Hi Everyone! I think I have come to the right place to ask this question... How do you know if your kid reached test ceilings?? This is for the WISC IV.... Here is the subtests scores:

Similarities 19
Vocabulary 19
Comprehension 15

Block Design 17
Picture concepts 19
Matrix Reasoning 19

Digit Span 14
Letter/Number 12
Arithmetic 17

Coding 9
Symbol Search 13

Obviously lots of scatter with the Processing speed stuff, so just curious to your thoughts!!
Thanks!!

From my limited understanding (DC did some recent testing), those 19s could indicate reaching the ceilings (I believe in some cases, even an 18 can be a ceiling, depending on subtest and age of the child). Did your tester use extended norms? In my experience, extended norms could make a big difference!

Also, I'm going to guess that as is common with MANY gifted kids, your DC's GAI would be even higher than FSIQ, and might be the better measure.

Those look like very high scores!

A ceiling is reached (from my understanding) under two circumstances. The child either answered all the questions with correct responses or the rule for stopping was not reached (like getting a certain number wrong in a row...)

This indicates that if there were more questions a higher score in that area might have been reached.

The 19s and yes some 18s indicate a ceiling may have been hit. Of course...19 may have been were your child naturally stopped. Given the number of 19s I'd highly expect ceilings were hit.

I would email your tester and inquire if ceilings were hit and with which subtests. Some testers will use extended norms to continue the questions and further gauge ability. And those can even be ceilinged!

As to GAI...with the processing speed I'd imagine GAI is a better indicator of overall academic success and intellectual ability. Ask your test if she/he will calculate GAI as well for you. Lower processing speed, from my understanding, is not unusual for young highly and profoundly gifted children.

Thanks!! They did calculate GAI and it's 155. FSIQ is 143. Qualified for everything we tested for so I don't think I NEED or even want (haha) higher scores but all the >16.10''s got me curious. He's 8.9.

No extended norms were used

Yes, score interpretation is best done by qualified professionals who've observed your child's testing. Interested parents who wish to acquaint themselves with general background information in preparation for discussion with their professional may conduct a websearch to find and read information in source documents found online and freely accessible to all, such as the WISC-IV Technical Report #4 General Ability Index dated January 2005 by Susan E.Raiford, Ph.D. Lawrence G.Weiss, Ph.D. Eric Rolfhus, Ph.D. Diane Coalson, Ph.D., and updated December 2008. This 20-page technical report states on page 3: and on page 17:
While referring practioners to their extensive score interpretation manuals for guidance, the technical report provides tables for assessing differences between predicted achievement scores and actual achievement scores for discrepancy analysis. Tables do not describe predicting GAI based on achievement scores.

My understanding is that VCI, and in particular working memory and processing indexes are important for success in a typical school. I certainly see this at home with my DS; he is intelligent but his deficits in executive function makes school and homework a challenge at times. Fortunately he is doing well now, but I worry about this impact as he gets older.

What really matters is understanding the root cause of the lower scores in working memory or processing speed, and understanding what's at the root of executive function challenges. There are many reasons that you may see these issues, and many of those reasons can be remediated/accommodated. Our dysgraphic ds has similar spread in coding vs other scores as the OP, and when he was in elementary school was extremely challenged with organizational skills. He receives accommodations for the challenge that impacts his coding subtest score (dysgraphia - he keyboards in all classes plus has accommodations for testing and other dysgraphic-typical accommodations), and we were able to remediate the organizational challenges by working with him intensively on those skills during 6th grade. Today he's doing very well in school, and isn't being held back at all (grade level/acceleration etc) by the disabilities that impact his coding score and executive function.

polarbear

eta - I'd add that in some ways, I feel like it was easier for ds to make progress in remediating and using accommodations as he got older, because he was able to understand how his challenges impact him and be an active part of coming up with his own solutions.

Interesting that most psychs do not seem to calculate extended norms. Both of my kids could have technically had this done (I think there needs to be a 19, or two 18's), but neither psych did it. I did it on my own for DD and it added about 5 points to her GAI and a few points to FSIQ.
I wonder why he had 3 tests done for working memory? And which were used in the scores? My kids each had two (DD had arithmetic and letter/number and DS had digit span and letter/number).

