The Children's Auditory Discrimination Inventory (CADI) was used in the longitudinal study to evaluate the child's ability to identify, between two pictures that have been given oral word equivalents, the picture that represents the orally presented stimulus word. The task, which takes 10 minutes, uses 38 cards with two pictures on each and 38 pairs of stimulus words. One picture in each pair is a real picture representing a familiar word, and the second picture is a nonsense design paired with a nonsense word. The real and nonsense words are randomly located on the right or left side of the cards, with the real word always presented first. The tester presents each pair of pictures, naming each one as she points to it; the child is then asked to point the picture named by the tester. An item may be repeated only once. The score is the total number correct (range = 0 to 38). The results of administering the inventory in the present study showed that although 98% of the sample attempted to respond to each item, those items involving nonsense words had a smaller proportion passing each item than did the items involving read words. Tables present means, standard deviations, and range of responses for the total sample by three-month age breakdowns and by sex. The data indicate that children's auditory discrimination increases with age for children between 42 and 69 months, and sex differences appear slightly in favor of girls. Analyses of variance (sex x age x SES) showed SES as measured by mother's education to be significant when scores were summed across years. (For related documents, see TM 003 174-176, 178-198.) (DB)