The aim of this investigation was to follow up a sample of exceptionally short but medically healthy children, and a normal comparison group, previously studied at 4 years of age. They lived in an inner‐city area which was, on objective criteria, seriously disadvantaged in socioeconomic terms. When first seen at 4 years, cases were significantly impaired in cognitive abilities relative to comparisons, although firstborns were much less severely affected. Of the original 46 cases, 45 were assessed again at 11 years. Most continued to live in the same geographical area. Case children remained exceptionally short, even when parental stature was taken into account, although a degree of catch‐up had occurred. One third had special educational needs, and a similar proportion had been referred for speech therapy. Verbal and nonverbal cognitive skills of both case and comparison children had, on the whole, changed little and group differences persisted. In conclusion, short normal children from socioeconomically disadvantaged backgrounds are at high risk of educational failure at elementary school.