In terms of the GAI, the NAGC has a position paper on using the WISC in terms of identifying children for gifted programs. I turned this into the school with my DD's report, since her GAI was a lot higher than FSIQ. In her case, the coding score really brought her FSIQ down a lot. We do see this "slowness" reflected in school and it is meaningful, but most psychs would say that kids should be put at the right level in terms of reasoning ability, and be given supports for weaknesses in other areas (like speed). Some gifted kids are very careful and deep with their thinking and that slows them down. I don't think this should be viewed as a "problem" but it can seem like a problem in terms of lower scores. In other kids it represents a disability, like ADHD or poor motor skills. My DS's coding score was low because when he was tested his ability to write was abysmal. The same thing happened with block design, which is timed and involves motor skills. Some psychs would have deliberately substituted a different test, but the goal wasn't to find out if he was gifted, it was to identify problems due to a brain injury and delays.
http://www.nagc.org/index.aspx?id=2455

@blackcat... I don't know why they did so many WM tests... The final soores are:
VCI 140
PRI 149
WMI 132
PSI 106
He is a slow mover and has handwriting issues for sure. I am less concerned then I would have been a few years ago. He does not have ADHD. We tested because we are relocating and needed scores for programming...
Got word that extended norms do not apply and to be honest, I am relieved!!!!

What do you mean they do not apply? The raw scores do not bring them up, or the school does not care about them?

The scores don't change.

"Does not apply" COULD mean that a child hit 19, but would not have any raw score points beyond 19, if criteria were "extended." In other words, a 19 on the WISC, from what I understand can mean a child just hit the 19 (just enough raw score points) or actually scored enough points that if scoring continued, they could have earned more scaled points (i.e., a 23 instead of a 19). It is somewhat similar to the concept of not all 99 percentiles being the same on a grade level test. On a MAP test, you will still have a 99%, despite the fact that the RIT score could be a 99% two grades up. Again - my limited understanding.

Jefferson - this may or not be the case with your DC. IF you have the raw scores, you could double-check the extended norms chart: http://images.pearsonclinical.com/images/assets/WISC-IV/WISCIV_TechReport_7.pdf
Also, a 19 is very high!

I think with DD she had two 18's and a 19. She ran out of questions on matrix reasoning which was a 19. So that brought up her GAI about 4 points. One of the 18's brought up her score 1 point and the other had no effect. I don't understand why an 18 would even be eligible for extended norms.

That chart seems to be helpful but I find it utterly confusing!! I do have raw scores if you're bored! HAHA

Here's my understanding of this based on what our neuropsych told us when testing (keep in mind, I'm just a parent and it's been several years since my ds was tested). On the WISC, ,there's an option to administer one of the optional subtests when a tester feels that the score on one subtest might not be representative of true ability in that area. If the score on the optional subtest is higher it can be substituted in calculating FSIQ - but with a note attached to the test report re which subtests were used in the calculation. My guess is that when the OP's child Letter/Number subtest score came in at 12 (low relative to other subtests), the tester opted to administer the optional Arithmetic subtest, that test score was higher, and was most likely substituted in place of the Letter/Number subtest score.

polarbear

Wonder if there is a test that can be substituted for "coding". Because both my kids do bad on it.

Because for some combinations of child age and subtest, it's not possible to score a 19 - getting all the questions right yields an 18.

I find it weird that it wouldn't be possible to get a 19, but I guess that explains it.

Parent as well and have a few thoughts that may help...

- agree processing and memory are important to understand how the cognitive abilities of a child will be expressed. GAI is a useful tool when trying to demonstrate the cognitive abilities in a child with a LD, ADHD, etc to ensure they are not overlooked when placement into gifted program is at stake. As well as it is used to ensure children with cognitive disabilities are not precluded from special needs services due to skewing higher on memory or processing. ADHD and LD are not the only reasons for depressed WM and PSI...timing anxiety; fine motor skills challenges, as well as I've read some believe processing speed develops slower in EG/PG young children...and the scores do increase especially in boys as they get older. Again, I believe I've read that.

