Summary.Verbal reasoning and semantic divergent thinking scores together with assessments of conventional and creative English attainment were gained for 331 10‐year‐old children. A factor analysis demonstrated a separation of the ability and attainment measures. A zonal analysis showed that highly divergent children gained better scores on the creative attainment criterion than did highly convergent children. The threshold hypothesis was tested by relating convergent and divergent thinking at different levels of VRQ, and was rejected. Evidence concerning construct and academic validity is discus