The Highly Gifted: definitions and methodological implications

Abstract
This paper is concerned with the methodological implications of various definitions of the term ‘highly gifted’. Three definitions of ‘highly gifted’ are presented. It is shown that the definition of the term ‘highly gifted’ has considerable methodological implications. These implications concern both choice of method for data analysis and results of analysis. Conditions for proper application of standard statistical methods including the t‐test for mean difference, the F‐test for regression slope differences, MANOVA, and discriminant analysis are reviewed. It is also shown that, under circumstances that are far from unrealistic, there are no methods with known properties available for data analysis.