This page covers the second (effectiveness) trial of Improving Working Memory and Arithmetic. The effectiveness trial is aimed at testing a scalable model under everyday conditions in a larger number of schools. To read about the first (efficacy) trial of Improving Working Memory – testing whether it could work in schools under best possible conditions – click here.
The Working Memory + Arithmetic (WM+A) intervention was designed by the University of Oxford to improve the numeracy skills of 7 – 8 year-old children who show low attainment in arithmetic at the end of KS1. The intervention seeks to improve pupils’ working memory, as well as the way they organise information about numbers and operations (addition and subtraction) in their long-term memory. It builds on evidence from cognitive science, which suggests that numeracy difficulties may be related to poor working memory capacity
WM+A is delivered through one-hour one-to-one sessions by Teaching Assistants (TAs) and includes the following components: a working memory programme, delivered over five sessions, and a number and arithmetic operations component, delivered over five subsequent sessions. During each session, the TA works for half an hour with one child while a second child plays specially designed online games in the same room to practice the strategies. After half an hour, the children swap activities. The sessions should take place once a week for ten weeks
TAs receive two days of training in the approaches – one day focussed on the working memory aspect and a second on the arithmetic elements.
A previous EEF-funded efficacy trial of the approach showed that the intervention had a positive impact on maths outcomes, with children in the intervention schools making the equivalent of three additional months’ progress in maths.
Considering the positive results, it was decided to test the programme in a larger number of schools through an effectiveness trial and assess the impact on pupils eligible for Free Schools Meal (FSM). This effectiveness trial will test a more scalable version of the intervention, which includes a revised ‘train the trainer model’, assessing its potential for further roll out.
The trial, led by RAND, will be a two-arm randomised control trial comparing outcomes for Year 3 pupils receiving the intervention with pupils that will receive business-as-usual maths teaching as the control group
The trial will look at pupils’ maths outcomes at the end of the first school year using a combination of two standardised maths measures – Hodder’s Maths Assessment for Learning and Teaching (MaLT) and the sub-test Arithmetic of the British Ability Scales (BAS, which was the assessment used during the efficacy trial).
The evaluation report will be published in Autumn 2023.