Gifted Issues Discussion homepage Forums Identification, Testing & Assessment Question about WISC-IV scores and extended norms

Hello,

My daughter (age 6 year, 9 month at WISC testing) suffers from Selective Mutism. At home, she talks non-stop, is a comedian, displays her intelligence, etc., etc. At school she speaks the bare minimum, rarely elaborates, etc., etc. She also suffers from high anxiety and is a perfectionist and hates to get things wrong.

In May we received the results of the school issued NNAT2. She scored really high. My husband and I knew she was smart but we did not expect her score that high. The school though sees her as "average" because of her quietness in the classroom. As a result, we decided to have a WISC-IV done on her. One of my thoughts was that we'd see a large score difference between the Verbal portion and the Working Memory/Perceptual Reasoning. From my understanding, the WM/PRI sections require very precise answers, where the Verbal scores depend more on how the child responds to the questions.

We got her WISC-IV results back. After going over them with the psychologist I realized that she scored 100% on four sections. I believe the psychologist also stopped testing her once she hit a score of 19 (is that possible? Or did she really just score a full out 19 on three sections?). My daughter sees a different psychologist bi-weekly for her Selective Mutism/Anxiety but this psychologist did not give the test. However, my daughters regular psychologist implied that my daughter reached the "ceiling" several times and the administering psychologist then moved onto the next section.

Even if extended norms should have been used, from my understanding right now it could not raise her score because they stopped testing once she scored 19 points and did not go farther. Am I correct? Had she been able to do the extended norm, then my daughter's PRI score could have been high enough to be eligible as a DYS. I asked the psychologist administering the test about extended norms but I have not yet heard back from her. I am wondering if she is not that familiar with them. I also was not expecting my daughter to reach the "ceilings", otherwise I would have asked before testing about extended norms. I know the extended norms would not push her FSIQ above 160, but when I looked at Technical Report #7 her WM score could have gone up and her PRI score could have gone up which would have also raised her FSIQ (if I understood correctly).

The psychologist's report indicated that her Selective Mutism and Anxiety probably played a part in her lower scores on those sections and that they would expect those scores to rise if my daughter's anxiety dissipated as she gets older.

So - I am looking for confirmation on what I understand, and any other advice. I plan to share her report with the school, but if there is a possibility her true score could have been higher, I'd like to point that out to them as well and/or have the administering psychologist add a sentence to the official report about that possibility. One thing I am not really sure on is if she just made it to a score of 19 or if they just stopped there. Is that possible to stop at a certain point knowing anything beyond that won't "make a difference"?

Anyway, here are her scores (Standard Score/Percentile):
FSIQ 137/99
VCI 110/75
PRI 143/99.8
WMI 150/99.9
PSI 112/79

Digit span 19/100
Letter number sequencing 19/100

Block design 14/91
Picture concepts 18/100
Matrix reasoning 19/100

Coding 12/75
Symbol search 12/75

Similarities 11/64
Vocabulary 12/75
Comprehension 14/84

Hello, BlueSky. Welcome!

I'm sure aeh will chime in here, but I don't think a 19 is "100%." I believe that is a percentile rank , which again, isn't "100%" but is more likely > 99.9 percentile (meaning, greater than 99.9% of the people who took the test at her age scored lower than her).

I don't believe this is correct. "19" is not the number of answers she got correct; it is a reflection of the percentile rank. There is very strict discontinuation criteria that has to be met before the psychologist stops administering the test. She may have hit the ceiling, but she may not have.

In your case, she has 3 19s. You should definitely ask for extended norms. That will scale her percentile rank on subtest scores up to a possible 28 (though each subtest has a different scale) and could very well take her FSIQ higher.

I'd also look into General Ability Index (though I'm not sure if you can actually calculate a GAI on extended norms).

More info here: http://images.pearsonclinical.com/images/assets/WISC-IV/WISCIV_TechReport_7.pdf

You can calculate a GAI based on extended norms - my daughter's tester did. If they don't want to or don't know how to calculate based on extended norms, ask for the raw scores to be included in the report, and you can do it yourself.

There's nothing in here that tells me that tells me whether the tester stopped when the child's score could get no better or not. But 19 is the highest possible score on a subtest without extended norms, and I see no reason not to assume that your daughter actually did hit a score of 19 on three subtests.

aeh may be able to calculate a GAI for you based on the information you have given so far. This profile looks like one that they should have routinely calculated it.

Thanks for the feedback so far. So it does seem I am not crazy for asking for additional information about her testing such as raw scores and of they stopped testing once she scored a 19. I will probably wait and ask her regular psychologist at her next appointment if I don't hear from the one who administered it by then. They are in the same practice.

Actually, you can calculate your own (non-extended) GAI by summing up the scores (not the percentile ranks) of Block Design, Picture Concepts, Matrix Reasoning, Similarities, Vocabulary, and Comprehension and then refer to the chart here:

http://images.pearsonclinical.com/images/assets/WISC-IV/80720_WISCIV_Hr_r4.pdf

If I read this correctly, there isn't a significant statistical difference between the two scores.

Actually, I take back the statement that they should have routinely calculated it - I'm not sure if they would or not. A more typical profile is to have the VCI and PRI high and the PS and WM low, in which case they definitely would have. But the GAI is quite close to the FSIQ with this profile, so it may not need to be calculated.

