Gifted Issues Discussion homepage Forums Identification, Testing & Assessment Large discrepancy in test results

DS is 8 yo/3rd grade. He took the WISC IV almost 2 years ago; at that time he was an 'older' kindergartner (late July birthday), and I was contemplating having him skip 1st grade and go to 2nd. He scored well - not off the charts (135 GAI), but well enough that I felt comfortable moving him up in accordance with his teacher/school's recommendation to the grade he should have been in if I had not held him. The only reason the school evaluated him was because I asked for some enrichment, spurring the conversation of 'maybe he'd be just fine being the youngest in his class' and subsequent testing/evaluation by the school (reading test, math test, a week in the 1st grade classroom at the end of K) at which point the school was confident he already knew the curriculum that would be covered in 1st grade. It was a good move to skip 1st and go to 2nd. He's the youngest and smallest in his class still, but he's fine maturity wise, and has done well academically. I was contemplating sending him to our local public school next year and he recently took the district's GATE qualifying test, the NNAT2. But he only scored a 69. I was pretty surprised. I don't believe in test prep (and he's not an anxious kid) so I told him the night before he'd take a test, told him it was multiple choice, and just fill in the scantron. After he took the test, he told me he thought it was easy/fine. After getting test results, I looked online and it turns out that are sample NNAT questions out there! I gave him about 7 questions and he got them all right (I only got 4 right!). So, I seriously wonder what is going on. Has anyone ever seen that big of a discrepancy? Or, could the WISC-IV have been that 'off'? It was administered by a highly regarded psychologist who is very experienced. I am trying to be objective but school's very easy for him, he was an early reader, etc. so his somewhat strong WISC IV seemed reasonable in that context. But the NNAT does not seem reflective. I would ask the public school to check his scantron but they'd laugh in my face...are there other factors that could cause a big discrepancy?

You didn't mention his FSIQ, which may be a stronger support if you want to appeal his disqualification. I know that some people swear by GAI but in reality he does need good working memory and processing speed for certain tests and increasingly as he progresses through school. Different tests also emphasize different strengths so both could be accurate. NNAT2 is very visually oriented (matrices) so that may not be as much of a strength? I would also not place much faith on sample questions because in my experience they are always easier (sometimes much easier) than the real test. On the other hand, trust your Mommy gut and ask them to check the scantron because I do have one high ability kiddo who could have potentially mess up a scantron at that age.

It's possible he did get off on the Scantron - I've had that happen with one of my kids and it made the difference between a typical score of 99th percentile to a 62nd percentile. However, I think it's really rare that this sort of thing happens, and my ds who had the issue has a diagnosed disability that impacts the ability to do things like fill out Scantron sheets soon... I wouldn't necessarily think that was the case with a nt kid. My gut feeling is that the difference in scores is reflective of strengths vs areas that aren't as strong for your ds. I'm not familiar with the NNAT so I can't help you with it, but hopefully aeh will stop in and help out with it (or you can also often google and get detailed info on what types of subtests are part of a test, how it is administered - group, individual, oral vs reading questions etc). Then look back at your ds' WISC report - if the GAI was used, that means there was a discrepancy in either working memory or processing speed subtests that made the FSIQ calculation not valid. There may be a clue in that info if you find a dip in either WM or PSI scores - it could suggest a relative weakness that might, in turn, have a larger influence on the NNAT (which I suspect is not as comprehensive of a test, therefore less subtests than the WISC, therefore a subtest that is weighted heavily on WM or PSI-type tasks might carry more wait in the full score).

I'd also look and see if there was any discrepancy between the VCI and PRI scores on the WISC - that also might correspond to a lower NNAT score.

Last thing - it's possible the WISC at a young age was an overestimate, and it's also possible that your ds simply wasn't performing up to his ability on the NNAT. The thing about tests is any one test all by itself without any other supporting data is just that - one piece of data. Have you looked into the requirements for admission to the public school's GATE program? The NNAT may be used as a screener to id kids who need to be tested further, but is it the only criteria for admission? If it's not - for instance, if it's given as a first-round screener and then other criteria are considered such as teacher recommendations, parent recommendations, other ability tests etc - then you can put together an appeal and advocate for placement.

Best wishes,

polarbear

Good point on the working memory and processing speed. I had to go back and dig to find the paperwork (i only remembered the highest # off the top of my head which was the GAI), but it shows:
FSIQ-128; PRI - 131; VCI - 128; WM- 113 and PSI (processing) - lowest at 109.

So definitely lower, but, again, enough to have such a low NNAT?

When DS took the WISC-IV, I literally gave him no heads up whatsoever...I did not optimize his rest, pre-test meal (none!), or otherwise make him comfy cozy. DS did complain a little bit that there was music in the background and that he had to 'turn the music off in his brain'. The doc did confirm there was background music the whole time, at a very low low level...and even offered to re-test in 6 months if DS felt distracted, but really, I didn't see the point..seemed unlikely to have a big impact, IMO, and more importantly, scores indicated to me what i needed to know- he did ok enough to suggest he would not struggle academically by skipping a grade.

