Elgazette Logo newtrans  The magazine for English language teaching and English medium education

A Compass To Navigate Education

Claudia Civinini presents an overview of a substantial body of research which analyses the evidence behind many educational practices and interventions

 The Education Endowment Foundation (EEF) is an independent grant-making charity founded by the Sutton Trust whose mission is to close the attainment gap for students from disadvantaged backgrounds. It aims to do so by collaborating with research institutions to test educational practices and interventions, and provide evidence about which of these work best – and at what cost. They then collect their results in the Early Years and Teaching and Learning (5–16-year-old) Toolkits ‘so that schools have the best possible evidence on which to base their own professional judgements’.

The Toolkits list practices and interventions such as ‘Teaching Assistants’ or ‘Block Scheduling’ (longer lessons but fewer of them), and for each of these they indicate the cost, strength of the supporting evidence and average impact measured in terms of months of learning they can add – or subtract – to the student’s education, from +8 to -4, taking average pupil progress over a year as a benchmark. In our bid to dedicate more space to evidence-based teaching, we believe the tools offered by EEF can make a good compass for teachers and policy-makers alike. On this page we offer an overview of what works really well with very young learners and students aged 5–16, and for the latter group we also show you which practices we thought would work wonders but actually don’t, which practices don’t work and which ones can be potentially damaging. Overall, the evidence points to something that we all know, but that curiously doesn’t get the same attention as the latest technological gadget: teachers and their expertise are the greatest asset in a school.

In our next issue we aim to provide readers with an overview of teaching tips connected to the most successful practices.

See the Early Year Toolkit here and the Teaching and Learning Toolkit here

Early years: what works well?


  • Self-regulation strategies: students’ ability to manage their own behaviour or learning. This is linked with successful learning, including pre-reading skills, mathematics and problem solving. It is a promising area although more research into its impact is needed.
  • Communication and language approaches: these emphasise the importance of spoken language and verbal interaction. Studies consistently show positive benefits for young children’s learning, including their spoken language skills, their expressive vocabulary and their early reading skills.
  • Earlier starting age: nursery or pre-school from the age of three. Although the impact seems to be high, this needs to be further investigated.
  • Parental engagement: a set of practices that ranges from encouraging parents to read with their children and supporting them with homework to providing counselling for families. This is closely linked to the child’s academic success.
  • Early literacy approaches: foster young children’s reading and writing skills through activities such as storytelling or group reading. An approach targeted at small groups and sustained overtime seems to bear the best results. Evidence is very sound.

Some surprises – approaches we thought would work wonders, and in fact they don’t, sometimes at a very high cost


  • Class size: size matters, at least for schools’ marketing departments. The impact this has on actual learning is not so straightforward. There is no improvement until the class is reduced to less than 15–20 students. Class size reduction may work better with younger learners. Another study (Rivkin Hanushek and Kain, 2005) proved that the effect of having a very effective teacher as opposed to an average teacher is the same as reducing class size by 10–13 students (see September 2015 Gazette).
  • Teaching assistants: although more research is needed, some studies say that TAs can have a positive impact on learning. The key seems to lie in the collaboration between the teacher and the assistants, who best support learning by working with individual students or small groups, whereas settings where assistants provide general support do not work as well. In some cases it has been observed that low-attaining pupils perform worse in classrooms where assistants are present.
  • Individualised instruction: the practice of providing different tasks for each learner and support at individual level. Overall evidence shows this approach does not work as intended, with some studies even reporting a negative impact. This may be because with individualised learning in a classroom setting, learners receive less direct teaching. Small groups could be more effective instead.

Teaching and learning toolkit 5–16. What works well, with sound evidence and low cost?


  • Feedback: according to evidence, feedback is crucial for good learning. However, it must be done correctly, otherwise it could potentially be negative: it needs to be accurate, specific and clear, and provide precise guidance on how to improve.
  • Meta Cognition and Self-Regulation: teaching students specific strategies to set goals, and monitor and evaluate their academic development. Strategies are usually more effective when taught in collaborative groups but the full extent of their impact can be difficult to achieve, as it requires students to take responsibility of their own learning.
  • Homework: yes, homework. But only for secondary school students. Benefits are higher when homework is used as a short and focused intervention, more modest, but still significant, when routinely set. At primary level, homework has a lower efficacy rate.
  • Peer Tutoring: effective for both tutor and tutee, peer tutoring works best when structured to ensure high quality of interaction. It should be used as a supplement and reinforcement of teaching, not as a substitute or as a way to introduce new material.
  • Reading comprehension strategies: these aim at improving the learner’s understanding of a text. Some examples: summarising, identifying key points, developing questioning strategies. The most successful strategies select reading activities in careful accordance with the learner’s abilities, in order to provide an ‘effective but not overwhelming challenge’.
  • Collaborative learning: small groups working together on a task that has been clearly assigned. Clearly is the key word – structured approaches, well-designed tasks and meaningful interaction between group members are necessary.
  • Phonics: an approach that teaches reading and writing by developing students’ phonemic awareness – helping them recognise the relationship between sounds and written patterns, or graphemes. It is mostly effective on younger learners (4–10) and teachers’ expertise in the technique is crucial.

Teaching and learning toolkit 5–16. Some things that don’t work


  • Block scheduling: an approach to timetabling whereby pupils have fewer but longer classes per day. It needs further investigation but evidence so far seems to show that this approach is unlikely to raise attainment by itself.
  • Performance pay (for teachers): the Teaching Toolkit points out that evidence does not support this approach, that money would be better spent investing in professional development for teachers, and that performance pay ‘may lead to a narrower focus on test performance and restrict other aspects of learning’. Hands up if this shocked you – however, it still needs to be further investigated.

Teaching and learning toolkit 5–16. Damaging practices


  • Repeating a year: evidence shows that repeating a year hinders the student’s chances of academic success. Students who repeat a year are unlikely to catch up with peers and are more likely to drop out of school. The Toolkit suggest trying intensive tuition or one to one support instead, which are also considerably cheaper.
  • Setting or streaming: based on the assumption that teaching will be more effective with a narrower range of attainment in class, setting or streaming works best with talented or gifted students, but leaves low attaining peers falling behind one to two months a year on average. It also has a negative effect on those students’ confidence.