Good luck!

BlueSky,

I will probably see this inside-out or upside-down compared to other folks here, but as the mom of a 2e child with an expressive language disorder, my first thought when looking at the scores is - is it possible that there is a component of expressive language challenge that is driving what appears to be selective mutism, which in turn is related to the discrepancy in VCI vs PRI scores on the WISC.


This was also true of my ds when he was in early elementary school. We had no idea he had any type of expressive language challenge because at home, he talked to us all the time - and I mean ALL the time. At school, he would clam up and barely be able to utter a whisper if he had to talk to a teacher, and in peer groups he was an observer, rarely contributing anything verbally unless directly asked. We chalked all of that up to personality. Parallel to that, he was not completing work we'd expected him to be capable of doing easily at school, and we chalked that up to what we thought was perfectionism. Because of the nature of our ds' challenge, his communication challenges first became apparent as challenges with written expression. It wasn't until he was around 9 years old, in 4th grade, that he was able to tell us that he didn't know what to say in the situations at school etc where he needed to talk but was so quiet. What looked like shyness or anxiety was actually an inability to generate words.
At home, he had no issues - but at home, we weren't requiring the same type of communication. He was talking non-stop about ideas he had in his head and could clearly express. At school and in peer groups, other people were controlling the conversation and asking him for information - it's a very different type of communication skill and knowledge.

I'd take a look at the types of communicating/talk your ds produces at home vs the types of communication expected of him at school. If you think that any of what I said might make sense, you might consider having your ds evaluated by a speech language pathologist (if he hasn't already had an SLP eval).


My ds also suffered from extreme anxiety in early elementary, before we fully understood the nature of his challenges. I probably wouldn't even think twice about the note that your dd has selective mutism or is a perfectionist except for the two relative "lows" showing up on her WISC - in VCI and also PSI. It was a challenge related to Processing Speed that led us (parents) as well as teachers to think our ds was a perfectionist. In reality he was struggling with output.

It's possible none of this is relevant for your ds, but fwiw, possibly something to consider.

Best wishes,

polarbear

PolarBear - yes the thought has crossed my mind, and hence another reason I wanted a Wisc on her. In the fall her grade takes the CoGat and there is a verbal section on that. It will be interesting to see if that lines up with the VCI or if it is more aligned with her higher scores. Anyway, could I ask the school to do an expressive language test on her? However the psychologist pointed out that she is still in the high average range for the VCI. How do I get my school to test her for that when she is still high?

I know all of last year she would ask me every Monday what to write for the weekend news she had to write. We are taking about writing what she did over the weekend. And her Reading level scores always come back lower than I expect because they are based in her ability to retell something.

As always, polarbear makes good points. You'll note also that the outlier in the PRI is block design, which is the timed subtest in the cluster, suggesting that the PRI may actually be a low estimate of her ability under untimed conditions. It would not be surprising if the VCI were a low estimate due to her presentation of selective mutism, but she did respond on verbal tasks, which is not as characteristic. There is the possibility that there is a genuine expressive language disorder feeding into the mutism behaviors. Did the psych make additional comments about her verbalizations during the evaluation, that support the contention that mutism lowered the quality of her verbal responses?

Others who have mentioned percentiles, extended norms, and the GAI are also correct. Technically, there is no such thing as the 100th %ile, as that would mean that one's rank order within the norm group would be better than 100% of the population. This is why the upper extremes are reported as >99.9th %ile. The description of stopping at the ceiling and moving on may not mean that extended norms are not available. There are two different kinds of ceilings: test ceilings, which means that the test does not have enough range to capture the skills of this individual, and ceiling (or discontinue) rules, which define when the examiner is supposed to stop (usually after a certain number of incorrect responses). It is unclear which of these two ceilings is referenced in the comment you received. The maximum possible scaled score is 19, which means +3 SDs above the mean. A range of raw scores can result in this scaled score, some of which can be converted to higher extended scaled scores. It is usually recommended that extended norms be used when there are at least two 19s on the WISC-IV. The GAI is usually reported if it is more than about 4 points off of the FSIQ. And yes, the GAI can be calculated with or without extended norms. Yes, you can download the tables for doing it yourself, and, finally, yes, I would be happy to do it for you, with the scaled scores (for the standard GAI) or the raw scores (for the extended GAI).

The anxiety could be driving the slow processing speed, and the lower verbal/mutism, or slow processing/motor speed could be driving anxiety and mutism, or language processing delays could be driving anxiety and slow processing speed. I think you may need a more comprehensive evaluation, including social-emotional and speech/language evaluation, in order to tease out what the core deficit is.

Oh, and the verbal tasks, while more open-ended, also have very specifically graded responses, some of which do not require a great deal of verbal output. Most of the Similarities items can be answered in one or two words, as can many of the Vocabulary items. Comprehension, which was her strongest subtest, requires the most elaborate responses. It also has the lengthiest and most naturalistic questions, which suggests to me that she benefited a little from more meaningful, contextualized verbal tasks, where she had more opportunity to understand and respond to the questions, without the precise language skills involved in Si and Vc.