I was only given my son's GAI as his processing was 97 and he has fine motor delays. His WM was in gifted range so just PSI was a concern.

As for extended norms...my understanding is you can hit an 18 and the ceiling at same time as the stop rule may be a certain number wrong consecutively versus overall. So child may miss enough to have an 18 without hitting the stop rule. My own child was a 19 on every VCI subtest and several PRI and extended norms were used. With the extended norms he also ceilinged 3 of the VCI (yet was at a 23 not the limit of i think 27 because of the consecutive rule) and 1 of the PRI so his extended norm score is significantly ahead of his fsiq and still a underestimation most likely...but once at 99.9 does it make a difference...in the usefulness of the score that is...too little known about the small population. Ha!

As for additional subtests our tester used them all on my son. She only used the core ones to obtain IQ scores. i realize the additional subtests are there for when substitutions need to be made for various and rare reasons but i understand they also are used to gain more insight into the child (outside of the calculated score). As my tester said...EG/PG children are so rare it is valuable to study everything about their cognitive ability we can to provide educational guidance and support. Of course she also charged by the hour

but those additional subtests have given us some insight...between the GAI and Extended Norms I never know what number to use. My tester however only reports the GAI with extended norms so the number looks super scary.

http://alpha.fdu.edu/psychology/WISCIV_Substitution.htm

So then did she pick the highest subtests to use in the FSIQ? It seems like this would unfairly bring up the IQ for kids who have psychs who do this...of course a child's full scale score will be higher if the poor scores are thrown out or not used. I was under the impression that they have to choose ahead of time which tests they are going to use in the scoring...so if a child has a known fine motor issue, the tester might decide to replace block design with something else, and never give them the block design test.

The VCI score looks wrong. Ask the tester to check.

What do you mean 22B?

Oh I know... I made the mistake! The Vocab is NOT 19.. its 16! Whoops!

The 140 is inconsistent with the 19,19,15. There is an error somewhere.

Hi not sure if question is directed to me or OP as I indicated our tester did all subtests. The additional ones were not used for scoring. Score was based on the typical subtests used. Additional ones were used for information gathering purposes only. She does reference them in her report...like how symbol search was lowest thus indicating processing is more than fine motor alone... Etc.

The neuropsych who did my DS6's testing did two substitutions because of a diagnosed fine motor issue. From what he told me as well as what I have read, there needs to be a diagnosis or reason to do it rather than just choosing to substitute to increase scores. Even though he did the substitutions, he did administer all the subtests -- even those that had fine motor components that in the end weren't used to calculate his IQ.

Ok, I got concerned thinking they threw out low scores and kept high ones. I think my DD would have actually done better on digit span than arithmetic. Not sure why the tester chose to make a substitution there.

Okay, that fixes it.

In our report the tester pointed out the difference in processing speed and the other scores and included his reasoning (because at 7 the score is not uncommon) and also that her working memory was very good (which rules out some other issues that they look for). He included her level of interest throughout the test, what really excited her, and when her attention began to flag. And he made a point of explaining her scores by comparing her level to the age of an average child at that level.

He also told us verbally that she did hit the ceiling and that he could do the extended norms (for a fee). We didn't see the need (particularly since we paid for the testing ourselves) at that time.

I point all this out because I've seen posts here where it seems that parents aren't getting that much data from their testers or have the ability to discuss their questions. Obviously we had a great experience, and I hope that level of engagement and information isn't unusual.

Thank you for sharing this link. Some may say it shows the effects of cherry-picking data in order to contrive a desired score. Super-scoring IQ tests in this manner may yield results which do not accurately reflect the intellectual profile of the child, and may be said to be gaming the system.

Ongoing research of "gifted" children yields skewed results when the study subjects may have been identified by artifice.

My DD had a substitution on her WISC and it seemed to make sense to me, although it doesn't really matter. She would have qualified for DYS without the substitution and didn't need the higher score for any reason.

