In addition, the clusters are not equally weighted in the GCA on the WJ, which is why it cannot be handscored. The TA and VA have higher g-loadings than the CE cluster does (in that order). Since the GCA global composite is supposed to be an indicator of g-general intelligence, the clusters with higher g-loadings are differentially weighted into its calculation.
GGG, my original comments on preschool testing are still applicable. I wouldn't rule out re-testing in a few years, when the scores become more stable. The new WJIV is a significant improvement over the WJIII, in terms of its assessment of general intelligence, and its ceiling, and also has the capacity to calculate scores without speed and memory (like the GAI on the WISC). The WISC-V is also a technical improvement over the WISC-IV.
The social-emotional indicators should not be considered measures of trait (intrinsic qualities), but state (current conditions). That he appears to be experiencing psychic distress right now is highly significant, but certainly does not mean 1) that his long-term prognosis is poor, or 2) that sending him to kindergarten would be an unhealthy choice. There are services for social-emotional needs in kindergarten, as well, and it might be an avenue for accessing additional services. I would definitely prioritize looking for solutions to the anxiety and depression (aggression often is a manifestation of the anxiety and depression, especially in children), keeping in mind that poor academic instructional fit will contribute to those symptoms.
