Mixed-achievement math classrooms are a net negative

Mixed-achievement math groupings reduce progress for high-achievers, have no impact on outcomes for their marginalized or lower-achieving peers, and negatively affect students’ self-confidence.

It’s much disputed in England whether grouping students by achievement level or mixing them together is a more effective strategy for teaching in secondary schools. This practice resembles achievement grouping in the U.S., particularly when done in specific subjects like math.

In a new study, researchers from several UK universities evaluated these two student grouping practices in mathematics. The sample included year 7 and year 8 (aged 11–13) students in non-selective state secondary schools in England, along with subgroup analyses for marginalized students and those with low prior achievement in mathematics. The evaluators recruited two groups of schools based on their usual approaches to grouping students for math lessons. Using a quasi-experimental design, the two groups were matched to ensure they were as similar as possible across key characteristics before comparing outcomes.

The researchers had three primary objectives: examining the effects of grouping practices on mathematics achievement, students’ self-confidence, and classroom teaching practices.

Several findings emerged. Most important is the statistically significant finding that high-achieving students in schools using heterogeneous grouping made, on average, two months less progress in mathematics compared to students in schools that grouped students by current achievement levels. Overall, students taught in mixed groups made an estimated one month less progress in mathematics than those in more homogenous groups. (This difference, however, was not statistically significant.)

Nor did analysts find consequential differences in progress associated with grouping practices for socio-economically disadvantaged students or those with low prior achievement. Some evidence suggests a narrowing of the achievement gap in mathematics between low- and high-achieving students in the heterogeneous-grouping schools, but this appeared to be driven largely by lesser progress among higher achievers rather than greater progress among low achievers. In other words, the gap was reduced by lowering the ceiling, not raising the floor.

Mixed grouping in mathematics also had negative effects on students’ self-confidence in mathematics, particularly for socio-economically disadvantaged students and low prior achievers. Within the case study schools, mixed-achievement mathematics lessons generally had a slower pace than that seen in higher-achieving groupings. And the researchers found that enrichment opportunities for higher achievers were limited and were often disconnected from the lesson content and rarely incorporated into class discussions.

Overall, the findings suggest that mixed-achievement grouping in secondary mathematics, where students of different prior achievement levels learn together, reduces progress for high-achieving students, has no impact on outcomes for their marginalized or lower-achieving peers, and negatively affects students’ self-confidence in the subject.

SOURCE: Jeremy Hodgen, Becky Taylor, Antonina Tereshchenko, Jake Anders, Laurie Jacques, Maria Cockerill, Rosa Kwok, David Sirl, Hester Burn, Nicola Bretscher, “The Student Grouping Study,” UCL Institute of Education (April 2026).