Performance on Dutch adaptations of Rey's AVLT was examined in a sample of 225 6- to 12-year-old Dutch school children, selected to be representative of the general population. With an interval of 3 months, they were tested twice, using two out of three test forms which proved to be parallel. No test-retest practice effects were apparent. Age had a considerable impact on test performance. Measured imperfectly, socioeconomic background and intellectual level had some additional influence, as did the examiner who administered the tests. Boys made more errors than girls. Normative data corresponded well with those collected in a smaller Australian sample, suggesting usefulness outside The Netherlands. With a parallel test-retest reliability of .70, which was reduced to .59 when the influence of age was taken into account the most reliable AVLT measure was the total number of words correctly recalled over the five learning trials. On the basis of reliability data implications for the clinical use of the AVLT are discussed. (C) 1999 National Academy of Neuropsychology. Published by Elsevier Science Ltd.