Talented and gifted students require educational modifications to realize their full potential. Educators must design the best possible education for talented and gifted students with the most appropriate environment. A modified curriculum which addresses the differences in the talented and gifted (TAG) student needs to be developed and implemented to provide the best education possible. The purpose of this mixed methods study was to examine the perceptions, attitudes, and beliefs of gifted students in addition to their achievement and the program model (pull-out versus cluster) at the elementary level. Quantitative data was obtained from the Harter Self-Perception Profile for Children (Harter, 2012) and the Georgia Criterion Reference Competency Test. This study used qualitative pieces of interviews and focus groups to tie the quantitative and qualitative strands together. The findings from this study were numerous. The interviews’ over-arching themes were: a) TAG students enjoy working together; b) they enjoy challenging projects and see them as fun; and c) they do not like being singled out. The over-arching themes from the focus groups were: a) TAG students preferred to work in small groups; b) they liked learning together; c) liked having challenging activities; and d) when pulled out for resource they did not like missing classwork. There were statistically significant interactions between all the subscales of the Harter Self-Perception Profile for Children and the cluster and pull-out groups. Grouping had a statistically significant effect on the perception scores of particular grade levels and subscales. Data for the content areas showed differences for the two grouping options. The indications from this study led the researcher to see there is a relationship between academic achievement and perception. In this study the pull-out group had higher perception scores and higher achievement scores compared to the cluster group. However, one must remember that a TAG student is not necessarily gifted in all areas and it is difficult at best to generalize that all students in pull-out models would have higher achievement scores than TAG students in cluster models.