Thanks.
So in terms of differences, the only subtest that both is more likely to have declined because of examiner differences, and actually declined in any notable way is Similarities. This doesn't definitively say that it was examiner error, though, since we already know there is an alternate explanation (ADHD). The other ones that decreased enough to make me wonder are pretty cut and dried in their scoring, with limited space for queries or clinical judgement (MR and Cd)--but do have in common readily identifiable vulnerabilities to unmanaged ADHD, in the case of MR because it is multiple choice, and can be impacted by impulsivity, and in the case of Cd because weaknesses in sustained attention can impact speed and accuracy.
The resulting VECI, however, is nearly identical to the previous VCI. (Remember that index scores are more reliable than subtest scores.) The consistency across most VECI subtests, including the ones that weren't retests, suggests that this is probably generally in the ballpark of his real performance.
The VSI subtests went the other direction, where BD rose quite a bit, with the similar score in VP suggesting that this might be a legitimate representation of visual spatial thinking (note that VP was not a retest, so there's no question about validity for it). There are some students with dysregulated attention for whom BD's timed aspect is a particular vulnerability, which may be one explanation for the lower performance the first time around. Another that should be noted is score inflation from retest effects, since the novelty of the designs is a significant aspect of the test. In this case, I tend to think that the similar score on VP is more confirmatory.
FRI is where the most dramatic fall occurred. I'll note that, taken in isolation, this EFI looks like the MR score is not a fluke, since three of the four EFI subtests are in the average range, while one is an outlier, in the extremely high range. The Ar score doesn't appear to be lower because of working memory artifacts, since both WMI subtests are in the extremely high range (and are consistent with the previous test). Notably, PC is also a variation of multiple choice. (To be fair, FW is too, and was strong both times.) So it's not entirely clear why the whole FRI is so much lower the second time, especially where these are not tasks with a lot of opportunity for examiner error. (This would be where any clinical commentary from either set of examiners would be useful.) One can speculate that fluctuating attention, fatigue or impulsivity may have been factors, given the diagnosis.
Based on our available data, I think it would be fair to say that the VCI, VSI and WMI at the time were probably in the MG range, with PSI in the average range. One might argue that VCI/VECI should have been closer to the HG range. FRI officially was at the MG/HG border. Playing around with his highest subtest scores for each subtest from each administration actually just slides him around from the lower to the upper end of the HG range.
In sum, this looks very much like a mostly MG learner with slightly higher abstraction (both verbal and nonverbal), drifting toward HG, with age-appropriate processing speed. Not, of course, taking into account impacts of ADHD. As you already know, this is a profile with many strengths cognitively; I don't see huge spreads between (or even really within) ability areas, with the exception of the FRI results, which we've discussed already. Is it possible there was more cognition not captured anywhere in these two assessments? Of course. If one were reassessing, it would be informative mainly in documenting improvements in the management of his ADHD, or if either of the two usual reasons for evaluation were pertinent (1, to elucidate/respond to a presenting problem impacting IRL function; 2, for access to necessary resources).
If you do decide to pursue updated evaluation, and he is receiving disability-based services from the district, have a thoughtful conversation with the evaluators he would have in-district for a 504- or IEP-based triennial before exploring private evals, and make sure to maintain clear communication, so that all assessment data is valid, and duplicative testing is minimized. I can't speak for others, of course, but I always try to work collaboratively with other evaluators; it's in the best interest of the child.
