Discrepancy Definitions of Specific Learning Difficulties

Abstract
The necessity to apply an ability‐achievement discrepancy model to the identification of specific learning difficulties has often been justified on the grounds that it is the best available and therefore constitutes defensible practice. This article re‐examines that contention in the light of findings from recent psychological research. The article starts with a comparison of the criteria being implemented in USA and in the UK. It then considers conceptual, statistical, empirical and ethical issues in relation to IQ‐achievement discrepancy definitions. Other discrepancy definitions are also discussed. Finally, an alternative is proposed.