The impact evaluation will start in the academic year 2017/18.