This paper deals with the first study on the construct validity of the GRS 2 Rating Scale for Teachers in Spain, on a sample of 348 subjects. A confirmatory factorial study, CFA, was carried out using robust least squares estimation procedures (WLSMV). A sixfactor structure has been selected that explains 66% of the variance. The fit indices of the model are satisfactory (CFI, .93; TLI, .93). Convergent validity (AVE, .66) and composite reliability (between .87 and .96) are adequate. This scale is the first validated in Spain and provides a relevant measure for the identification of high ability students. [ABSTRACT FROM AUTHOR]