0 members (),
86
guests, and
12
robots. |
Key:
Admin,
Global Mod,
Mod
|
|
S |
M |
T |
W |
T |
F |
S |
|
|
1
|
2
|
3
|
4
|
5
|
6
|
7
|
8
|
9
|
10
|
11
|
12
|
13
|
14
|
15
|
16
|
17
|
18
|
19
|
20
|
21
|
22
|
23
|
24
|
25
|
26
|
27
|
28
|
29
|
30
|
31
|
|
|
|
|
Joined: Mar 2021
Posts: 1
New Member
|
OP
New Member
Joined: Mar 2021
Posts: 1 |
Our newly identified exceptionally (or profoundly? I am not sure) gifted, 2E 6 year old son was recently administered the WISC-V (in addition to several other elements in a full neuropsychological exam.)
I am working now to gain a better understanding of his scores and their significance. His GAI is 168, but the neuropsychologist explained that this is not a true representation of our child's full intellectual capabilities; he maxed out the extended norms of the Matrix Reasoning subtest, so we don't know his actual "limit." A GAI was calculated due to what's indicated on his report as a "clinically significant >23 standard score difference" between two indexes; his FRI is a SS of 183, and his WMI is a SS of 130. That's a whopping 53 points between the two! I gather that a discrepancy like this must mean *something* given the notation that a >23 pt difference is "clinically significant," but can somebody expand on what this means?
He was indicated as being mildy-to-moderately ADHD as well, and his executive functioning was described as being "disproportionately low" compared to his intellectual capabilities. Is the discrepancy between his FRI and WMI an indicator of the executive functioning struggles?
All indexes came out at >99.9th%, with the exception of PSI and WMI (99.9th% and 98th%.)
Would love any and all help understanding his scores and what they actually mean. This is uncharted territory for us!
|
|
|
|
Joined: Jun 2016
Posts: 78
Member
|
Member
Joined: Jun 2016
Posts: 78 |
Hello! I’m not sure why you haven’t received any responses, and I’m not the most qualified to answer, but I’ll make a short reply and maybe it will also help bump up your post for others to see. First, I think in just about any circle, your child would be considered profoundly gifted with those scores. You may want to apply to Davidson Young Scholars if you are in the US. Secondly, I am aware that ADHD can contribute to lower PS and WM scores, but it is also not uncommon for very highly gifted children to have a discrepancy in those areas. In your child’s case, the WM and PS are very high, despite the discrepancy. Are you seeing frustration or challenges in school that are concerning?
Hopefully others will be along to contribute. When my child was tested and had a very large gap there as well, I read an interesting article that explains how very gifted children are trying to take in the world through a funnel, or something like that, and I will try to find the link and post it.
Last edited by ChasingTwo; 03/26/21 12:52 PM. Reason: Spelling typo
|
|
|
|
Joined: Apr 2013
Posts: 5,259 Likes: 8
Member
|
Member
Joined: Apr 2013
Posts: 5,259 Likes: 8 |
Welcome, RaisingHumans! You've received a great reply above. I'll just add that it may be interesting to: 1) read descriptions of the test which are found online, for example - https://handinhandhomeschool.com/testing/wisc-v/2) ask the professional who administered the test for detailed feedback, as they also observed your child.
|
|
|
|
Joined: Jan 2020
Posts: 53
Member
|
Member
Joined: Jan 2020
Posts: 53 |
Would an adult who maxed out the matrix reasoning test on the WAIS-IV max out the extended norms of the WISC-V matrix reasoning test? I was able to correctly answer the last question on the matrix reasoning test, so I would have easily maxed the section if I were to take the WAIS-IV.
|
|
|
|
Joined: Nov 2012
Posts: 2,513 Likes: 1
Member
|
Member
Joined: Nov 2012
Posts: 2,513 Likes: 1 |
Raisinghumans - While I don't have any insight to add re: those specific scores, welcome to the community! I think you'll find you'll be quite at home here.
What is to give light must endure burning.
|
|
|
|
Joined: Apr 2011
Posts: 1,694
Member
|
Member
Joined: Apr 2011
Posts: 1,694 |
Hi RaisingHumans, one of my children has a 73 point difference between VCI and WMI. And various other glaring gaps between measures, so FSIQ was not calculated. Including a 30-point difference between FRI and WMI, which the report also adds this statistic: (FRI>WMI by 30pts, 2.5%). Did you get something similar in the report discussing the statistical rarities of the various gaps and why a FSIQ was not given? This might help provide some context?