I see the difference between high achiever and 'gifted' in my own kids, which is why I am questioning the NNAT in DS's case, and what could cause it to be so low. Oldest DD showed similar traits (early reader, got concepts easily, etc. though not as rapidly as DS) and had similar but slightly higher scores (139 GAI on WISC-IV and a 145 OLSAT when she applied to a magnet school at age 10..they also gave her the NNAT as an initial screener and it was 94). Her WM was higher than DS though on WISC-IV (129). My second DD is a high achiever (straight A's) but never exhibited the classic signs - not an early reader, curious but not intensely so, etc., really works to learn new concepts, etc. and I've never had her tested.

Who knows, maybe DS was 'out of sync', and a little ahead at age 6, and not now that he is 8. But he's still an avid reader (a good two hours a night); jumps on Khan academy for fun (not far ahead, just working through 4th grade math); will still work through a huge lego set alone and voraciously for hours unassisted, and has yet to have to 'review' or 'study' anything to understand it. That 69 doesn't match up with that profile, though clearly I'm not objective about the situation.

Thank you, Polar Bear. I was typing and then saw your response, and it's helpful. I'm leaning toward a private school with IB curriculum so that regardless of whether or not DS 'needs' a more intensive environment (maybe he doesn't!), I am believe it will be more interesting/challenging than what our local public school offers in the traditional classroom.

Polar Bear, in our current district, the NNAT is 'it' for gate qualification. It's all of a 30 minute test. In our former school district, NNAT was just the screener test given to all 1st graders and then again to all 5th graders; those scoring at 90 or above would get additional testing (OLSAT by district, then WISC -IV if parent wanted to provide).

The NNAT-2 is a single task, most closely resembling the Matrix Reasoning subtest from the WISC. Adding in the scantron aspect of administration, processing speed-related factors do enter, depending on the reasons for his low PSI. Even if he didn't make errors entering his responses, the additional effort of visual scanning, discrimination, and bubbling may have slowed him down, leaving him less time to reflect on and complete downstream items. It's also often administered in a group setting, which may affect children who are not as efficient at screening out background noise or visual distractions.

Was there ever any follow-up on his low PSI? Though solidly Average, it is markedly lower than his GAI index scores.

Thank you for your thoughtful response; I truly appreciate it. DS has never used a scantron before (unless he used it for WISC-IV, but has never seen one school because his school does not give standardized tests, and his teachers do not do multiple choice tests. Ever.) That might have been an issue. No, I did not follow up on his PSI (bad mom!). This may sound odd,and I'm not sure if it's because he's the youngest of three kids, but...I have always known/sensed that he would not have a hard time learning what he was expected to know. I never sounded out a word with him, yet he could read pretty fluently at age 4 picking up his oldest sister's books. I have never explained a math concept to him, and I do not know he learned to add/subtract/do basic math skills that he was not taught in K but yet knew when school tested him to make sure he knew 1st grade math. Given the 'he seems to know everything without even learning it', I did not focus in on the PSI, and I think maybe now would be a good time to look further, as he is getting older and he will undoubtedly be challenged, at some point. He attends a school that is fairly alternative and provides alot of opportunity for learning in a non-traditional sense...gardening, knitting, multiple foreign languages, mindfulness, cooking, etc. He does do reading-writing-math in school, obviously, but acceleration in primarily those areas is not the school's focus for engaging kids, and that approach has worked for him. But he has to exit the bubble, so to speak, and that PSI may come into play. Thank you again for your helpful response.

Here are my thoughts and I hope that they help you.

So, the 135 GAI should not be disregarded. Also, don't feel like you have to qualify it because it is not 'off the charts.' Think gifted or not-gifted - better accommodated in gifted classes or non-gifted classes. Someone made a specific decision to make 130 the cutoff; it has to be based on something (when I get a chance I want to research this for myself; it might be in the legislative history of the different state statutes). It may just be that represents the top 10% say for cognitive ability in any one class by grade level.

When you mentioned that specific other test I read a few sentences about it on Wikipedia, which has a criticism listed that something about that test (maybe it was new and they were working out the problems) could lead to unusual high or low scores. That is definitely worth looking into; you might help fix the problems and that can help everyone.

Then, I remembered that test might have been the one that was picked-up and discussed by the press - major magazine article (New York Times ?), radio discussion and even a small amount of television coverage. Parents were very definitely prepping their students for that test. It came up before on this forum and I remember thinking, okay, if you orchestrate your child 'passing' a gifted test, isn't anyone going to notice (especially that child) that the child isn't thriving in a gifted environment or is the parent going to force the child to work extra hard just to keep up - not a good idea for the happiness factor.

Your child is so lucky that you are aware of everything that is going on, following up is exactly what a gifted adult would do (We are constantly trying to figure everything out correctly.) Not everyone will enjoy your questioning, but hopefully they will respect it.

It is my understanding that on the "normal curve" of IQ distribution, 130 historically represents the upper limit of the 2nd standard deviation from the norm, the top 2% of the population, therefore identifies individuals with a relatively rare intellectual profile.