Summary. A sample of 105 adolescents who had abused solvents and an equal number of controls matched for school, school year and sex were identified by means of a two‐stage screening procedure involving (a) administration of a self‐completed questionnaire to approximately 5,000 third, fourth and fifth year pupils in 16 secondary schools, and (b) individually conducted interviews. These two groups were followed up at the end of their final year of compulsory education when details of their fifth year public examination entries and results were obtained from their schools. The examination attainments of the solvent abusers were generally lower than those of the controls. They were more likely to leave school at Easter of their fifth year and a much higher proportion of them were not entered for any examinations. Those who were entered for exams were less likely to pass at O level and, regardless of the type of exam, much less likely to obtain higher examination grades. Poor examination attainments among the solvent abusers were associated with adverse domestic circumstances, frequent self‐reported behaviour problems and low scores on educational tests administered at the time of transfer from primary to secondary school. After controlling for these factors, the difference between groups in examination attainments was no longer statistically significant. 1990 The British Psychological Society