The abilities of 17 emotionally disturbed children enrolled in a day psychiatric treatment center were studied using the Goldman-Fristoe-Woodcock Test of Auditory Discrimination, replicating an earlier study (Blager, J. Audit. Res., 1978, 18, 221-227) and extending it to other factors possibly influencing auditory discrimination abilities. It was confirmed that, as a group, such children perform less well than the GFW normative sample both in quiet and in noise conditions. Despite normal HTLs, their performance more closely resembles that of the GFW "poor discriminator" normative group. Age, sex, socioeconomic status, severity of disturbance, and intelligence were shown not to be significantly associated with test performance. Although there were 7 children whose rank orders in the group of 17 were better when the GFW was administered in noise than in quiet, and might have been thought to form a subgroup of those less impaired by noise than usual, this hypothesis was not borne out (p = .07) as in the previous study it had been. Several questions regarding the GFW were raised and suggestions for future research discussed.