As other's have said your child can clearly be quite reasonably be described as profoundly gifted. And should qualify for programs intended for PG kids like DYS. But the uneveness can, depending on the child, make life hard, or make it hard to be "seen" fully for their strengths. That said, it's actually "normal" for gifted children to have uneven profiles and it seems the more extreme the results the more common some big spread can be.
|
|
|
|
Joined: Apr 2014
Posts: 4,074 Likes: 6
Member
|
Member
Joined: Apr 2014
Posts: 4,074 Likes: 6 |
Nice to "meet" you, Raisinghumans!
For some context on your child's scores, it's good to remember that all of them are extremely high (as in officially, the classification of those scores is "Extremely High"!), so none of the skill areas can really be described as weaknesses per se. Typically, the impact of a relative difference, even of this magnitude, is more noticeable when the lower skill is Average or below. Although it may still reflect the profile that led your neuropsych to give the ADHD diagnosis.
As to what "clinically significant" means--conventionally, a clinically significant difference should be statistically significant, meaningful and rare: 1. Statistical significance usually is at p< .01 or .05 (for those of you who care about stats!). 2. Meaningful is the actual clinical part, which is a little more subjective, but usually means there is some perceived or likely impact on function. You can find a statistically significant "weakness" in any set of traits if you run enough comparisons, but some of them really don't have much impact on life. This is a little akin to the question of causation/correlation, in that numbers are distinct from real-life meaning. 3. Rare usually references the base rate, since there are some combinations that are statistically significant, and yet surprisingly frequent in the standardization population. Most interpreters use <10% or <5% base rate as an indicator of rarity. Your child's FRI>WMI difference occurred at a 0.6% base rate in the GT standardization sample (with the caveat that the rates were derived from standardization scoring, not extended norms).
I should also point out that, according to what you've reported, there is really only one outlier score among the reasoning tasks: MR. It appears that all of the other subtest scores included in the GAI are within a point or two of each other. This is not to downplay the phenomenally strong FRI that results, but simply to note that the difference between FRI and WMI may be less extreme than at first appears. For instance, the base rate for a 25 point difference (which is what it would be if the FRI were more like the other FRI subtest) would be a not very more moderately notable 22% 8%.
More context: your child is still very young, so some of the areas that appear relatively lower may simply be later developing. One should also keep regression to the mean in mind, as it is not impossible (in fact, nearly inevitable, given the construction of the standardization table; at older ages, it will become impossible to obtain a max extended scale score of this magnitude) that the extended FRI may decline somewhat in the future. Regardless, it is evident that your child is extremely bright!
A couple of other thoughts: were the two WMI subtests consistent with each other, or were there differences between or within them? And did the NP test under true standardized conditions, or did they make modifications for COVID protocols?
Last edited by aeh; 03/27/21 09:05 AM. Reason: oops...read the wrong row.
...pronounced like the long vowel and first letter of the alphabet...
|
|
|
|
Joined: Apr 2013
Posts: 5,259 Likes: 8
Member
|
Member
Joined: Apr 2013
Posts: 5,259 Likes: 8 |
HighIQ - Wanting to respect the OP's thread and not veer off-topic, I will just answer you very briefly: 1) I am not aware of a correspondence table or validity test which correlates the scores of the two tests you mention. 2) Here is a bit of information about WAIS-IV, courtesy of Pearson: http://images.pearsonclinical.com/images/Assets/WAIS-IV/WAISIV2_6_08.pdf (2008). 3) It seems that you could have the Adult Intelligence Scale (WAIS-IV) administered, if interested to do so? 4) Your post was a bit unclear to me, as to which test you had taken, WISC-V or WAIS-IV: "Would an adult who maxed out the matrix reasoning test on the WAIS-IV (sounds like you took the WAIS-IV) max out the the extended norms of the WISC-V matrix reasoning test (sounds like you are asking what the equivalent childhood scores might've been)? I was able to correctly answer the last question on the matrix reasoning test, so I would have easily maxed the section if I were to take the WAIS-IV (this sounds like you have not taken the WAIS-IV). If you wish to discuss this further, please consider creating a new thread, so as to be respectful of the OP's topic of interpreting their child's scores, and not veer off topic, hijacking their thread.
|
|
|
|
Joined: Jan 2020
Posts: 53
Member
|
Member
Joined: Jan 2020
Posts: 53 |
Someone leaked the last question of the WAIS-IV matrix reasoning test to me, and I got it correct easily. If I can get the last question on the WAIS-IV, then I should be able to easily get the WISC-V last matrix reasoning test question correct since the kids test should be easier than the adult test.
I don't plan on taking the WAIS-IV since I know that I would easily get the maximum FSIQ and GAI and it's expensive.
|
|
|
|
|