Group composition and National Curriculum assessment at seven years

Summary

Children in primary schools spend much of their time working in groups. Groups can differ in terms of whether they are composed of friends, children of similar ability or children assigned on a random basis. Instructions for the first National Curriculum assessments conducted in 1991, at the end of Key Stage 1, suggested that Standard Assessment Tasks (SATs) were often to be conducted in small groups, but there was no guidance given about the control of group composition. This study was designed to find out whether the composition of groups affected performance, on‐task behaviour and group involvement on SATs. The study was based on maths tasks given to seven‐year‐old children in one class in an infant school in London. Results showed that children placed randomly in groups performed less well than children in friendship and ability groups on all three criteria. Children in friendship groups had higher performance scores and were more involved in the group than children in random and ability groups. The study indicates that the social setting within which assessment takes place is important, and its possible effects on performance, task involvement and social relationships should not be neglected.