Evidence for review of mathematics teaching: Improving mathematics in key stages two and three

This document presents a review of evidence commissioned by the Education Endowment Foundation to inform the guidance document Improving Mathematics in Key Stages Two and Three (Education Endowment Foundation, 2017). The review draws on a substantial parallel study by the same research team, funded by the Nuffield Foundation, which focuses on the problems faced by low attaining Key Stage three students in developing their maths understanding, and the effectiveness of teaching approaches in overcoming these difficulties. This project, Low attainment in mathematics: an investigation focusing on Year 9 pupils includes a systematic review of the evidence relating to teaching of low-attaining secondary students, which the current report builds upon in the wider context of teaching maths in Key Stages two and three. The Education Endowment Foundation and the Nuffield Foundation are both committed to finding ways of synthesising high quality research about effective teaching and learning, and providing this to practitioners in accessible forms. There have been a number of recent narrative and systematic reviews of mathematics education examining how students learn and the implications for teaching (e.g., Anthony & Walshaw, 2009; Conway, 2005; Kilpatrick et al., 2001; Nunes et al., 2010). Although this review builds on these studies, this review has a different purpose and takes a different methodological approach to reviewing and synthesising the literature. The purpose of the review is to synthesise the best available international evidence regarding teaching mathematics to children between the ages of 9 and 14 and to address the question: what is the evidence regarding the effectiveness of different strategies for teaching mathematics? In addition to this broad research question, we were asked to address a set of more detailed topics developed by a group of teachers and related to aspects of pupil learning, pedagogy, the use of resources, the teaching of specific mathematical content, and pupil attitudes and motivation. Using these topics, we derived the 24 research questions that we address in this review. Our aim was to focus primarily on robust, causal evidence of impact, using experimental and quasi-experimental designs. However, there are a very large number of experimental studies relevant to this research question. Hence, rather than identifying and synthesising all these primary studies, we focused instead on working with existing meta-analyses and systematic reviews. This approach has the advantage that we can draw on the findings of a very extensive set of original studies that have already been screened for research quality and undergone some synthesis. Using a systematic literature search strategy, we identified 66 relevant metaanalyses, which synthesise the findings of more than 3000 original studies. However, whilst this corpus of literature is very extensive, there were nevertheless significant gaps. For example, the evidence concerning the teaching of specific mathematical content and topics was limited. In order to address gaps in the meta-analytic literature, we supplemented our main dataset with 22 systematic reviews identified through the same systematic search strategy.