Using the CogAT 7 screening (one subtest from each cluster) is considered a valid use for gifted identification. Obviously, the more data one has, the higher the degree of confidence. But HMH & the authors do condone using the screener this way, based on its reliability data, and correlation with the complete battery, with the caveat that the catchment range should be about three times the cutoff range that would otherwise be used, if using the comprehensive battery. IOW, 90th %ile and above would be a reasonable screening cutoff, if the comprehensive cutoff would have been 97th %ile. If the comprehensive cutoff would have been 90th %ile, the screener cutoff should be 70th %ile.
The publishers/authors do not recommend releasing score breakdowns when using the screener, as there are no valid area scores.
http://www.hmhco.com/~/media/sites/home/hmh-assessments/assessments/cogat/pdf/cogat-cognitively-speaking-v7-aug-2011.pdf?la=en
(p.7-9)
Other than the testing conditions themselves, which appear to have been beyond the control of the proctors, there is nothing strikingly untoward in your account of the use of the test or the reporting of its results.
Whether the entire administration was valid is another question, but that would apply to every member of his class (and likely many other classes).
Among the reasons he might not have reached the cut score this time:
1. Classroom performance and teacher ratings are among the weakest predictors of GT-level cognition.
2. The primary level of CogAT 7 is entirely nonverbal, which means that students who present as higher-functioning due to advanced reading and writing skills, or oral performance skills (often overlapping with soft skills, such as assertion/confidence, or fluency in oral English) will receive no advantage over peers who have the same level of abstract cognition, but have not yet attained those levels of specific academic skills.
3. Young children have differing levels of familiarity and comfort with computer-administered assessments, resulting in a notable percentage underperforming their "true" abilities in their first few encounters with formal testing.
4. The percentile cutoff may have been a local percentile, rather than a national percentile. If your child attends school in a high-performing community, it is possible that a child who would fall above the cutoff nationally falls below it in this specific community.
None of this is incompatible with you and your child's teacher's experience of him as a capable and motivated learner. Nor does it invalidate his gifts, of whatever nature and extent, or his right to have them nurtured and developed in the context of his development as a whole person.
