BERA Review of Education, 7(3) pp 570-592

Debates on how best to educate young children have been raging over the last 100 years—more often fuelled by ideological preferences rather than empirical evidence. To some extent this is hardly surprising given the difficulty of examining pupil progress in a systematic and comparative way. However, the introduction of a new child‐centred curriculum in Wales provides the opportunity to undertake just such an examination. The Foundation Phase curriculum, introduced in 2008, is designed to provide all 3‐ to 7‐year‐olds with a developmental, experiential, play‐based approach to learning. Evidence from a major 3‐year evaluation of this intervention finds that, overall, pupil progress and well‐being is fostered in those settings where the principles of the Foundation Phase have been most closely followed. However, the evidence also suggests that even within these contexts, progress is uneven and that some kinds of children seem to gain more from this approach than others. The ‘losers’ appear to be boys and those living in poverty. Drawing on the theories of Basil Bernstein, the paper explores why this may be the case and examines the relative significance of teacher dispositions, teacher–learner dynamics and the availability of resources. The paper concludes by arguing that these issues will need to be addressed if the benefits of child‐centred approaches are to benefit all.