I don't remember the details, but the tester wrote a paragraph explaining the reason for the substitution. Apparently DD correctly completed all of the block design and was able to solve all of them, but went very slowly and carefully. So the report described her score with the time limit and then that she was able to complete it when given more time. She then had a substitution for that part. The resulting score was almost identical to her score on other tests, so it seemed reasonable to me - but, again, it didn't matter that much as we had all of the scores on the report and it was easy to see exactly what had been done.

I did once have a tester who adjusted my daughter's score on a test to help her get a higher score. That really bothered me and I never used the score from that test. My daughter had met the termination criteria for missing questions, but was allowed to keep going and then given points for the additional questions. I was unhappy about that. So I'm sure cherry picking does happen, unfortunately.

I don't know why there was a substitution as I have not yet received the whole report. I do not think the tester cherry picked... My child is already a DYS. We were merely retesting to update scores due to a move and a program requested scores within 2 years.

I had my daughters initial testing done at a Ph.D. Level school psychology program at a university. This university employs several very well known creators of these tests and when I asked about the extended norms the evaluators checked with them and this is what I was told. " Extended norms are only to be used when an examinee receives full credit on every item in a subtest and receives a perfect raw score, otherwise extended norms are not statistically necessary, they are for the students who max out an area of the test, not students who do very very well for their age." The person who answered helped create the norms, so I imagine this answer was accurate. In short they said most of the time extended norms really should not be calculated.

In addition, he did say that the norms had been marketed in a way that the creators had not intended and they would be glad when the WISC 5 came out in the fall and the extended norms were no longer a factor.

Are you sure that's what they said? Mathematically this is quite bizarre.

He said they were initially intended to distinguish among PG (GAI 99.9 percentile before extended norms) children and not moderately gifted individuals. That they were created for research purposes and marketed quite differently. Once you sell your intellectual-property you do not get to determine its use.
I got the impression he was disturbed that they were being used to raise the scores of HG individuals. I do trust the university that did her testing as they were very informed.

I asked my mathy husband your question 22b and i am sure his explanation will be more clear:). Per the evaluator: In order to to initially be included in the extended norming sample you had to have a GAI at the 99.9th percentile and both the PRI and VCI needed to exceed the 98th percentile and you needed to have received the maximum raw score on at least one subtest (all before extended norms). When the norms were published the criteria for the use of extended norms was listed differently. His argument was unless you meet the above criteria you should not use extended norms because you do not meet the criteria of the norming group. I can see how my wording was not clear, sorry for the confusion.

sallymom - are qualified testers using them outside of these criteria, though? I imagine, statistically, they are seldom used. This list probably has a much higher than normal group of DC who would be eligible for the extended norms.

He seemed to indicate that they were but it was because the parameters were not listed appropriately by the publishers of the WISC-IV, not necessarily because anyone was knowingly using them incorrectly. He did use them for DD (she met criteria) and as you said I am sure there are many other children on this board who meet this criteria. But if there is, for instance a VCI or PRI that is 99.9 but the other score is 97 percentile then extended norms should not be used even if the GAI is high enough, I am guessing that is probably when they are most often used incorrectly. I am guessing the maxing out of at least one subtest is only important because without doing so someone would have been excluded from the norming sample. I am far from a statistician but it was very obvious the psychologist explaining this to us was.

With those kinds off cutoffs to qualify for extended norms, a single question could make a huge difference in score. That is not a robust system.

Really it would have been better if they'd tried harder to avoid ceiling effects in the first place.

The impression that I got from reading about it obsessively back when we first got DD's results is that the WISC was never intended as a tool for recognizing the gifted but rather as a tool for diagnosing issues at the other tail. Consequently, its ceilings are too low and things get too 'compressed' at the tip of the RHS tail.

This is the case--both psychs who saw DS gave him the WISC (or WPPSI back in preschool) for other reasons, not to figure out if he is gifted. The first time because he had odd delays, the second time because he still had odd delays, I noted that there had been discrepancies (huge gaps) on the WPPSI a few years earlier, and he had a brain injury. He likes using the working memory section to identify possible ADHD.
But the whole test needs to be done to find discrepancies.

