Moving students between sets during the school year “may carry an attainment cost rather than deliver gains”, according to researchers for the National Institute of Teaching (NIoT).
The study did not find meaningful rises in attainment when students moved up or down a set during the year, after analysing data from one multi-academy trust covering 21 secondaries and more than 10,000 students.
The have come from the Teaching Improvement through Data and Evaluation project, a five-year initiative funded by the Nuffield Foundation that began in 2024 and aims to identify the characteristics of effective teachers in England.
In what the NIoT describes as the “most comprehensive analysis of within-year class moves in England to date”, researchers found that moving up a set during the year was not followed by better attainment in English, maths, history or geography.
Disruption of moving sets ‘rarely repays itself’
Science was the only one of five subjects analysed where improvements were seen in attainment after mid-year moves between sets.
Moreover, in each subject reductions were generally seen in the attainment of students moved down a set during the year, with the exception of geography.
The data indicates that the disruption associated with changing classes part-way through a school year “rarely repays itself”. It raises questions over whether this disruption outweighs the potential benefits of teaching students alongside peers with similar attainment levels.
Caution over findings
The team behind TIDE includes experts from organisations such as the Education Endowment Foundation, Evidence Based Education and UCL Institute of Education.
The NIoT stresses that the results shared through the annual TIDE report today “do not establish as fact that changing sets caused the differences in attainment and should not be read as a blanket recommendation to stop set moves; the results should be considered in the context of wider literature and schools’ practical realities”.
Calum Davey, executive director of research at the NIoT and principal investigator on the TIDE project, said: “These findings are an example of the kinds of the rigorous, data-driven insights we are generating through the TIDE project.
“While they need to be treated with caution - as this is an observational study from one trust and one academic year - they offer the sort of information that schools, trusts and the wider sector can benefit from to improve teaching quality and ensure that every child in England has access to the excellent education they deserve.”
He added: “The next step is to extend the study to more trusts and add a further year of data, to strengthen the statistical reliability of the findings and test whether these patterns hold in different school contexts.
“That will help us to continue building a stronger evidence base to inform decisions about how best to support pupils’ learning.”
GCSE outcomes a future priority
John Jerrim, professor of education and social statistics at UCL Institute of Education, who led the research on between-set moves, told Vlog: “Our hope is that we will be able to extend this analysis over time to closer to 100 schools to improve our confidence in the extent we can generalise our results.
“We also hope to be able to explore how moving classes between academic years is also related to academic progress, including eventually GCSE outcomes.”
Becky Francis, CEO of the Education Endowment Foundation, said: “While the causal evidence base on pupil grouping remains relatively limited, this new observational study offers insights from one trust into the potential implications of moving pupils between groups during the school year. Schools also need to consider the wider implications of grouping decisions, including access to higher-tier qualifications and pathways such as triple science.
“Our own recent study, which compared attainment grouping with mixed-attainment teaching in maths, highlighted the complexity of these decisions. We found some evidence that higher-attaining pupils benefitted from setted maths classes, while lower prior-attaining pupils made similar progress in both mixed-attainment and setted classes.”
Professor Francis added: “What is clear is that decisions about pupil grouping should be informed by schools’ specific circumstances and implemented carefully, with close attention to ensuring that all pupils are appropriately supported and challenged. Whatever approach schools adopt, the evidence remains clear that high-quality teaching remains the strongest lever we have for improving outcomes, alongside targeted support for pupils who need it most.”
The TIDE annual report also looks at other issues: analysis of data from one trust with more than 20 secondary schools and 17,000 students, for example, found no evidence that students with special educational needs and disabilities were systematically allocated less effective teachers.
TIDE draws from the Teacher Education Dataset (TED), which brings together data from seven multi-academy trusts: around 165,000 pupils, 8,500 teachers and 2.5 million assessment results, encompassing 91 primary, 124 secondary and seven all-through schools across every region of England.
The NIoT underlines that all data in the TED is pseudonymised and that researchers conduct analysis without access to raw data - in a similar way to studies carried out in the NHS - so that it can never be used to identify, rank or manage the performance of individual teachers.
The TIDE project is expected to come to a close in 2028.
To get essential briefings from Vlog magazine delivered directly to your inbox, sign up to our newsletters. And stay up to date with the latest education news via the Vlog magazine mobile app - available on and on
