Findings of two studies which examined the identification of prekindergarteners with potential learning problems are presented. The first was a pilot study, in which a short screening device and a parent questionnaire were administered to a large, randomly-selected sample. In the second study, these instruments were validated against extensive diagnostic testing on a smaller sample selected to highlight children with high and low potential for school learning problems. The results indicated that (1) the screening device identified the learning disabled children, as defined by the diagnostic testing, with over 90% accuracy, (2) teacher ratings of the children were inaccurate as compared to the results of the testing, and (3) the parents were not a reliable source of information about their children. Recommendations are made for the evaluation of five different intervention models to determine which types of learning deficits are amenable to remediation. (Author/MV)