My thought on it is that once you reach a certain point like 99.9th percentile, does it really matter about getting an exact number? It's not like someone is going to say "this person has an IQ of 200 vs. 160 so we will grade accelerate 4 years instead of 2 years." The same issues would be present in both cases. DD has a GAI >99.9th percentile--personally I'm surprised she tested that high in the first place, there are a lot of factors going into it (like her being super careful with her answers or maybe just luck on that day), it can change a lot over time, and I'm not going to get too hung up on what is her exact IQ....it just doesn't matter that much. I now have a general idea of where she stands and that the non-verbal questions come easier to her which is good info to have.

blackcat, I can't speak to why this particular examiner gave three wm subtests, but when I do it's often because a) there was a marked difference between the standard two wm subtests (ds and lns on the WISC, and ds and ar on the WAIS), b) I want to investigate rote memory vs memory with higher cognitive load (possibly because there were differences between digits forwards and digits reversed), or c) one of ds or lns was spoiled or suspect (in which case that scaled score should not have been reported at all, or with an asterisk). Some examiners cling to the outdated AC(I)D profile, which was once thought to be diagnostic of ADHD; that would require giving Arithmetic.

The score tables in the report should list which subtests were used in the composites. It might be in tiny print under the table of IQ/Index scores.

N, there are two kinds of ceilings being mildly conflated here. There are ceiling rules, which are essentially as you describe at the beginning of your post, and have to do with rules of test administration, and there are test ceilings, which have to do with the limits of the normative sample. A gifted child is more likely than others to -fail to ceiling- as far as test administration, by not receiving the necessary number of zero responses to trigger the discontinue rule before reaching the end of the subtest. Jefferson appears to be inquiring about reaching a test ceiling, which is a situation where the test/norms are insufficiently high level to capture the full range of an individual's ability. This is the situation for which the extended norms become relevant.

I have inserted a link to Pearson's official extended norms, in case your examiner doesn't have them:

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

As you will see from the TR, N is correct about the two-subtest criterion for using extended norms.

Jefferson, Coding is fine-motor heavy, which makes it a mixed measure of processing speed. Especially with younger high-ability students, I sometimes substitute Cancellation, which has fewer fine-motor demands.

aeh, I am trying to understand DD11's recent WISC scores (for which we don't have the final report yet). Apparently she did not reach the discontinue criterion on two or three different subtests, and yet she only got one 18 scaled score. Does this make sense? How can you run out of test without maxing out the score - especially when the test goes up to age 16 and you're only 11? Shouldn't the test have enough range that this can't happen?

My DS got a 15 while not reaching the discontinue criterion (that is, he reached the end of the question list). It can easily happen by getting several wrong answers, but not consecutively.

How bizarre. OK, thanks.

Can you clarify what you mean by this?

Once upon a time, some examiners observed anecdotal patterns of depressed scores in Arithmetic, Coding, Digit Span, (and some include Information), in the context of at least average performance on the remaining subtests, all associated with inattention and impulsivity (aka ADD/ADHD). If a particular examiner thought there was a question of ADHD, and they also believed the ACID profile was legitimate (newsflash: it's not), they might throw in the optional Arithmetic subtest to validate their behavioral observations of inattentiveness and impulsivity.

Yes. That is correct. In some cases, this is described as intrasubtest scatter, and may suggest that the measure is a low estimate of ability, the examinee was not fully engaged or attending, or that their access to instruction has been inconsistent (which might happen for gifted youngsters because they are exploring knowledge on their own, in sequences not typical of the general population).

