The authors present a Rasch scaled instrument for the standardized assessment of children's learning potential. The test was constructed on the basis of theoretical and empirical considerations regarding the evaluation of general cognitive abilities in children with learning difficulties. It can be administered both individually and in groups. The results of the first two studies show that the pretest—training—posttest procedure allows discriminating between learners with and without learning disorders. Significant training effects are found in the experimental group and the posttest scores of these children improve the prediction of school related criteria. The test appears less related to non-cognitive factors that may influence performance than a classical intelligence testing procedure.