ABSTRACTThe impact of illustrations on the reading performance of children with reading difficulties has been examined. Nineteen children from the first year of three comprehensive schools each read illustrated and unillustrated fictional passages at three levels of difficulty. The reading performance was assessed using miscue analysis, Goodman and Burke (1972) and a two‐way analysis of variance was applied to analyse the results.Reading performance was assessed in the area of: total errors, time taken to read the passages, comprehension, grammatical relationships and errors excluding omissions and insertions. The results suggested that, in the majority of these aspects of the reading process, there was a statistically significant interaction between illustration and difficulty levels. At the lower level of difficulty, reading performance in the illustrated conditions was superior to that in the unillustrated conditions. At the higher level of difficulty the reverse trend was apparen