students already in a GAT program) with GAT students from a less highly selected population (e.g.
a GAT child may have a high academic self-concept, but only average physical and social self-concepts).
In both studies: (a) GAT students attending a GAT class were matched to students of equal ability who were attending mixed-ability classes; (b) students in the GAT setting experienced significant declines in three domains of academic self-concept over time and in relation to matched comparison students but there were no significant differences in four non-academic self-concept domains; and (c) this general pattern of results was reasonably consistent across gender, age, and initial ability.