I have also seen gifted students exhibit scatter because they employed a more simplistic problem solving approach on easier items, and then switched to a more efficient, high-level approach only after failing a number of items. One student I recall nearly reached a ceiling on the memory for beads subtest on the old Stanford-Binet-4, because he was using a brute force rote memory approach, but suddenly succeeded at quite a number of items in a row when he changed his approach to one involving grouping and mini-patterns. (Not sure if my explanation of his approach is clear, but) the takeaways are 1) his actual working memory was much higher than that represented by the scaled score, and 2) I would not have derived as much information from this test if I had not explored the reason for his abrupt increase in performance near the end of the test. The qualitative information about his thinking and reasoning was much more interesting than the scaled score.

That makes sense and is fascinating. Thank you!

Thanks for all the info... it is fascinating.

OK, so extended norms are being over-used, but in the Pearson link posted earlier it says the highest score in the norming sample was 151. So how is a score from 150 to when extended norms kick in valid? I do understand it's all pretty timey-wimey wibbly wobbly over 99.9 percentile, but extrapolating beyond the top end of your sample seems even more suspect.

And that also became, de facto, the top score in the gifted validation sample, using the standard norms. That is to say, we have no idea whether that 151 was really 151, or should have been 161 or 181. The idea of the exercise with extended norms is to more finely distinguish all the examinees who bunch up at between about 140 and 151 (if you factor in the confidence interval). If you look at the increased spread from standard to extended norms (e.g., FSIQ top score goes from 151 to 159, VCI from 155 to 188), clearly not all 150s are created equal. Also, extended norms are a way of rescoring the raw data into scaled scores above 19. It overlays the standard norms, rather than purely replacing them (you can think of the 151 on the standard norms as 151+). The rule about needing two max scaled scores is supposed to restrict use of extended norms to those children whose confidence intervals include the maximum Index score under the standard norms. But I grant you that the extended norms have more clinical utility than strict psychometric robustness. I'm sure there was a fair amount of fun with curve smoothing involved.

Pearson also really had to release some technical guidance, because people were starting to make their own rules for this. Plus, more cynically, the Stanford-Binet LM had been clinging to the gifted eval market for a long time, and part of its appeal was undoubtedly the possibility of scores in the 200s. The extreme agedness of the LM created a hole in the market, which had not been definitively captured.

I hope what I just wrote makes sense, because it is definitely past my bedtime!

Thank you, it does make sense. It's like above grade testing instead of grade level testing. Is that a fair comparison?

I don't know that ranking of the kids in the over 150 area matters as much as the individual child's personality and quirks (which is probably true of all children, now that I think about it).

That's what happened with my son, twice he got 60 out of 60 on the math part of the Cogat. We have no idea what it means, because what would he have gotten if there were 80 or 100 questions?

I always scored 13th grade on state standardized tests at my school, from 6th grade (the first grade I took it) on. They never increased the maximum level and didn't tell us what our raw score was.

My opinion is that if your child is tested, especially if you are paying for it yourself, they better give you a raw score too. This is also true with SAT subject tests too, for some tests, you could get 1 wrong and not get a 800, but others you could get 7 wrong and get an 800. The raw score will tell you more than the converted reported score.

(note that the G&T criteria for JH CTY seems very low to me, compared to what little they give much more gifted kids in public schools)

"The raw score will tell you more than the converted reported score."

Not really. The converted score is the part that tells you where the student stands in comparison to the national norm population (hence, are they statistically unusual enough to be considered gifted). It also compensates for differences in forms of the same test (such as for the SAT/ACT/GRE, etc.), by rooting them in the same normative sample. I do, however, always include raw scores in my score tables, in addition to all the converted scores, because I would like that information if I were receiving the report, both as a parent and as a professional. I would agree that the raw score provides additional information about ceiling effects.

I know that this is a very old post, but anybody claiming that inclusion in the extended norms sample required 98th %ile in VCI and PRI, a 99.9th %ile GAI, and a ceiling raw score in at least one subtest is clearly having you on. According to Technical Report #7, norming data came from both the original standardization sample and data provided by the NAGC. NAGC data included a VCI of 110, PRI of 102, FSIQ of 118, and GAI of 120. Even the subject of the case study provided with the report (with a GAI of 208, although the reliability of norming data is likely quite dubious in that range) did not attain the maximum raw score on any subtest.